Why Growth Stalls, and the Five Questions Jason Cohen Asks to Diagnose It
Lenny's Podcast
Jason Cohen has founded four companies. The last two, including WP Engine, became unicorns; the earlier one was bootstrapped. Cohen has also invested in about 60 startups and has written at A Smart Bear for nearly two decades. In this conversation with Lenny Rachitsky, almost the entire discussion centers on one problem: a product starts working, grows for a while, and then growth stalls.
Cohen's answer is not a checklist of tactics. It is an ordered series of diagnostic questions: whether customers are leaving, whether pricing is right, whether existing customers are growing, whether acquisition channels are saturated, and whether the business needs to grow at all. The order matters, he argues, because a problem higher on the list makes fixes further down largely irrelevant. As Rachitsky notes during the conversation, the same sequence works just as well for a company that simply wants to grow faster.
A Writer Who Publishes Only What He's Proud Of
The episode opens with Cohen's writing. By his count he has published roughly 300 to 350 pieces over about 18 years, of which 150 to 200 are ones he is proud of. That is not many for the time span, and he says it is deliberate. Conventional advice when he started blogging, and still today, was to publish on a regular schedule so readers could fit your work into their routine. Cohen never followed it. His rule was to publish only when a piece was the best he could do, and leave it to readers to judge whether it was useful. Some years that meant one or two posts, other years around 40. During that period he was also founding and running companies, which limited how much he could produce.
Rachitsky says he used to tell people that success with online content came from quality plus consistency, and has increasingly concluded that quality is what matters. Cohen agrees but names the cost. Wanting everything to be your best work keeps the bar high, and it can also tip into paralysis. He says he still struggles with that tension.
How Growth Actually Slows
Cohen lists several reasons growth can stall right now: an economy that he suggests is weaker than headline indicators, AI and uncertainty about it, and plain size. Nobody grows 2x a year forever. Sometimes a single event causes a sudden drop, such as an algorithm change. More often, in his experience, growth decelerates gradually. Teams watch growth constantly, so it doesn't catch them off guard, but it feels like "running through mud": the same amount of work produces less and less.
His framework is ordered like a marketing funnel. If one stage is where everything falls apart, tuning a different stage won't help enough. You go to the biggest problem first.
Question One: Are Customers Leaving?
The first question is about logo churn, meaning the number of customers who cancel. Cohen calls it the worst problem for several reasons. Once a customer is gone, nothing more can be done: no saving them, no expanding their revenue. Churn also tends to come with negative reviews and social media complaints, which suppress growth further.
He also makes an emotional case. Think about the gauntlet a customer went through to become one. They had to hear about you at all, which is improbable. They clicked, didn't bounce off the homepage, reached the pricing page without being scared off, had the budget, bought the product, and spent time onboarding. Almost no one gets through all of that. Someone who did clearly wanted the product to work, and they still left. Cohen's reaction as a product person is that something fundamental is broken: the company is not fulfilling the promise it made, or the one the customer thought it made. That may be a product failure or a communication failure.
The ceiling that churn imposes
Online debates about acceptable churn rates tend to be abstract, Cohen says. He considers anything above about 3% monthly cancellation terrible, while others argue that 5, 6, or 7% is fine. He prefers a metric that makes the stakes concrete.
It rests on an idea he thinks people underappreciate: cancellations grow automatically as a company grows, and marketing does not. Imagine your customer base tripled overnight with the same kinds of customers. Next month, AdWords, SEO, and your other channels would deliver the same number of new leads, because they don't care how many customers you already have. But with a 5% monthly cancellation rate, the absolute number of cancellations would triple. Marketing grows only as fast as you can improve it, which is hard, roughly linear work. Churn is a percentage of the whole base, so it is exponential. Churn therefore always catches up with marketing eventually.
Rachitsky compares it to a leaky bucket. Cohen adds that the leak widens automatically as the bucket fills.
The ceiling is easy to compute: new customers per month divided by the monthly cancellation rate. A company that adds 100 customers a month with 5% churn will never have more than 2,000 customers. As it approaches that number, growth slows toward zero, because nearly as many customers leave each month as arrive. Cohen presents this as an explanation for why growth slows automatically at SaaS companies in general. The same calculation works in revenue terms. He prefers customer counts because "we will never have more than 2,000 customers" hits harder.
Why the cancellation survey lies
Once churn is identified as the problem, you have to learn why people cancel, and customers who are already leaving don't want to spend time explaining. Cohen tells a story from his earlier company, Smart Bear. Its cancellation form had a dropdown of reasons ("too expensive," "project ended," and so on), and one reason was chosen more often than the others. It was first on the list. When the company randomized the order, every option was chosen about equally. The data had been noise. He says other companies have seen the same thing.
His first recommendation is to use open-ended questions. Most people won't answer, but the answers you do get come from the customer. Phrasing matters too. Instead of "Why did you cancel?", ask "What made you cancel?", which points the customer toward what about the product or situation caused it. He took this from Groove, which published a case study: its cancellation email got 10% usable responses asking "why," and 20% after switching to "what made you cancel."
"Too expensive" is never the reason
Cohen says "too expensive" is usually one of the top reasons in cancellation data, and he considers it never the real one. The customer already saw the homepage, understood the promise, looked at the pricing page, and bought. Their actions show the price was acceptable for what they believed they were getting. Something else happened: the product didn't deliver what they thought it would, or something didn't work. They may have lost their budget, but that is a different reason from the product being too expensive.
He compares this to medicine, where a death certificate records the proximate cause and doctors also try to capture the underlying one. "Stopped breathing" corresponds to "too expensive." Asking further might lead to a car hitting a telephone pole, then to the driver passing out at the wheel, then to undiagnosed diabetes, which is much more useful. Cohen stops to say he dislikes the idea of a single "root cause." Complex systems have many interlocking factors, each of which could be detected earlier or changed. In the medical example, a healthcare system that isn't preventive and a patient who didn't see a doctor both play a part. A good analysis should produce an array of contributing causes, which he jokingly calls "rooter causes." He is also wary of the "five whys" technique for the same reason, since it implies one cause at the bottom.
In a software example, the real reason might be that the product integrates with Jira but the customer uses Linear. Whether to build that integration is a separate decision, but that is the reason, not the price.
"Project ended" gets similar treatment. Cohen mentions seeing a founder on an entrepreneur forum say that more customers were citing it and nothing could be done. Cohen disagrees. If the product had been more useful, would the project have continued? And did the company choose a segment, such as small businesses or consumers, where projects end often because small entities have high variance? That could be a mistake in choosing the ideal customer profile. None of this proves the market should change, he adds, but saying nothing can be done closes the door on options that might be right, and the chosen segment has a lot to do with retention.
Catch customers before they leave, and invest in onboarding
Cohen also recommends talking to customers who are in trouble but haven't canceled yet. They can sometimes be saved, and they are still willing to talk. The warning signs depend on the product: never uploading data, calling support too often, not calling support at all, not logging in for a while. With enough data you can statistically correlate behaviors with cancellation. Without it, you can form a theory, act on it, and revise it as data comes in.
Rachitsky mentions the long-sought product that watches metrics and flags customers about to churn. Cohen separates two things. Talking to struggling customers doesn't require scale; statistically identifying the predictive behaviors does. He also refines a common piece of advice. People say to find what good customers have in common, but much of that is shared with bad customers because it's simply what all customers do. What matters is what distinguishes the good ones from the bad ones, and in churn analysis, what distinguishes customers who actually cancel.
His default when unsure where to start is onboarding. Most companies see far more cancellations in the first day, 30 days, or 90 days than in the rest of a customer's life, and small onboarding changes can have large effects. He uses his own YouTube videos as an analogy. About 50% of viewers leave in the first 30 seconds, and roughly 20% remain at the end of a longer video. Improving the late part of the video barely changes the final number. Raising early retention from 50% to 55% could raise end-of-video retention from 20% to about 25%, an increase of 20 to 30% in the final outcome. In SaaS, early churn is also the most unprofitable kind, because the company paid to acquire those customers and they didn't stay long enough to pay that back. Even when you know what else to work on, Cohen says, he would still bet on onboarding.
Cohen also treats this first question as a threshold rather than something to keep polishing. For small and midsize businesses, he says 2% monthly churn is good. There is a floor below which churn can't reasonably go, and if you're near it, churn is probably not what's holding growth back.
Where AI fits in analyzing feedback
Asked whether AI could summarize cancellation feedback, Cohen answers carefully. In his experience, LLMs are good at identifying themes and bad at surfacing the specific, actionable details. His explanation is that an LLM predicts the most likely output, which makes it something like an averaging machine: useful when you want the average, weak at finding what is unusual. And the themes are usually predictable anyway, such as "couldn't figure out this integration." The specific details are what make a person think, "Wait, that means we should do this."
His workaround is to ask the model for themes, then ask it to list every specific detail under each theme with which customer said it and a link to the source, so a person can read the specifics. AI can also clean up bad grammar. But he would not rely on it to do the thinking, and he suggests just reading the feedback. Rachitsky adds that outside large consumer apps, the volume of cancellation responses is usually small enough to read directly.
Question Two: Is the Pricing Right?
The second question is whether pricing is correct. Cohen quotes Patrick Campbell, whom he describes as having data on 4,200 startups: "Your prices are way too low because you just guessed and you haven't changed them." Many companies, Cohen says, copied a competitor or adjusted a number for vague reasons.
People fear raising prices partly because of the textbook demand curve: higher price, lower demand. Cohen says that often isn't what happens. Frequently, a price increase leaves signups unchanged, or signups rise. He says this happens even to solopreneurs on Twitter who raised prices nervously and found that sales went up.
The founder who charged $300 a year
Cohen tells the story of a founder, whom he declines to name, who sold to enterprise and government customers for $300. Cohen assumed that meant per month; it was per year. The product got one or two signups a week, which is normal for a startup selling to enterprises. Cohen suggested changing the price from per year to per month, a 12x increase. The founder did, and signups stayed at one or two per week.
The founder's next plan was to use the new revenue to hire an engineer and invest in marketing. Cohen stopped them: the next step was to raise prices again. A 12x increase with no visible change means the company still hadn't found the right price. It didn't have to be another 10x, maybe 2x or 50%, but pricing wasn't finished. Cohen notes that this hadn't occurred to the founder even then.
Pricing selects the market
Cohen's explanation is that price determines which market you're in. At the bottom are customers with small budgets who get limited value, and they do behave like the textbook curve: raise prices and they leave. But a company with a thousand employees and $400 million in revenue, looking at a product priced at $2 or even $100 a month, concludes it can't be good enough. The vendor must be immature, support must be weak, governance policies must be missing. Whether or not that's true, the product looks low quality and aimed at small businesses, so these companies don't buy.
For those buyers, demand rises once the price enters a range that makes sense for what they need, stays up across that range, and falls again at some high point; nobody will pay $10 million a year for it. Cohen describes the resulting shape as more like a mesa than the downward line in textbooks. Only the lowest end of the market, the part that is also worst on most metrics, behaves as the microeconomics curve predicts. Raising prices moves you into a different market, which is why signups can rise. And higher-paying customers, Cohen says, tend to retain better and improve other metrics too.
He adds that pricing is as much art as science, and that the market changes over the years even after you get it right. Price also includes structure: per site, per use, per seat, and so on. The same product at the same price can feel fair or unfair depending on how it's structured.
Double Down: 8x the revenue from repositioning
Cohen's main illustration is the case in the episode description of repositioning a product to charge eight times as much. He says it's based on something that happened to him, simplified because the real details are complicated and uninteresting.
A fictional company, Double Down, makes AdWords spending more efficient and advertises "cut your AdWords cost in half." A customer spending $40,000 a month on AdWords would save $20,000, but won't pay all $20,000 to Double Down, because then they'd save nothing. Suppose they pay a quarter of the savings, $5,000 a month, and net $15,000. Everyone is reasonably happy, and nothing is wrong with the business.
Cohen then imagines the CMO or chief product officer reporting to the CEO at year end. "We saved money" is welcome. But a CEO would rather hear about growth, which he calls healthier for the company, better competitively, and more valuable. So the pitch changes. The company currently pays $200 per lead. If Double Down halves the cost per lead, the same budget buys twice as many leads. The pitch becomes "double your leads." The company is already willing to spend $40,000 a month for its current leads, so it would be willing to spend $40,000 more to double them. Double Down now earns $40,000 instead of $5,000 for the same product. The CEO, told that leads doubled at the same customer acquisition cost, asks how to get more of it.
Cohen acknowledges he's exaggerating to make the point. The lesson for product managers is to sell more of what the customer company already values, whether growth, customer retention, or competitiveness, rather than cost or time savings. Savings are real value, but they cap what the customer perceives as your worth. Delivering more of what they value raises that cap by perhaps an order of magnitude, in Cohen's estimate.
Rachitsky mentions Jen Abel's observation from an earlier episode that enterprises typically buy SaaS contracts in the $75K to $150K range, and that selling becomes easier once a product fits there.
Pricing can't be changed in isolation
Cohen warns against treating price as a separate knob. A new market brings new requirements: SOC 2, governance features, integrations with systems you didn't know about, professional services. Your competitive advantage may exist only in your current market. The reason may also be cultural. His example is Buffer, which he says could move upmarket but decided it is a company for small users, doesn't want to build for large companies, and finds its current work fulfilling. Changing prices is a decision about the whole strategy, not just a number.
Question Three: Are Existing Customers Growing?
The third question returns to the churn math. Marketing can't keep pace with cancellations because it doesn't scale with the customer base. What does scale is revenue from existing customers. If 2% of customers leave but some of the remaining 98% upgrade, pay more through usage, or otherwise expand, that growth is proportional to size, and it triples if the base triples.
The metric is net revenue retention (NRR): take current revenue from existing customers and see what it becomes one year later, excluding new customers. Cancellations and downgrades lower it, upgrades raise it, and it can end above 100%.
Cohen explains why NRR alone isn't enough and why he still ranks logo churn first. If NRR looks healthy but too many customers leave, not enough remain to upgrade. He also uses an investment analogy. A $100 stock that falls 20% is worth $80; a 20% gain from there brings it to $96, not $100. Recovering from a 20% loss takes a 25% gain. NRR treats a 20% loss from cancellations as offset by 20% from upgrades, which Cohen says overstates the situation. He recommends tracking both metrics while remembering that NRR is somewhat less favorable than it looks.
His evidence that NRR is essential for scale: if churn really caps growth, large public SaaS companies should almost all have NRR above 100%. He says that of more than a hundred public SaaS companies, about two have NRR below 100%, and those have poor financials and valuations. He cites the median NRR of SaaS companies at IPO as 119%.
He considers the tactics for raising NRR (new features, tiers, usage- or seat-based pricing) fairly obvious. The principle he emphasizes is that customers should agree they're getting more value when they pay more, ideally far more: if the price doubles, they should feel they're getting five times the value. So companies should measure the value customers receive, not just usage statistics that are easy to collect. If that value can be a single number, it might be a North Star metric. If not, use proxy metrics and qualitative conversations. He says many important things aren't numbers, giving market differentiation as an example. Most people, he notes, have left a product they loved when its price rose faster than the value they felt.
He summarizes it as creating more value for the customer and then splitting it with them. The company isn't a charity and is entitled to a share, but it has to create the value first.
Rachitsky raises land-and-expand and Jen Abel's point that a low initial price becomes a reference point that's hard to move from; a customer who started at $10K will resist a jump to $100K. Cohen agrees and adds that you may not deserve the increase. Investors and other forces that aren't the customer can push companies toward policies that are bad for customers. His test for any internal proposal is whether it is actually good for the customer, and whether the customer would say so. Proposals are often framed as if they are when they only benefit the company. His standard is that a change must be good for both. Companies that don't hold to it may succeed short term, he says, but it isn't a good long-term strategy.
Does this apply outside B2B SaaS?
Rachitsky asks whether expansion revenue is harder in consumer products. Cohen agrees it's harder, but says the rules come from the financial mechanics of any business, so they still apply. If you can't find an expansion lever, move to the next question, but a product manager who does find one creates far more value. As a consumer, Rachitsky doesn't want to pay AT&T more, since it offers no more value, but Cohen points to Amazon as a company that does. A second product sold to the same customers is another route, common in consumer goods; Rachitsky mentions AG1 adding a sleep supplement, and Duolingo's many small paid upgrades.
Cohen adds another lever that scales with the customer base: existing customers bringing in new ones through referrals and word of mouth. This is common in consumer products and harder in B2B, where "a midsize company" doesn't refer a friend the same way. The question of how the existing base can drive growth is the same; the answer depends on the segment.
Question Four: Are Your Acquisition Channels Saturated?
The fourth question is whether marketing channels have hit their limits. When growth stalls, Cohen says, the usual reaction is to push the AdWords team for more leads and the SEO team for more searches. But channels have real limits. Magazine advertising called it "inventory": there are only so many searches in your category, and you can appear only once per keyword. There's a theoretical maximum and a lower practical one, and you may be near it.
The elephant curve
Worse, channels decline. People describe channel growth as an S-curve that rises and then plateaus, followed by optimization. Cohen says the curve actually sags after the plateau. He named this the "elephant curve": a trunk rising, then the back sloping down. He says any marketer has stories about it.
He gives several reasons. Audiences saturate: people who have already seen your message many times and didn't act mostly won't act later. Channels themselves decline without telling you. When he bought magazine ads, publications reported rising circulation every year until they shut down. Conferences reported strong attendance until they went out of business. He says the same decline happens with AdWords, Facebook ads, SEO, and affiliate programs. On AI's effect on channels, he says he doesn't know how it will play out now or in two years, only that it won't make every channel grow.
For Cohen, this reinforces his first point: a company can't rely on marketing indefinitely. There are a limited number of channels your customers use, and they sag over time.
What to do about it
He asks listeners whether they know right now which of their channels are saturated. If not, the answer may be all of them. In that case, adding a small feature and pushing marketing harder won't work, however good the feature. He says this is often what companies do anyway, so the question isn't just whether you know this but whether you're acting on it.
Adding indirect channels to direct ones, or vice versa, is one option, but Cohen says, without data, that products that sell well through direct advertising often don't do well through social and SEO, and that SEO-driven products often find ads expensive and ineffective. Rachitsky's view is that most companies get some results from many channels but usually one channel (sales, word of mouth, or ads) drives most growth.
Cohen suggests getting more creative about what counts as a channel. Constant Contact, which sold email marketing to small businesses before modern tools existed, restarted stalled growth by holding in-person workshops in cities, teaching restaurant owners and dentists email marketing using Constant Contact. It seemed impossible that in-person events could pay off for a roughly $20-a-month product, but Cohen says it worked, partly because the company recruited power users who were also agencies to run them. HubSpot tested selling through agencies instead of only direct, and Cohen says that became 50% of revenue after four or five years. At WP Engine, many sites were sold through agencies that build WordPress sites.
The alternative is a new product or market. Cohen mentions a framework he has written about for these expansion decisions. The idea is to keep one foot planted in an existing strength or asset while moving the other into riskier territory with high upside. When acquisition channels are full, he says, these larger bets become necessary because pushing marketing isn't an option.
Rachitsky mentions upcoming or published posts on new growth channels, including Emily Kramer's on ecosystems as a channel, with a quote from Wiz's head of growth about starting with an existing audience of 10,000 rather than zero, and a post on ChatGPT's app store as a possible channel.
Question Five: Do You Actually Need to Grow?
The last question assumes the previous four have been answered satisfactorily: do you need to grow? Sometimes the answer is that this product in this market doesn't need to grow, and the company needs a new product or market. Or the goal can change from revenue to profit. Cohen points to 37signals and many bootstrapped companies that hit a limit and found it acceptable, with founders earning millions a year in dividends. Growing larger might mean running an organization they don't like or serving customers they don't want to serve.
He then takes up the saying "if you're not growing, you're dying," asking whether it's true or just something investors say to push founders to grow when they shouldn't. His view is that it is fairly relevant even for bootstrapped companies that reject growth at all costs. In a company stagnant for years, is the founder fulfilled doing the same thing every day? Do employees get to advance their careers, learn, and innovate? For some people, such as a CPA with a stable client base, stability may be what they want. But he thinks many people drawn to building products and starting companies don't want to repeat the same work for 20 years.
He suggests reading the saying as being about the person, not only the company. Many people, he says, have spent their whole lives in "grow, get promoted, do more" mode without asking whether it makes them happy, or have gotten stuck pursuing growth without stepping back. He points to people who say they hate work but decline after retiring once they lose their purpose. If the business is truly stagnant after everything else has been tried, something drastic may need to change: leaving, restructuring, selling, changing jobs. He calls this question existential rather than another metric, and says you should ask it at least yearly, not only at the end of this sequence. It applies at every scale, he adds; plenty of public companies aren't growing either.
Rachitsky relates this to his newsletter: he'd be sad if it stopped growing, but it has already given him a great life. Cohen extends the point to how you grow. There are probably ways to grow the newsletter that Rachitsky wouldn't be proud of, and success lets you refuse them. Early on, Cohen says, people may work to a looser standard, for example copying other bloggers' styles, and that's fine for getting started. After 20 years, having no voice of your own would be a problem. His conclusion: "if you're not growing, you're dying" is fair, but it has to come with limits, and the more successful you are, the more seriously you can hold to them.
Deciding When to Quit, and the Limits of Probability
Rachitsky mentions advice from Matt MacInnis of Rippling that if a startup hasn't clicked after four or five years, it may be time to move on. Cohen says his next book will be about decisions under uncertainty: whether to quit, move cities, marry someone, or launch a company. The standard advice is to use probability and expected value, but Cohen argues we don't know the probabilities or their distributions, so expected value can't actually be calculated. And even if it could, an individual either sells the company or doesn't; statistics about many cases don't apply to one life. He says probability isn't the right tool for these decisions, though he doesn't yet say what is. Rachitsky notes it raises more questions than it answers, and Cohen agrees.
Hidden Multipliers and the Common Thread
Cohen's first book, Hidden Multipliers, was available for preorder at the time of recording. The title refers to small actions or decisions with large effects on revenue and profit, such as reducing cancellations from 5% to 4% or focusing on onboarding instead of later stages. He says the goal is to go deep enough with examples and specific steps that readers who already "know" these ideas actually act on them. He calls it his best work.
Asked for a final takeaway, Cohen says the common thread is customer value: the product promises the right thing, delivers it, customers can onboard and use it, they recognize the value, and the company measures it. If the North Star is creating value the way customers understand it and in their language, then good pricing and good retention become much more achievable, though not automatic. He acknowledges this sounds like a platitude and that platitudes aren't very actionable, but says it's a reasonable summary given how much detail they've covered. Rachitsky adds that the value may already exist but be aimed at the wrong customers or positioned poorly. Cohen agrees: all of these things have to be right, including communicating the value clearly enough that a visitor understands it on the homepage.
AI Corner: Turning Charts into Data
For his writing, Cohen wants real data, but much of it on the internet exists only as images. He's found AI, especially Gemini, good at converting a chart image into a table he can paste directly into Google Sheets. This lets him gather ten examples to test a theory where previously it was too much work.
Contrarian Corner: A/B Testing Mostly Doesn't Work
Cohen's contrarian view is that A/B testing works poorly and doesn't apply to most things. It can't test strategy, vision, or insight, which he considers the things that matter most. "You don't A/B test whether Uber is a good idea." When teams do test details, they try verb after verb and ad after ad, adopt the seventh or eighth "winner," and repeat. A year later, stacked wins should mean 50–100% improvement, but conversion rates are unchanged. His explanation is that the tools are often not statistically accurate, and that even an accurate tool produces many false positives when real effects are rare. He says real effects are rare in typical A/B testing, so most "wins" are false positives. Unless a team is very sophisticated, he considers it a large waste of time. He uses the poker saying: if you don't know who the sucker at the table is, it's you.
Rachitsky says A/B testing has been useful in his career at scales where small percentage gains mean millions of dollars, while agreeing most people don't work at that scale. Cohen responds with an earlier episode featuring Shopify's CTO, who described a team of about 100 people keeping holdout groups and finding that roughly a third of measured effects disappeared on follow-up. Cohen's point is that if you're working at that level, good for you, and if not, he's skeptical.
Lightning Round
Cohen recommends William Zinsser's On Writing Well ("on my best day I write like that") and Geoffrey Moore's Crossing the Chasm, which he says many people know only from a diagram without having read it. He values Moore's definition of a market, which includes that people in a market respect each other's opinions. That explains why moving to a new market is hard: your case studies stop working. Rachitsky summarizes Zinsser's lesson as "cut," and they recall Zinsser on a panel with an amateur writer who said he never knew writing could be hard, to which Zinsser replied he never knew writing could be easy.
Cohen's favorite show is ER, which he's watching with his 16-year-old daughter, now around season 13; he takes her view that it holds up as evidence that it's good. His favorite products are Wispr Flow for dictation and Anker chargers. His motto is "Be yourself; everyone else is taken," commonly attributed to Oscar Wilde, though while checking citations for the book he found no evidence Wilde said it, and no evidence of who did. He adds that people who like your work want you to be yourself, because that's what they like.
The award on his desk is the 2017 Ernst & Young Entrepreneur of the Year for Central Texas, which he shared with WP Engine CEO Heather Brunner. Cohen calls her a "late-joining co-founder," borrowing how Reid Hoffman described Jeff Weiner at LinkedIn, given her impact over 11 years as CEO. He notes that co-winners are rare enough that he can't think of another case.
Cohen closes by noting that all his articles at A Smart Bear are free, with no ads or courses, and that the book is the only thing he has charged for.
A lot of product teams, a lot of founders build something, it starts to show some success and then all of a sudden it just stops growing.
There's a series of questions that I ask to diagnose why is growth slowing. The first question is, are customers leaving? Think about the gauntlet they went through to get to the product. How do they even find out about me? That was hard already and improbable. They didn't just bounce off the homepage, which is again improbable. And they got to the pricing page that didn't scare them off. They actually had the budget and bought the stupid thing. And after all of that, which clearly means they wanted it to work, they're like, "No, bye." What? Like, just on an emotional level, you got to go, "Wait a minute, that's terrible."
Step two is pricing positioning.
Your prices are way too low because you just guessed and you haven't changed them. What often happens is you raise prices and signups don't change. Just think about a company with 1,000 employees and 400 million in revenue or whatever. If they see a product that's $2 a month or even $100 a month, the thought is like, "That can't be good enough." We position this conversation as how to deal with stalled growth, but it's actually just as useful for how do I grow more?
Do you know right now which channels are saturated and which aren't? You can't just rely on marketing forever. Just adding one little feature and then hoping we can flog AdWords is not going to work.
What comes next?
The last question is, do you need to grow? We all have heard the phrase, if you're not growing, you're dying. Is that true or is that the kind of thing that investors use to make founders try to grow even when they shouldn't?
Today my guest is Jason Cohen. Jason is a four-time founder, including two unicorns, one being WP Engine. He's not just an incredible builder and entrepreneur. He's also an incredible writer and sharer of product wisdom. He's been sharing his advice online for over 20 years now. I've been a huge fan of Jason's from afar for so long, and it was such a treat to have him on the podcast. There are a million things we could have talked about. I'm definitely going to have him back.
In this conversation, we spent the entire time talking about his very actionable and very helpful framework for what to do when your product's growth stalls. I found his way of looking at the problem incredibly practical and real and actionable. And if you're looking for ideas for how to rekindle your product's growth or just accelerate the growth of your product, you're going to walk away from this conversation with your mind buzzing.
Also, I'll add that after 20 years of blogging online, Jason is about to publish his very first real book. It's called Hidden Multipliers. You can now pre-order it online at hiddenmultipliers.com. I am going to grab a bunch. I bet after listening to this conversation, you will too.
If you enjoy this podcast, don't forget to subscribe and follow it in your favorite podcasting app or YouTube. And if you become an insider subscriber of my newsletter, you get a year free of over 20 incredible products, including a year free of Lovable, Replit, Bolt, Gamma, n8n, Linear, Devin, PostHog, Superhuman, Descript, Wispr Flow, Perplexity, Warp, Granola, Magic Patterns, Raycast, Mobbin, and Stripe Atlas. Head on over to lennysnewsletter.com and click product pass. With that, I bring you Jason Cohen after a short word from our sponsors.
This episode is brought to you by 10Web, the company that pioneered AI website building before ChatGPT. In the last 3 years, over 2 million websites have been generated with 10Web's vibe coding platform. 10Web's vibe coding platform is a powerful way to build websites. Think of it as Lovable for WordPress front end and backend. Users can build any website at any complexity: e-commerce, portfolios, information websites, blogs, and it comes with the WordPress admin panel and thousands of ready-to-use plugins.
10Web also offers website generation as an API as a service for SaaS companies, marketplaces, hosting providers, MSPs, and agencies. SaaS companies can embed it via API so that users can launch AI-generated sites directly inside of their platform connected to their own data. Agencies and MSPs can get a white-label dashboard to manage clients and resell under their brand. Hosting providers can self-host the API builder on their own infrastructure.
Check it out at 10web.io/lenny and use code lenny for exclusive free credits and 30% off API or white-labeled solutions. That's the number 10 web.io/lenny. Vibe coding platform as an API.
This episode is brought to you by Strella, the customer research platform built for the AI era. Here's the truth about user research. It's never been more important or more painful. Teams want to understand why customers do what they do. But recruiting users, running interviews, and analyzing insights takes weeks. By the time the results are in, the moment to act has passed.
Strella changes that. It's the first platform that uses AI to run and analyze in-depth interviews automatically, bringing fast and continuous user research to every team. Strella's AI moderator asks real follow-up questions, probing deeper when answers are vague, and surfaces patterns across hundreds of conversations, all in a few hours, not weeks.
Product, design, and research teams at companies like Amazon and Duolingo are already using Strella for Figma prototype testing, concept validation, and customer journey research, getting insights overnight instead of waiting for the next sprint. If your team wants to understand customers at the speed you ship products, try Strella. Run your next study at strella.io/lenny. That's strella.io/lenny.
Jason, thank you so much for being here and welcome to the podcast.
Thank you. It's an honor to be here.
It's an honor to have you here. I have wanted to get you on this podcast for so long. You are both an incredible builder and a founder and you are such a great communicator. You have been writing at A Smart Bear, asmartbear.com, which I want to get the backstory on, for so long. How long have you been writing there, by the way?
Almost 20 years. I started when blogging was cool and I'm still waiting for blogging to come back and be cool again, but it's not yet.
I think it is cool, when newsletters are cool now. Yeah,
Newsletters are cool.
I don't know if you saw, Twitter now is encouraging long-form writing. There's like this articles feature. So I think it's cool. I think you've survived the
Rough. Okay.
I was also talking to Gemini trying to figure out how many posts you've written. I was like, count the number of blog posts on asmartbear.com. Do you have a sense of how many things you've written on there?
Yeah, it's not that many. It's something like maybe a hundred, well, I would say between 150 and 200 that I'm proud of and probably about 300, 350, and that's it over, you know, about 18 years. And that's because I only write in depth. Some are long, not all are long, but none are short, I guess.
And I've always had a rule, even though you're supposed to write really regularly, and not just for algorithms, but they used to say, again, back in the aughts where I started, oh yeah, it needs to be really regular so people know when to expect your thing and they plug it into their day and all this. So it's always been true that you should be regular, and I never was because my attitude was always, I will only put out stuff if it's the best that I can do. It's up to the reader to decide if it's good or useful. And if I don't have that, I'm just not going to publish. That's the way it is.
And so there's years where I've published once or twice only the whole year. Maybe I was busy or didn't have the energy. Other years, yeah, I posted, you know, 40 times or something. But even then it's only that because I can't do something of that magnitude and also found unicorns, which I did during that same time, and run them. You know, I can't do that all at the same time and produce a lot. So it's fewer and hopefully better, but that's in the eye of the reader, of course.
I like when you say I've done 300, not too many.
Well, not for 18 years, right? Like over that time you
But I think this actually, this is where I was going to go, but I think this is a really important lesson I've also learned. I always used to tell people the key to being successful writing stuff online and just content in general is quality and consistency. But I've just more and more realized quality is actually the only thing that matters and the consistency doesn't matter. So the only difference is, the more rarely you write, the more awesome it has to be.
It is a lot of pressure. I feel that, and then I tell myself that will just prevent you from writing anything, and that's not good. So yeah, you tend to want every thing you make to be the best thing you've ever made. And on the one hand, I want to hold on to that because it's motivation to be good and not to let the bar slip. On the other hand, it can go into paralysis, which is obviously bad. So yeah, I still struggle with that, but I think that is the tension.
Okay, so with 300ish posts, 200 you're proud of, there are so many directions we can go. There's a few that I've picked that I want to spend most of our time on. The first is you have a really pragmatic way of approaching growth stalling. And the reason I want to spend time here is because a lot of product teams, a lot of founders build something, it starts to show some success there. It's going, it's growing, and then all of a sudden it just stops growing. And I think that's one of the most painful things to go through.
And I've never come across a way to think about how do I solve this, because I think a lot of people are just like, okay, I guess that is not working, let's move on to something else. You have a very specific way of approaching this problem.
And I want to read actually a quote from Will Smith, and this is something that has stuck with me ever since I read it because it's so true. So in his biography he has this line. People ask him what's it like to be famous, and his answer is: becoming famous is amazing. Being famous is a mixed bag. Losing fame is miserable.
That's funny. So first of all, I think a lot of people are experiencing this right now. You have a lot of companies that have reasonable products and their growth has slowed. Why? Could be the economy, because it's not as good as a lot of indicators say. We all know that, for example, jobs are not as good as the indicators say. It could be because of AI or the threat of AI or the expectation of AI, blah blah blah. Who knows?
It also can just be size. As you get bigger, growth slows because, you know what? You're not going to grow 2x a year forever. So it slows. There's just mechanical things. So there's many reasons why things slow, and sometimes it's all of a sudden, although then maybe there's some event like an algorithm changes or something happens. But actually I think what's really common is it just slowly gets slower.
In other words, it decelerates, but it kind of, I wouldn't say sneaks up on you, because most people are looking at growth all the time. So it's not sneaky, but it is a little bit more gradual, right? You just feel more like you're running through mud, like, ah god, we're still doing so much work and it's not having as much of an impact. And so that's what I see.
And when I say that's what I see: so I've built four companies. The last one is a unicorn. The one before that was also a unicorn. The previous one was bootstrapped. This one was VC funded. And I've invested in about 60 startups. Some of them failed completely. Some of them were very successful, some in the middle, of course, right? And so when I say that's what I've seen, that's the context of what I mean by what I've seen.
So there's, I wouldn't say a checklist, but there's a series of questions that I ask to diagnose why is growth slowing, in this order, because it's one of these things where the first one that's a problem, if you don't fix that, it doesn't matter if you fix one of the ones below. Just like, I don't know, maybe if you had a marketing funnel and there's a step where everything falls apart and you're like, well, I'll just tune the bottom of it a little. It's like, that's not going to work. It's not going to help enough. You got to go where the biggest issue is. So this is in that sort of order.
So the first question is, are customers leaving? I.e., logo churn, right? Churn with N. You can do churn with MRR too, but just for simplicity, let's say with customers. And it's the worst problem for a couple reasons. One is there's nothing you can do about it once it happens. They're gone. There's no saving them, increasing their revenue. There's nothing in the future you can do. Also, it's often correlated with things like negative reviews or other things on social media, which is another kind of preventing growth. So it's kind of the one-two punch thing of like, they're not here and they may be actively hurting your growth. So that sucks.
The math is undeniable, which I want to talk about because there's a metric I like that is unusual and people find useful. But before I get to the metric, there's also this kind of visceral thing, which is the customer saying, this product, I don't want it.
And when I think about the gauntlet they went through to get to the product: how do they even find out about me? That was hard already and improbable, that they see an ad or hear it. And then they clicked, which is improbable, and then they didn't just bounce off the homepage, which is again improbable. They actually were like, oh yeah, this sounds pretty good. And then they got to the pricing page and that didn't scare them off. They actually had the budget and bought the stupid thing. Then they went through onboarding and invested their time, etc., etc. That is a crazy gauntlet that almost no one gets through.
And after all of that, which clearly means they wanted it to work, they're like, "No, bye." What? Like, just on an emotional level, you gotta go, "Wait a minute, that's terrible. I'm fundamentally not fulfilling whatever promise I made or they thought I made," whether that's a product issue or a communication issue. Okay? There's lots of... But one way or another, something is really fundamentally broken.
Just in terms of, I'm a product person. So what I want to do is make a product that other people want to buy and use. And if they don't, no matter what the metrics say, we're failing our mission, our customers, whatever. So there's even that non-mathematical reason to go, oh my god. Right? So to me that's already enough reason, but the math is very interesting.
And what I find is when I talk to people, especially on Twitter or something where people are just, you know, yapping around whatever they're doing, I say things like, you know, anything above 3% per month cancellation is terrible. And people are like, oh no, it's okay. Five is fine, seven, six. Everyone's yapping about what they... And it's very abstract. Like, is four better, much worse than five? I don't know. And I heard someone else, and blah blah blah. So it's very, I don't know, generic and rough.
So there's a different metric that I like to use, which keys off of this idea that I think, again, people don't appreciate, which is cancellations grow faster than marketing. And so cancellations overpower the growth of the company and slow it to a halt, i.e., growth slows, right, to where you literally cannot grow anymore. There's a maximum ceiling of how big you could ever be thanks to cancellations. And when you know what that number is, it's much more real and visceral and scary. And so, just to kind of justify
What I just said, just imagine any company and imagine you just tripled the number of customers that are there and paying, the same kind, the same age, you know, just the same kind of stuff, just tripled, right, overnight.
So, the next month, would marketing deliver more new customers than a month before? No, because marketing doesn't, none of your marketing efforts care how many customers you have. AdWords delivers the same number of leads and, you know, SEO delivers the same, like it does not care how big you are, these efforts. So you're still going to be growing at the same rate as you were the previous month, but cancellations in absolute terms, like the number of customers who leave, will triple, cuz you have 5% cancellation and triple. Okay, so still 5% of a tripled number is triple, right? So this is the point, is that cancellations automatically grow as you grow, even if you're doing everything right, but marketing doesn't. Marketing grows only as fast as you can improve marketing. We all know that's quite hard actually. It's linear. It's hard to find new channels that aren't trivial. Like, it's hard. Of course, we're going to do it, but it's hard. Whereas cancellations grow automatically as you grow, right? So cancellations always overtake marketing for this reason.
Like the metaphor here is a leaky bucket, where are you adding enough water to keep up with the leak, essentially.
Right? Except the leaks automatically increase, and that's what people don't appreciate.
Because it's a percentage of your entire customer base.
Yes. See, in marketing we say things like I'm adding 100 leads a month, but in cancellations we say 5%. Why do you say percent? Because it's based on your size and it's exponential. That's what 5% is, an exponential. And so there's this maximum size you could ever be. It's when churn equals growth, right? Like that's the m...
So how would you compute that? It's actually quite simple, because let's say you have this 5% per month just as a number. So it's simply the amount of new customers you add divided by that cancellation rate. That is the limit. So let's suppose you add 100 customers a month and you have 5% cancellation. So 100 divided by 5% is 2,000. So a company like that will never have more than 2,000 customers. And by the way, as you approach that number, growth is very slow, because you bring in a bunch of customers and almost the same number leave. So growth is slowing. Ah, look, we diagnosed why growth slows automatically at all SaaS companies. So that's why this is the first thing, because it's such a hard cap limit and it means that people don't want your product. These are two reasons why it's the most important thing.
Just to clarify, this is logo churn. This is like number of customers, not revenue churn.
Yeah. Well, it is both logo churn and revenue churn. You do the same math. You could say dollars in divided by dollars cancellation rate, or number of... I've been saying number of customers just to keep it simple, because I think when you look at it and say wow, we will never have more than 2,000 customers, it's just such a visceral "oh my god, we had to do something about that." Now of course one thing you could do is have more marketing, but you know that already. If growth is slowing, you're already thinking how do I get more out of marketing. You knew that. The point is that cancellation is this hard limit pulling you down, with all these other really bad either implications or side effects, which is why it's so important.
Cool. And when you say marketing, just to clarify, this includes basically all growth work, PLG stuff, marketing, sales.
Yeah, right.
Great.
Yeah. PLG is nice, but you still need marketing to bring the people in in the first place. PLG just means there's not a salesperson, unless you're expanding or some other segment.
Cool. Yeah. It's like the whole bucket of just bringing new customers in.
Yeah. Yeah.
So, okay. So assuming you agree, like, "Yeah, I don't like customers leaving. That sucks." So obviously you got to find out why they're canceling and do something about it. And the kind of root issue here is they don't want to tell you, like they're already out the door. They've already stopped investing in you, like, mentally. So the last thing they want to do is spend time with you or really think about it and diagnose it with you.
And I have a funny story about this for myself. So at Smart Bear, people would cancel. We put up this form and a drop-down list, you know, too expensive, project ended, you know, this stuff, like we do so we could gather data, and one of them did have more selection than the rest. And I realized it was the first one on the list and I thought, huh, I wonder if people are just picking the first one. So then we randomized the list so everyone saw a different order of the list, and now all the items were picked equally, like, oh, right, it's complete noise. And I know other companies have done similar things also with similar results, that this is a global phenomenon. So okay, so what do you do? Like, the point is it's hard, right?
So the first thing is you want to ask open-ended questions. I know you want to just get a list, but this is the problem. At least with open-ended questions, most people won't answer, but at least you might be able to get some kind of thing that they generated. And when you do this, the wrong way is to ask "Why did you cancel?" Because again, this allows them to say something really simple like budget, which may or may not be true. I'll get to that in a second. What you want to do is say "What made you cancel?" In other words, what about the product or situation or whatever caused the cancellation? Just phrasing it that way, you get much better results. And I stole this from a company called Groove, who has this great case study online about this very thing. They had an email that they sent out, which is a great email, and they started by asking "Why did you cancel?" They got 10% usable responses. They changed it, same email, to "What made you cancel?" and it's 20% usable responses. So this is, there's like, I guess maybe some anecd...
...as far as you can into there, because most people won't talk. The temptation is to hear what they generate at first and say that's the answer. So a really common one is it's too expensive. I think anyone who's looked at cancellation data at any company will agree that too expensive is often the number one or at least top three reason in one form or another. And that is never ever ever the reason.
How do I know? Because they already looked at your homepage, read all the stuff, saw what you promised, looked at the pricing page, and decided to buy it. That means whatever was in their mind of what it is is not too expensive. They already decided with their actions. It was not too expensive. Something else happened, like you didn't fulfill the promise that at least they thought you made, or something else didn't work, or, you know, now it is possible they lost budget, but that doesn't mean you're too expensive. That means they lost budget. That's a very different reason, right?
That's not that you're... So it's sort of like, this happens in health care, for example. So when someone dies, the doctor has to write what's called approximate cause, which is why did they actually die? But then you try to also write down the real reason. So let's say someone comes in and the approximate cause of death is they stop breathing. Well, you could stop there, and that's like listening to "it's expensive" and going that's it. Well, why did they stop breathing? Because they ran their car into a telephone pole and were injured so much that eventually they stopped breathing. Well, why did they run their car into a telephone pole? Because they passed out at the wheel. Why'd they pass out at the wheel? Cuz they had undiagnosed diabetes.
Now we're getting somewhere. It still isn't just one root cause. As a sidebar, I hate the idea of a root cause. Complex systems do not have one root cause. They often have many interlocking things that could be done to detect earlier or to change it or to reduce, and not one root cause. So root cause analysis to me is, by the way, an incorrect thing. I'm explaining why right now with the health care, right? Because, well, what about the second diagnosis? Well, maybe part of the problem is we have a health care system that isn't preventative, and part of it is that, but they didn't go to the doctor anyway, and, you know, okay, so there's all kinds of things that could be useful and interesting to prevent this or make it better. That's the point. That's what an analysis should be, is this array of things, not the root cause. Anyway, something along the lines of undiagnosed diabetes is much more of a cause than stopped breathing.
So when we say it's too expensive and that's the reason, you're making this fallacy. You got to go into, well, they wanted this stuff, but it didn't work with Linear, which is what they use. It only works with Jira. And so there's a lack of integration. Now, maybe we should write that integration, maybe we shouldn't. Of course, it depends on how much we hear about it, and, you know, of course it's going to depend on other things, but that's the reason, not "it was expensive," right? And so this idea of getting into not even the root cause, but let's say rooter causes.
The root.
Yeah, the more root. I think some people probably say five whys and just paper over what I just said with that, and maybe so, but let's not be so simplistic about that, because again, five whys sometimes implies that there's some root cause at the bottom of the whys. Let's be a little bit more... Let's be a little more smart about that.
So anyway, these things, too expensive, this is not it. Maybe project ended really is project ended. Okay. But even there, I see just today on an entrepreneur forum, someone said, "Yeah, you know, we're starting to see more people have project ended as the reason, and so there's nothing we can do about that." Now see, that's incorrect. That's only true if you only look at the approximate thing, which is project ended. You're correct that you can't make that project not end. Exactly. Yeah. Okay. But wait a minute. If your software was more successful and the project was more successful, would it have ended, or is that actually an indicator that your product wasn't that useful or didn't do its job? It's possible. Like in this case, who knows, right? But it's possible that it really is your fault.
Another example is, but you picked what target segments you were going after. Did you pick a market segment that was easier to sell to but their projects end, like small business and consumers, where very often the small business does go out of business or the project ends, etc., because when things are small they have high variance and lots of things can knock them off the path and so on. And so is it your fault for picking the wrong ideal customer profile or target segment? And so yes, that one case of that one project, it's not your fault, quote unquote. But by saying that, you're just ignoring the fact that there is maybe something to do about it. Now, all this is maybe. None of this proves you should change your market, right? But when you say there's nothing we could do about it, you are closing the door on these things that might be the right thing. And very often, as I think probably a lot of people listening to this know, the market segment you pick has a lot to do with your retention rate, because everyone acts differently, right?
And so anyway, I know it's a lot on this topic, but I just feel constantly people make this particular mistake of not getting, you know, just abdicating responsibility or just listening to the first thing they hear and saying that's the reason, and that's not right. So that's the big thing about listening.
Another thing is you got to ask when people are in trouble but not yet canceled. You might be able to save them. You certainly can learn more, because you can talk to them, like they're not shut off yet from you. So this might be they never uploaded their data, so they're not being successful. They are calling tech support too much. They're in trouble. They're not calling tech support enough. You know, they're not engaged. They didn't log in for a while. Like, there's all kinds of things where, now of course the details are going to depend on the product, obviously, but there are signals that are correlated with cancellation. Now, if you have a lot of data, you can literally correlate signals with cancellation and try to extract that, you know, precisely. But even without data you can guess, and guessing and having a theory, acting accordingly, and as you get more data adjusting your theory, this is a wise way to proceed even without data. So if you can catch them when they seem like they're off the happy path, they're in trouble, that's a better time to do it.
And then the last thing I would say about this detection is if you don't know what to do, or all else being equal, then focus on onboarding. Almost all companies have a whole lot more cancellation in the first day, 30 days, 90 days, depends, right? But the first period, than the whole rest of the customer's life. And also small changes in the onboarding can have large effects on cancellation. Whereas later on, that's not necessarily true. It could be, but it's not necessarily true.
So a really dramatic version of this is if you've ever done YouTube videos, which I know you have, but if a listener has ever done a YouTube video and you see the retention, quote unquote, of the viewer on a YouTube video, it has this thing where it falls just so much you can't believe in the first 30 seconds, and then if it's a decent video, it'll flatten out as people, you know, decide to watch the video. So in that crazy looking curve, for the people that watch it for 15 minutes, maybe there's something you could do to keep a few of them staying to the end, but that's not going to change very much how many people get to the end. Whereas, like for me, I've only done a few, but what I see is about 50% fall off in the first 30 seconds. Well, if I can get that from 50% to 55% stay, that's an additional, and at the end of the line, I only have 20% still there, which is pretty good for a longer video. But if I get it from 50 to 55, I might get it to go from 20 to 25% staying. In other words, if I shift 10% at the front, which maybe I could do, like I can't be dramatic, but maybe a little, then in the output I might be able to increase it by 20, 30%.
So that's a huge change. And the SaaS equivalent is, as we all know, if they leave early, not only is it bad, but it's super unprofitable, because you spent all this money to acquire them and then they never stayed around long enough to pay it back, much less to be profitable. So if you can do a little bit in the onboarding, or shift the onboarding percentage a little bit, it pays off enormously in revenue and profit over time by making them successful. And so again, if you don't know what to do, onboarding is a good bet. And even if you do know what to do, I'll still bet that onboarding is a good bet for where to go.
Oh man, I'm so happy we're spending so much time on this very specific first step of logo churn, because the way you described it is so visceral. It took so much. It's like impossible how far this customer got already. Like, they are using your product and understand it
Mostly, and then they still decide to leave. So brutal, the way
And you're going to believe them when they say it's because of the cost,
Right? Like it just doesn't even make sense when you put it that way, right?
So, let me kind of summarize the advice you shared here because this is so good. So, step one is look at logo churn. The way to understand, and essentially to understand how big of a problem this is and why you need to spend time here, is look at this, basically do the math: how many new customers you're getting divided by the cancellation rate, and that essentially tells you, if that doesn't change, what's the maximum number of customers you will ever have.
Exactly.
That's going to be a sad number, and then the question is, okay, cool, how do I reduce the cancellation rate? Obviously, as you said, everyone wants new customers, more new customers.
Yeah, and I know you're going to do that anyway, but you got this cap.
Exactly. Okay. So, a few things you've shared here. One is instead of asking people multiple choice, why did you decide to cancel? You make it free form and you make the question, how would you say it? Was it what made you cancel?
What made you cancel?
What made you cancel? Great.
Yeah.
And then you could use AI to help summarize these things, I imagine, instead of
Yeah. I think what I find with AI with this sort of thing, with surveys, is this.
AI is good at picking out themes.
Yeah. It is bad at picking out details that are actionable.
When I say AI, of course, I mean LLMs, which is probably what we mean when we're looking at natural language, right? And if you think about it, it sort of makes sense because the LLM is an averaging machine, right? It's predicting the most likely; that's an averaging kind of a thing. And so when what you're looking for is a kind of average, it's usually pretty good. So, summarization, topics, themes. But when you're asking for like what is interesting and not average, it's actually pretty bad at it.
One way that I found that's sort of useful is yes, I'll ask it about themes, but then I'll say now pick out every specific detail that goes under one of these themes, put it along with like which customer said it and the link to, you know, blah blah blah. So, you'd have to, you know, play with this to tune it, right? But like that kind of thing so that a human being can then still see the detail, which is what triggers in your mind, wait a minute, but that means we should do that, right? Because the topics won't do that. The topics will be, I already know what the topics will be. It'll be stuff like I couldn't figure out how to do this integration, right? Like the topics are actually not going to be that surprising, probably. It's the details that are going to be the triggers for actionable stuff or patterns or something like that.
So yeah, AI is not useless, but it's not as useful as it sounds. It's probably still a good idea to just read all this stuff. Although AI might be able to clean up, you know, maybe people's grammar is bad, it's a weird language. Okay. Yes. Like that's annoying. You could clean that up, but I wouldn't rely on AI to do the thinking for that reason.
That's such good advice. I actually have a really cool guest post coming out soon that gives a bunch of really specific techniques to avoid AI hallucinating or just giving you really bad results from this very specific synthesis work, because it turns out AI is not great at actually being honest about some of this stuff. So we'll link to it if it comes out before this. And like I think in real life, most people don't have that much, the volume of these cancellations, unless it's like a super consumer app, is not that high. So, you don't even need AI for this. Just like read it.
And then this is a good segue to your next piece of advice, which is essentially the five whys, but not the five whys, where you kind of force yourself to dig into what's the real reason that forced them to cancel. It's probably not pricing. It's probably not the project ended. There's something deeper.
Yeah.
And then advice number three is try to catch people early. Try to catch them before they churn. If you don't have a lot of customers, it's a lot easier. If you have a lot, it's obviously harder. There's always been this like holy grail idea of a product that just like watches metrics and tells you this person is about to cancel. I see that.
What I would say is it is not hard. You don't need a lot of customers to go talk to the ones who are in trouble.
Mhm.
You do need a lot of data or customers to mathematically know what behaviors are correlated with cancel, and therefore to spend your time wisely. Then you need a lot more data. But to your point, even if you have the data, it's not entirely clear whether some kind of mechanistic thing is all that important.
One way I look at it is, you know, it's very common advice: you should try to get more good customers and fewer bad customers. Of course, you should. And so therefore, they say you should see what the good customers have in common. But that's not the end of the sentence, because a lot of the things the good customers have in common, they also have in common with the bad customers, because it's just what your customers do, just what anybody does. So, it's what the good customers have in common that are different from what the bad customers have in common.
Okay. So, with that in mind, it has to be both, or else you're just getting correlations that are not helpful. The cancellations, or talking to people who are in trouble, is another application of that. So, what is correlated with people who actually end up cancelling, not just what you know. And so I think that mindset is correct if you add the other side of that to it.
A really important nuance.
Yeah.
Okay. And then the final step, just to close this out, is onboarding. Work on onboarding, activation. Something that's one of the most recurring themes on this podcast is just the power across every dimension of improving onboarding, improving activation.
Yeah.
Sweet. Okay. So, this is just step one, which is already full of gold if your growth has slowed. So, step one is focus on your logo churn, the number of customers leaving, people leaving, actually canceling your product.
So I kind of look at it like a question. So like the first question is, are people leaving too much? Because if your monthly cancellation is 2% for SMB, that's good. So you could try to work on it, but since it's already good, it's still probably a good idea to work, it's probably a good ROI for you to work on it, but it's possible that you've got diminishing returns and that this isn't really the reason, or it's not really reasonable for it to go. I mean, how low can it go for SMB? Like there's some floor and you might be near it. So the first question is like, is logo churn too high, and trying to set a threshold that's, you know, lower than what people normally want to do.
So the next question I have is, is the pricing correct? Which of course pricing is a perennially interesting topic. I know there's this funny thing, especially with newer companies, that the pricing is always too low. It's not always, but like that's the common thing. Patrick Campbell, who has 4,200 data points about startups, let that sink in a little, has this great quote, which goes like this: your prices are way too low because you just guessed and you haven't changed them. It's like, yeah, if you really look deep within, you realize like, yeah, or we just picked whatever our competitors are doing and that's it, or we added or subtracted something because reasons, and right, that's probably not good.
And people are scared to raise prices for obvious reasons. But if we set aside the emotional reasons, whether they're correct or not, the sort of economic reason people normally give is they have in their mind this microeconomic supply and demand curve thing. And the demand curve says that if you raise the price, demand goes down. That's why the demand curve is always going that way, right? And so they understand, I think everyone understands, right? But maybe you raise prices by 10%, but signups go down only 5%. So overall, it's better. But the opposite could happen too if I'm on the other side of the demand curve. And okay, so that's how most people think of it.
However, this is not how it works. So that's how it works in microeconomics 101 textbooks. That's not how it works in the real world often. So what often happens is you raise prices and signups don't change. When I say signups, I mean like signups per month, you know, the rate of signups. Or signups go up. This happens all the time. Even for like solopreneurs on Twitter who have, you know, strange projects or everything. Happens all the time. They raise prices. They're like, I was scared, but then signups went up.
I once talked to a guy. This is really funny. I'm going to not say the name to protect the name, right? But so he had a product that he was selling essentially to enterprise and government, so larger companies. And it was, to me, way too cheap. So he said something like, "Yeah, I charge $300." I'm like, "$300 a month? That's not enough." He goes, "No, per year." Like, "Okay, wait." I said, "Okay, just do me a f... how many signups do you get a week?" And he's like, "One or two?" Because this is enterprise and it was a startup. I said, "Okay, just for fun, just change it from per year to per month." So, in other words, we're 12xing the price, right?
So he did, and he still got one or two per week. Like nothing changed. I'm like, "Okay, what are you going to do next?" And he goes, "Oh my gosh, well now I have so much more money and profits. I'm going to like hire an engineer. I'm going to do this marketing." And I'm like, "Time out. What you're going to do is raise prices again." Like you just told me you 12xed the price and nothing observable changed. That means you're not near the price yet, right? You don't have to 10x it again necessarily. Maybe 2x, maybe 50%, but like you're not done. I mean, you can do those other things too, but you're not done with the price. Like, it didn't even occur to them still.
Okay. So, why does this happen? The reason is that pricing selects the market. So if you only think of the market as people with very limited budgets, barely can do anything, not getting much value out of it, then it is true that if you raise prices, you'll get fewer of them, because they were never getting that much value out of it anyway. They don't have that much money. So if you raise prices, they're gone.
But think about just even a mid-size company, forget about enterprise, just think about a company with a thousand employees and 400 million in revenue or whatever. And if they see a product that's, you know, $2 a month or even $100 a month, their thought is like, well, that can't be good enough. They're not mature enough. It's not going to do enough. The support's not going to be good enough. They probably don't have good governance policies or other things that we need, you know, etc. Whether that's true or not, like this is what it looks like: it's low-quality, cheap, whatever, aimed at SMB. So, they just won't buy. They're not in the market for the thing.
So it's not true that they have this demand curve where, oh, since it's cheap, they all want it. That's what the microeconomics curve says. It's so cheap that they should all want it. No, they don't. None of them want it because it looks bad. So as it gets into a price range that makes sense for the kinds of things that they need, then their demand actually goes up. Then it can stay up while it's in a good range. And then of course at some point you are priced out of them; that particular kind of company's like, I'm not going to spend $10 million a year on it. Are you kidding? So yes, it does slope down and go away. So it's not a normal curve, but it is like it slopes up and then it's something and slopes down. Who knows exactly what shape it is? Probably none of us know, but it's more like a mesa and not a line that goes from up to down like we see in the textbook.
For that market, it's only the very lowest, you might even say worst in terms of metrics, end of the market that has the microeconomic slope that you're worried about. So what happens is you raise prices and you enter a different market, and that's why the signups go up. Or, okay, you leave behind perhaps a worse market anyway. And of course everyone will tell you, you know, the more they pay, the higher retention is, and that, you know, all the kinds of stuff gets better when you charge more.
So this question, is pricing correct? This is kind of what's in my mind when I ask that question. It's like, probably the answer is no, because pricing is very hard. It's just as much art as it is science. You've had some really good people on here on pricing. In fact, so good that I've bought some of the books that those people have talked about because I loved the interview so much. Right. So I believe in all that. No problem. I believe in it. Nevertheless, they also say it's art and science and it's very difficult to, and also once you augur it in, the world changes; like 5, 10 years later, the market is different. The world's different, and so it's still unclear.
Also, price is not just the number on the web page. It's easy to think that, right? But how it's structured is just as important. How the product's positioned is just as important. So, for example, this example I've written about before online, it actually was something that happened in my life, but I changed the story to make it simple and real and clear without having to get into lots of detail. So, the sort of story version is how this company was able to charge eight times as much for the same product just by talking about it differently. So just by positioning it differently, eight times as much. Again, this happened to me, but it's too complicated. Those details are not interesting.
So say there's this company called Double Down, and the idea is that it halves the cost of your AdWords because it makes it so efficient. So that's what it says on the web page: cut your AdWords cost in half. Which is a very good pitch, isn't it? Simple, obviously valuable. But let's suppose I'm a customer and I spend $40,000 a month on AdWords. What am I willing to pay for Double Down? Well, if you do cut my AdWords in half, then all right, I save 20K, but I'm not willing to give 20K to Double Down, because then I'm not saving any money. In order to actually save money, I need to give Double Down less money. How much less? I don't know. Let's just call it a quarter. So, I pay Double Down 5K to save 20. So, I'm really saving 15. Double Down's making 5K a month. That's pretty good. Everyone's pretty happy at this 5 grand a month price point.
So, there's nothing wrong with this. No one's doing anything wrong. Like, that's a perfectly valid company. However, think about these two situations that the CMO or the chief product officer might be in, in talking to the CEO at the end of the year. Well, scenario one goes, we started using this tool, Double Down, and it halved our cost. So, we were able to spend that money on some other stuff. We were able to save money. And the CEO would say, great, that's good. We're going to renew, and I'm happy to hear it. Again, nothing wrong here.
But let's take a different tack altogether. What does the CEO want to hear more? Growth or saves money? Both are good, but I know which one is healthier for the company, increases market share, is better competitively, and also makes the company more valuable. It's the growth. So, what we'd really like the CMO to
tell the CEO is, I increase the growth rate of the company, not so much I save money. That was way better. So, here's how we could do that with Double Down. Yes, Double Down halves the cost, but what that means is right now, right now, the company is paying 200 bucks per lead. Let's call them leads, right? Whatever this is outputting, right? Well, if I halve the cost of a lead, I could get twice the leads for the same money. I'm already willing to spend $200 a lead and I'm already spending 40k a month for it. So, if Double Down halves the cost, it means I can get more leads.
So the way I could pitch Double Down is double the leads per month, period. Now if I'm willing to spend 40k for this number of leads, how much am I willing to spend to double the leads? 40k. I just said I'm willing to spend 40k for this number of leads. So doubling it, I'm willing to spend 40k to double it. So if I give Double Down 40k, not five, for the same product, which is the leads are cheaper, but then now the pitch is I doubled the leads for 40k instead of halving the cost for 5k, so Double Down gets 8x the money because it gets 40k for this product, not 5k for this product.
Amazing story.
And everyone's happy because the CEO goes, what did you do? And they said, oh my god, I doubled leads. What? Yeah, I mean at the same ROI as we had before, same CAC, I doubled leads. CEO goes, how could we do more of that? I mean, just everything is so much better, same product. Now, I know it's a little bit of an exaggeration, etc., because I'm trying to make a point, right? But the big point, or the largest point, is pricing is not just the number on the page. It's positioning. It's how their budgets work. It's how it's structured, like it's per site or it's per usage or it's per seat. All of this stuff is part of what pricing is. And often even if it's the same price or the same product, depending on how that's structured, it either seems fair and good or it seems unfair and too expensive or whatever.
And so in particular with the positioning, the big lesson for product managers is sell more of what the company values, like growth. It doesn't have to be growth. It could be something else. The retention for their customers, how competitive they are in the market. There's various things they could value. Growth is an obvious one. Sell them that they're going to get more of what they value, as opposed to saving, cutting, ROI, saves time, saves money, more efficient. And again, there's nothing wrong with saving money, saves time. It's just that it caps this price and this value that they perceive that you do. Whereas if you deliver more of the value that they already value, I don't want to say uncapped completely, but the cap is maybe an order of magnitude higher than saving.
So again, it's valuable to save. You're not doing anything wrong. Like, that's not how to talk about what it is and therefore help set the price. So yeah, it's a long way of saying, again, when I think is your pricing correct, I'm thinking in maybe a more general way than just the number. I'm thinking about the structure, the positioning and all that, and my guess is when growth is slowing, there's a lot of improvement that could be had there.
Oh man, this story is so powerful. Who does not want to change some copy on their website and double their growth and triple their price? And the biggest takeaway here is when you say is pricing correct, it isn't like what is the number, 20 or 25 or 100. It's almost, is the market we are going after correct? Is the way we are selling to them correct? Is this price communicating the right sort of story? And then also, is the positioning of what problem we solve for you correct? So there's a lot here, and luckily I've done a bunch of episodes along this stuff, which we'll point people to, to go much deeper, because this is a very deep skill and there's a lot to do here.
One thing I'll mention specifically, so Jen Abel, a recent podcast guest, it was her second visit to the podcast. She has a lot of really good advice on this of just how to price and how to reposition the way you're selling it. Specifically, she had this really interesting insight that enterprises, their sweet spot for contracts is like 75 to 150K. That's how they normally buy SaaS software. And it sounds absurd, but you want your product to be in that bucket versus like a thousand a month, 2,000 a month, and everything just gets easier if you're like, "Okay, this is one of those. Okay, cool."
It gets easier. I don't want to go off too much of a tangent, but the thing you have to remember is that pricing is not this knob that you can turn separately from the rest of your strategy.
So when you say, even when I just said raise prices or whatever, you can't just raise prices. These new customers have different demands. Now maybe you need SOC 2. Now your other governance stuff matters. Now integrating to certain systems you didn't know about matters. Maybe now they need professional services. You can't just raise prices and change market and that's it. And maybe that's wise to do, but maybe it's not. Maybe you realize that, sure, of course, that other market has certain advantages, but they also have disadvantages, and we don't want those either because we'll no longer be competitive, because the way that we're distinguished, competitive, special, interesting, valuable is only valuable in the market we're in. The next market does not value it like that. And so, uh-oh, actually, that would be really bad for us.
Or it could even be cultural. Buffer is a great example. Buffer could go up market and try to sell, you know, social media tools or whatever. But they realized we are a company for the little people. I don't want to sell to a big company. We're never going to make a product for them. We don't want to. This is who we are. This is what we want to do. This is what's fulfilling to us. So, we're not going to go there. So, it can be cultural. It can be certain goals. It can be other aspects of the business model or the strategy, but it's not this kind of like, oh, I'll just change this. It's a decision about the whole strategy, and that too we could talk about for hours. But let's just caution that. "Oh, I'll just go enterprise" is of course not how it goes. Sorry. We got this.
Yeah, I love Jen. Her pricing and sales stuff is good. She's great on Twitter, too. I love her. I love Jen. And that is such an important nuance, you know. Don't build the thing you're miserable building and just like, okay, I listened to a podcast, we're going to raise our prices 10x and life will be grand. There's also downsides.
Yes.
Yeah. Okay. Amazing advice. Okay, there's so much here. Again, this could be its own conversation. Pricing, positioning. Yeah, we'll link folks to a bunch of cool advice that we've covered on this podcast, too.
If you're a founder, the hardest part of starting a company isn't having the idea, it's scaling the business without getting buried in back office work. That's where Brex comes in. Brex is the intelligent finance platform for founders. With Brex, you get high limit corporate cards, easy banking, high yield treasury, plus a team of AI agents that handle manual finance tasks for you. They'll do all the stuff that you don't want to do, like file your expenses, scour transactions for waste, and run reports, all according to your rules. With Brex AI agents, you can move faster while staying in full control. One in three startups in the United States already runs on Brex. You can too at brex.com.
Let's keep going through this checklist. So, one was look at logo churn. Yeah. Two is look at is your pricing correct?
Is pricing right? And why do we think it is? Because we probably don't have good reasons yet.
Okay. So, the third one is are existing customers growing?
I think everyone probably knows this, but just to say it. Okay, if cancellations overtake marketing in magnitude, one way to combat, I mean one thing is, okay, make cancellations lower, but they can't be zero. So what else do we have to combat cancellations that would be proportional to our size, so that it keeps up, unlike marketing, right, or you know, basic direct marketing? So one would be, all right, 2% left, but of the remaining 98%, some of those upgraded or otherwise paid us more. Maybe it's usage based, whatever it is, they're paying us more. And so that covers the gap, and yes, if I tripled the company overnight, that would triple. And so that's the answer. So that is an answer. And maybe that's obvious, but you know, it's useful to tie it back to the sort of mental model we've got going.
And of course the metric here is NRR, net revenue retention. And the way that's computed is you say, what is the revenue of customers right now, so existing customers, just the whole total, and then one year from now, what of that remains? So not new customers coming in, not talking about them, because we're asking about the cohort that exists. What remains? So with cancels it goes down, with downgrades it goes down, but with upgrades it goes up. So when I say remains, it could end higher than we started if upgrades exceed cancellations and downgrades. And now we are talking about MRR and not N, because N doesn't have an upgrade and only goes down, which again is why I think the N is actually the most important one.
Because think about it, a lot of times people think, so if you've heard of NRR, you might think, well, that's my golden metric, I'm done. But the issue is if NRR is positive but N goes down too fast, it doesn't matter, because not enough people are left. And so there's not enough people left over to upgrade, and so actually you're wrong. And so NRR does not include that, and therefore it actually undercounts what's going on in a bad way, in a way that hurts you.
There's yet another way to see why what I'm saying is right. You know, there's this thing in investments where, let's say I started out at $100 and the stock goes down 5%, so now it's at 90, or no, it goes down 20%, now it's at 80. Then it goes up 20%. Is it back to 100? No. Because 20% more than 80 is 96. So if it goes down 20% and up 20%, it does not come back to zero. It's worse. When you have a percentage loss, you have to have a greater percentage gain just to get back to where you were. In this case, a loss of 20% requires a gain of 25% to get back to where you were. This is why NRR isn't quite right, because NRR is saying that a loss of 20% from cancellations is offset by 20% from upgrades. As we just saw, no, it's not. That only gets us to 96% actually. So, this is why, again, I believe in NRR. I'm saying you've got to track it, it's good, just in the back of your mind realize it's not quite that good, and looking at N keeps you honest about what's really going on with these customer cohorts, right? So that's why they're both useful, in fact.
So NRR of course is important. A nice way to see this is, if everything I'm saying is true and there's these limits and stuff because of cancellation, then there should be no way to get a big company like a public SaaS company unless NRR is greater than 100. Otherwise cancellation should just win. And that is in fact the case. There's over a hundred SaaS public companies and something like two of them have NRR less than 100%. That's how it goes. And those companies have horrible financials and their valuations are bad. It's not a good thing, right? And in fact the median for an IPOed SaaS company, at IPO the median NRR is 119%. So yes, that's what it takes. You can't do this. You're limited unless. Now your goal may or may not be to get that big, but the point being it's mandatory for growth. Now if the literal customers are just leaving, you've got to plug that hole first, like you said. But okay, that's why this is in order. If that's generally okay, now we turn to NRR to say, okay, but the ones who stay, they're hopefully happy, they need to grow. Okay, so that's kind of the full story of NRR. I think people who've heard of NRR don't necessarily think about all those things and realize that.
So a good question is, okay, what do I do with NRR? But I think the answers are pretty clear. You add features, you have different tiers, you change the pricing in some way with usage or seats or something that kind of goes up automatically as they get more value out of it. So I don't think that's terribly interesting to double click into. It's sort of obvious. I would say, oh look, it's tied into pricing because of their behavior. But the main thing is you want it so that the customer themselves would agree, when they pay more, that they are getting more value. Hopefully, they even think they're getting far more value than the price going up, right? I'm going to say this as if it's precise, which it's not. But they need to feel like, if the price doubles, yeah, but I'm getting five times the value, so that's fine. That should be the feeling, whether they can measure it or not.
A good way to do that is to say, well, then you should be measuring whether they're getting value out of it. Often we measure usage metrics and other kinds of metrics within our product because we can. But actually what's really important is to measure how does the customer value this, and we need to measure that so that we make that go up, because if we make that go up, they'll be willing to pay in whatever structure. And if that isn't going up, they won't be willing to pay. So even if we start making them, they'll leave. And we've all probably done it ourselves. We've all had products we love, but then as we scale, the price goes up faster than we feel the value is, and then we start looking for other products. We've all experienced that.
So that's what I'm saying. To do that, you want some sort of measure of the value the customer is getting. If you're really lucky, that can be a number. That would be wonderful. Then go do that, and maybe that's your north star. But admittedly, it's not always possible. So then the question is the usual questions of metrics. Are there proxy metrics that we understand? They're not the full picture, but they're helpful. And I'm a big believer in saying not all important things are numbers. I mean, even things like how differentiated are we in the market, not a number, but it's very important. So this might be one of those things that's important but not a number. So okay, can we get some proxy metrics, even of behavior, other things, something better than some metric that's just operational? And even if it's qualitative, okay, can we do that? Can we talk to customers and ask them qualitative questions? I would just say, do your best here, because only when you generate more value for the customer can you then decide how to split that with the customer in terms of things like price, right?
That's in fact how I think of it, that very phrase: how do we create more value for the customer and then split that with them? And when you do that, you're keeping the customer forefront in mind. Splitting means you get some. Let's not forget it's not a charity. And on the other hand, first we should think how are we generating value for customers, and then we've now earned the ability to take a little piece of that. So to me this is the right way to think about NRR, not just we'll add a feature and make them pay. True, but let's
Actually take it from this different, let's get there from this different perspective.
Amazing. In the recent Madhavan episode where we go into pricing, he actually has some really good tactical advice for measuring the value that you're giving to a company, to quantify that, which feeds into this idea of how do I create more value for you and then how do we split it?
Yeah.
The other element of this that's top of mind is just this land and expand strategy. There's a lot of companies that are just like, okay cool, we'll get in with some price, we'll expand, that'll be amazing, which is essentially, expanding is NRR going above 100%. Yes. Something Jen actually shared in her chat that was really important is that you can't expand that much, at least for a while, because if you get in for like 10k, if you go up to, okay, now it's 100k, someone's going to be like, what is this? Why is this, it can't go up 10x, are we getting 10x value from this? You can't just raise prices later. You're kind of stuck at that reference point, so you have to be really careful there.
Yeah, I think that's right. And maybe you don't deserve it. In other words, especially when there's investors or other sort of forces saying like, "Hey, we need X, Y, and Z." The forces that are not the customer saying that. Yeah, you can be coerced into making pricing or other kinds of policies that in fact are not good for the customer.
So one tool I use is when anyone claims anything really: is that really true, or is that really actually good for the customer? Because look, we are going to do things that are selfishly good for us. We have to. We can't just do things that are bad for us. But is this in fact good for the customer? Because often, even in an internal proposal, we say it as if it is. Oh, this pricing will be good. It'll raise prices on everyone, but it's better because of this reason that we're sort of justifying, right? It's like, well, will the customer say this is better? If the answer is no, it's like, okay, it's better for us, but sorry, we have an equation where it has to be better for us and better for the customer. Sorry, it's an AND, you know. And of course, not all companies do that. And we experience that, all of us, as consumers on the other end of that, and we don't like it. And that's not a good long-term strategy, even though it might work in the short term, as many bad long-term strategies are.
I love just how this third step just reveals how powerful the sequence is that we're going through. Step one is this logo retention, essentially, do we have product-market fit? Step two is pricing, positioning, essentially, are we going after the right market and charging them the right amount, roughly? And then here it's just, can we grow? Is there something here that can continue to expand? Because you're going to get eaten alive if you're... especially, and this is just to be clear, this is B2B SaaS primarily that we're talking about here. It's harder to grow NRR if you're a consumer product that has, I don't know, just like a tier two.
What I would say is the rules are true everywhere because they're based in the mechanics of finance in the business.
Now you are right that in the consumer segment, or small business segment for that matter, they tend not to grow. So that doesn't mean the NRR question is invalid. It means, dang, we can't think of anything.
That's okay. Then you go on to the next question because you can't think of anything. But you know, it would be more strategic if we could.
Mhm. Yeah.
And are we trying hard enough? As a consumer, I do not want to spend more with AT&T, but they're also not giving me any more value.
But there are other products as a consumer, like Amazon, where they do. So you're right, but as a product manager, it would be 10 times more valuable for you to think of something like that, you know, than to move on to other things, etc. Or there's other ways, like other products. We didn't talk about it, and it's okay because of course each one of these, you could go on forever, right? But another way is a second product. A second product sold to the same segments that you're in, so that your existing customers can buy it. Well, I don't know why that wouldn't work in consumer. It certainly works in consumer and apparel.
I think about AG1, which has all these new, like I'm doing their sleep supplement, and there it goes. They're just like, here's a new thing you can buy.
Yeah.
So is it harder? Yeah. Of course. Of course. All of these are easier or harder in different segments and ways. Of course. I'm not trying to say otherwise.
But I would say the mechanics of how the finances work is the same. You're just saying I don't have this lever to pull. And then I would say, okay, then you need different levers. One that we didn't talk about is, one way to offset the cancellations is existing customers grow. But another way is if existing customers bring in new customers. So they didn't grow, but they brought their friend. Now this is absolutely something that happens in consumer. But it's also an answer to this thing where cancellations grow exponentially, you know, because existing customers bringing in new does triple if you have more existing customers. Aha. So this is stuff like refer a friend and all this kind of stuff. Again, some of this is obvious. We don't need to enumerate that. So those things are good.
And so in consumer you might say, oh, it's easier to try to get someone to invite a friend with a coupon and blah blah blah than it is to try to get them to grow. But in B2B that may not be true, right? I don't get a mid-size company to refer. That doesn't make sense. So once again, this question of how do we have the existing base help us grow is still correct in consumer, but its manifestation could be very different. Of course, I agree with that. But I mean, how could we not say, of course things like word of mouth and inviting a friend, of course that's enormous with consumer, and this is one of the reasons why.
I just love picturing the people listening to this, especially product folks, founders. I imagine many of them are just sitting here taking all these notes of how to help grow their product, because we've positioned this conversation as how to deal with stalled growth, but it's actually just as useful for how do I grow more.
Oh for sure, right, right. It's more growth. It's just maybe a little more evocative.
It's because if growth is good, yeah, sure you want to grow more, but it's not the problem.
If growth is really good, the problem is generally operationally scaling to meet the growth, and so you're focused on that. It's when growth slows, you're like, whoa, wait, wait, wait, right? We have to focus on growth now. Must, as opposed to, of course it's always nice. So yeah.
Yeah. And I was thinking, as we were talking about consumer NRR, if you look at Duolingo, they've done a great job here. There's so many ways you can pay them more for all these little
advances. Get all these gems. Change the color of your app to something fancy.
Yeah. Yeah.
Yeah. Okay. So, there's more. Let's keep going. So, we've done three. There's more you can do.
So, okay. Logo churn.
Pricing.
Pricing.
NRR. And then maybe it's stalled. So, this is really a stalled question. Maybe your acquisition channels, your marketing channels are saturated. We're done. I mean, what we tend to do is flog the people doing AdWords, get more; flog the SEO people, get more searches, right? It's possible that this is it. Maybe this is it in some sort of physical law. There literally isn't anything else. Or maybe this is it just, this is how good we can be. This is it, right? But there really are limits.
You know, there's different words for it. Inventory is sort of the old word, like with magazine ads, inventory was the word. But there's just this amount. There is only so many searches in your area, and you can only appear once in the search results for a given keyword. So there is this limit of what you can get. Even if you were number one position for everything, you know, other things that are not even practical, there's still a limit, and there's some kind of practical limit that's below there that we don't really know. But, you know, maybe we're there or maybe we're close.
And worse, channels tend to decline over time. So, I think people talk about S-curves, right? Oh, I didn't figure out this market, then I figured it out. We unlocked it. Now we're getting, you know, 100 leads per month through Facebook or whatever. But then it kind of taps out, and then we go into this optimization mode. Can we eke out another blah blah blah, which is right, that's right. And you call that an S-curve because it's shaped like that. But that's not what happens. What happens is it starts with an S-curve and then it starts sagging. Its butt starts sagging down. So, I wrote an article about this called the elephant curve, which is what I named it, right? Because it's like this trunk, but then it's this butt.
And there's different reasons why this happens, but if you talk to any marketer, they'll all tell you, "Oh my god, let me tell you this story," right? They all have stories about it because yeah, this is what happens. There's different reasons. First of all, the audience gets saturated, you know, because there's all these little marketing-isms, and I don't know if they're true or not. I don't think any of it has any data that actually proves it, but whatever. Like, you know, if words start with the same letter, that's better. I don't think anyone's ever proved that's true, but marketers seem to think so. Okay. Well, one of the things is someone has to see it seven times before they act. Okay. All right. Well, maybe they saw it seven times already, so they don't want it, you know, or they saw it 20 times, right? So initially you hit people who hadn't seen it yet, right? But now you have.
And you know, especially with magazine ads, as I used to do before there was no such thing, that's exactly what would happen. You'd have this nice surge, and then it's like, well, we've seen this, we've seen it before. So you still get a little trickle. You still get some because, you know, it was just that moment they needed to see it again. Okay, but a lot of people have already seen it and they don't want it. Or the channel is declining and they'll never tell you.
When I did magazine ads, every year they would tell you what their circulation is, and every year it would go up, and then the magazine would go out of business. Conferences are the same way. Attendance is great. Attendance is great. Oh, wait. We're out of business because we couldn't get enough people to come. What? They said it was great. But, you know, more quietly, AdWords, Facebook ads, even SEO searches, this does happen all over the place. Affiliates, it can happen.
And now with AI, I mean, I don't even know, right? It's disrupting everything. We've all heard stories going all different directions. I think the answer is I don't know. Shrug is the answer. And what will it be like in two years? Another shrug. But the point being, they're not all going to be growing a lot. That's not one of the futures that AI will bring. So it has this sag. That's just a very long way of maybe trying to prove this point. But it's yet another reason why you can't just rely on marketing forever, because not only do you try to stack things, but there's not an infinite number of marketing channels you could advertise in that your customers are actually going to, and they sag. Oh no. It's even harder to keep up. So it's kind of reinforcing my very first point here with this.
But here, are they all saturated? Because we could flog marketing all we want. It's not going to work. So maybe growth is slowing because all our channels are saturated, possibly even sagging, but even if not, okay, not growing. And so, we can't just flog the marketing department. It's going to take something else, right?
Then the obvious thing is get more channels, but again, maybe there aren't any. So again, there's many possible things to do, but this is this critical thing to notice, because I guess I would put it this way. Do you know right now which channels are saturated and which aren't? If the answer is no, I'm like, well, okay, maybe that's because the answer is all. You know, that needs to change how you think. Just adding one little feature and then hoping we can flog AdWords is not going to work. Even if the feature is great, not going to work. So there's different things that could work, but that's not one of them. And yet, that's probably what we're doing. "Let's add another feature and marketing can flog it" is often the answer. But if you're in this state, that isn't the answer. So this is why, you could say it's obvious to say this or that, but are you acting like this is true? You know, often we don't.
Okay, so there are many things to do. Again, we don't have to enumerate all of them or something. It's just simply the right question to ask. But for example, some people are like, we've done direct, maybe we should try things like SEO and social and these other indirects, or vice versa, we're really good at SEO but we've never taken out ads. Often if you've done ads and they're optimized, you know what content might be good to write stuff for SEO, and maybe even vice versa. So it's a good idea, and sometimes it works. But here I have no data. I only have my feeling here, and actually you probably have a lot more visibility into this, but my experience is a product that sells really well direct actually doesn't do well in things like social and SEO, and vice versa. If you're getting a lot of traffic through SEO, adding ads often costs a lot and doesn't really move the needle. You can tell me, because I don't have data to support that theory.
Yeah, my take is you can always get some percentage of wins from all these different channels. Usually one channel is where most of your growth will come from, and so over time everyone just adds every channel. Everyone's doing ads. Everyone's doing SEO in some form. But it's usually sales or word of mouth or ads that drives everything. Everything else is kind of a little layer on top.
Yeah. So, you know, should you do that? Yeah, probably, especially if you're at some scale and you can just afford to, because it's such a clear thing to do. But probably you'll have to get more creative about what it means to add a channel, or something like a new product or a new market where it's actually new. It's expanding in a new way rather than trying to incrementally expand what you're already doing.
So an example of getting creative on a channel is what Constant Contact did when they had this very problem of growth is slow. We don't know how, we sell email marketing newsletters to small business before all these modern tools existed. And one of the things they did that restarted growth is they physically went to a bunch of cities and held workshops showing here's how to do email marketing for your small business. So the restaurateur and the dentist and everyone would come to these sessions, and they'd teach them how, of course teaching them how with Constant Contact. So they became customers, right? You would think there's no way this is cost-effective, physically being in these cities and dragging people in for a, you know, $20 a month product. Like, no way. It was very effective and actually solved in that moment their restarted
growth. They were very clever about... first of all it's a clever idea, but then they were clever about how to do it. They took like power users who were also agencies. So like these could become customers of their... okay, all like you could be clever about like how is it that we do something different, something new. So that's possible. Of course, it's always hard to say, "Think of something clever." That's a weird finger-wagging thing to do, but okay, it's true. But yeah,
it could be a different type of channel. For example, HubSpot famously tested selling through agencies instead of direct. It ended up being 50% of the revenue after four or five years. So, it's one of the main reasons why they're able to continue growing. Same thing happens with my company, WP Engine. Tons of our websites are sold through agencies that create WordPress sites. So there could be something that's not direct anymore, that's, you know, another channel of human beings or something like that, could in fact dramatically change your growth rate. There's lots of examples like those.
So, but it could be like it's time for the next product, like, and I said that earlier because it's always possible. Of course we all know that's very hard. It's risky. I have sort of a framework that I use to think about that kind of expansion, which I'm happy to provide. I've also written it up, but I'm happy to say it right here. But usually you want to stay in the target market you're already good at and grow from there. But sometimes the whole point of the expansion is to change the... or change target market, or add something where you're leveraging something else about the company that you have as an asset, going somewhere else. So this is what this framework helps decide. But one way or another you probably want to plant one foot into some strength or asset that you have, move the other foot, which is the risky part, but the idea is that, yeah, well, we have this big upside, so we're taking that bet, and like that becomes a smart bet.
So with things, of course that's true for any of these things, but especially if acquisition channels are full and it's like we literally can't ask the marketing department, like that's just not one of the choices. It almost forces us to start taking these more drastic bets to say, "Well, we got to do something and that's not one of them."
I just want to keep saying how awesome this advice is and how many people are going to benefit from... like, you know, none of this is like, "Oh, I've never ever thought of any of this." It's just like the very methodical sequence of questions you should be asking yourself to help you not just undo stalled growth, but also just come up with a bunch of great growth ideas. And this specific section, it makes sense. Somebody's discovered alpha in a growth channel. Say, SAPI just launched. TikTok just, you know, there's like, oh, cool. What's the new thing? Okay, let's get there quick. And then you drive a bunch of growth. It's awesome. Eventually, everyone's going to start doing that. And so, you should assume everything that is working for you now will slow down. I've never... I've missed your post on the elephant S-curve. What do you call it? The elephant curve.
Yeah, elephant curve.
It's so... Yeah, that's so real. It's not just this S-curve that will forever continue to drive win. It will actually dip and decline over time because other people discover it and start using it.
Yeah.
I love that. So, the idea here is, and the classic advice is, like, not whether it's an S-curve or an elephant curve, think about are you starting to approach the apex of that, and start to explore other channels before you slow down or start to dip.
Yeah, it's easy to hear stuff like, "Oh, if marketing is full, do something else." And you go, "I know." But then you look at people's behavior and it's like, "Well, you're not acting like you know," [laughter] you know, so maybe what needs to be said in enough detail that you actually do something about it.
And this is... it's important to note this is a very hard problem. Most companies do not really solve this. They... something worked for them.
Sure.
And then it stops working, and then like, "All right, well, we found something and then it just kind of went away." There's a couple posts we're going to link to in the show notes that will help you come up with ideas that are all around new growth channels that are emerging. One is by Emily Kramer around ecosystems as a new growth channel. And there's a lot of really cool advice there around this kind of emerging combination of influencers and content and partners where it's your ecosystem that helps grow. Basically, there's a quote from the head of growth at Wiz where it's like, "Why start with zero when you can start with 10,000?" Essentially growing through someone with an audience already. And then there's going to be a post out by the time this comes out around ChatGPT's app store, which is going to let you submit apps. And that's a really interesting, potentially huge new growth channel for companies. So, cool stuff happening there.
So just to summarize: logo retention, pricing, NRR, marketing, channel saturation. What comes next?
The last question is, do you need to grow?
So okay, growth is stalled. And if we assume every question before has been answered in a satisfactory way, you could ask, hey, [clears throat] is that a problem? What do we mean by grow? What do we need to do exactly? Now of course you should know these kind of things with goals all the time, but... and again, like, obviously the answer could be, once again, oh, new products, these other things where, like, this company... when we say do you need to grow, if we define you as this product in this market in this company, the answer might be no. What we need to do is have a different product, or in a different market, or a different thing. Or you could change the word revenue. If you say do you need to grow revenue, you could change the word revenue and say, you know, but what we could do is maximize profit instead of revenue now. We've been maximizing revenue, but maybe we maximize profit instead.
And so this is a company like 37signals, or really lots of bootstrap companies who have hit some sort of limit and realize that's okay. Like the founders are getting paid millions of dollars a year in dividends and like it's okay. [laughter] I don't have to get... in fact, if I got bigger it might be an organization that I don't like, or serving a market segment that I don't want to serve, or whatever. And so maybe growing forever isn't the goal actually, or growing revenue isn't.
You could ask philosophically, why grow anything? Why isn't it just okay to have stasis? And we all have heard the phrase, if you're not growing, you're dying, right? This is a classic company thing. Is that true, or is that the kind of thing that, like, investors use to make founders grow, or try to grow even when they shouldn't?
It might be, but I would submit that even at a bootstrap company that has other values and culture other than growth at all costs, that that phrase is still fairly relevant, because if the company's stagnant for years, is that a great environment for everyone? As the founder, did you start this company in order to do the same thing every day? Is that why you did it? Do you really want to do... is that fulfilling for you? What about everyone else? Nobody wants to further their career? They just want to do the same thing every day and never further their career, not really learn anything, not really innovate? Does it feel good to just not be growing? The answer could be yes. You know, if I'm a CPA and I have some clients and life is good, the answer could be yes. Like, you know, I'm not saying... I'm not dictating the answer here, right?
I'm just asking because a lot of times, whether it's our careers or, as founders, our companies, a lot of times we've just been in the mode of I've got to grow, I've got to get promoted, I've got to do more, I've got my resume. We've got in that mode for so long, like maybe our whole life, that I was going to say lose sight of, but maybe we never had sight of: wait, does this make me happy? Is this what I really want? Am I fulfilled doing this? Or even if I do have these goals, have I gotten stuck in a rut where my goal is growth and, you know, I don't know, more money, more everything, but I'm stuck in a rut here? Sometimes we forget to take a step back and go, wait a minute, what is this? Is this still right, or do I need to turn the page and have a new chapter of life right now, you know?
And so this question, do you need to grow, or if you're not growing, you're dying. Well, for some people, no, they like doing the same thing forever. And that's great actually. That's nice. But for many people, especially the kind of people who want to get into product and build stuff and innovate, and people who start companies, a lot of people like that are not the kind of people that want to just, you know, kind of do rote things for 20 years.
And so the not growing part, what I like to say is maybe the "you" in "if you are not growing, you're dying" is you the person as opposed to you the company. It's also you the company, right? But like, what if we took it to mean you? If you are not growing, then in some sense, maybe for some people, you're dying. Maybe if you're listening to this, that's you. You're a shark and you got to go. And we all know people too who claim they hate work, or maybe they do hate work, let's not say claim, they do hate work, but then they retire and kind of go downhill because they don't have a purpose, or this or that and the other thing. In that case, it was true. If they're not growing, they're dying, literally.
So again, I don't mean to overstate this, and I certainly don't mean to claim that there's some answer that's right for everybody, of course, but surely this is the right kind of question, and surely for many people who are listening to this, the answer is, yeah, I mean, in some sense, some very rough sense, that's probably right for me. And so if I'm in a stagnant situation and really every other option has been exhausted and isn't going to happen, this is simply a stagnant thing, maybe something else needs to happen. I need to leave. The company needs to change in some drastic way. I sell the company. I change jobs. I don't know. Like, of course, it's going to be super context-specific and personal, right? But like, something dramatic may need to change because nothing incrementally is changing.
So this final question: do you need to grow, or if you're not growing, you're dying, is that true? And are you therefore dying? And what needs to happen? You know, so if you were looking for more metrics in another framework, sorry, that's existential, but it is existential.
So, do you have to only ask this at the end of the chain? No. Of course, you should feel fulfilled. And, you know, of course, you want to be checking in with yourself at least annually, of course. But I sort of put it at the end of my list in the sense that I'm assuming the original question is about the company, [laughter] you know. But especially with smaller companies, but also with big public companies. There's plenty of big public companies that aren't growing, aren't there? [laughter] So like this is true of all scales, because there are natural sizes for things. So yeah, it's a little philosophical, but I think it's quite important.
Such a beautiful way to wrap up this piece. A lot of people listening to the podcast are bootstrap founders, and for them this is actually very much an option. They can just be happy with the revenue they're generating. Like with my newsletter right now, I'd be very sad if it stopped growing, but also just, it's amazing the life it has created for me. And even if it did stop growing and just stay flat and doesn't become an elephant curve, that'd be incredible. In practice, psychologically, still hard for that to be the case. And that's why this component of the sequence is really important. Like, why do you actually need it to grow? Is that just your ego? Is that just like I'm used to growth?
It can also help you avoid doing unnatural things that you actually regret to grow. So like if growth at all costs is just the thing, like there's probably ways you could quote-unquote grow the newsletter that you would just say, I just wouldn't be proud of that, and the newsletter is doing so well, I don't need to do that. And so again, maybe that's a softer version because growth hasn't actually stopped. But okay, it's a softer version of... you might say, I certainly don't agree with growth at all costs. I want to grow as much as possible within the things that I'm proud of. Like if we grew fast but the content was crappy, I'm just not willing to do that. It's like not the point, you know? And so it helps set up these boundaries of like, wait a minute, not if dot dot dot. And, you know, early on we may not have that flexibility. You could argue that you should have those values early on because that's who you are and that's what you're doing and people respond. So I could argue you should have that all along.
But I could also argue that at the beginning you're just trying to do something where you don't die. You're starting to blog. You're probably copying other people's style. You probably don't have that many unique things to say. So there's a lot of reg. That's okay. You're just trying to get going. It's okay. Now, 20 years later, if you have no style and no voice of your own and nothing new to say, that's probably not good. But to get going, sure. So sometimes we have this thing where we get going with maybe looser... I don't want to say values, I'm not saying it's unethical, but like a looser sort of bar, or a pride that we have in our own work, and we tighten it up as we're sort of able, as we can afford to, you might even say. So good, but then that becomes a nice filter here of like, what is it in a greater sense I'm trying to do here, I'm willing to do here? So if you're not growing, you're dying, fair, but like that has to come with these limits, and the more successful you are, the more you can be serious about those limits.
I think an important element of this is also the product you're currently working on. Maybe it's okay for it just to not grow. There's a good opportunity to do something else, have this thing maybe running on the side, maybe sunset it at some point, but it's a good opportunity to be like, okay, wait, what else is out there? We had a recent podcast conversation with Matt MacInnis, CPO at Rippling, and there's some really good advice he shared on just when to quit, when to quit your startup. Just like, you know, if it's like four or five years in and it's just not clicking, maybe it's time to move on, and even though people do succeed years in, most likely it's not going to be you.
I have a book almost out now that's on pre-order about topics like what we've been talking about. The next book I want to write is on this topic of how do I make these decisions of uncertainty. Like, maybe it is time to quit, maybe I should move to a different city, maybe I should marry this person, maybe I should launch this company, maybe I should use this strategy, where you want to use probability and expected value. It's unlikely that dot dot dot, right? But the truth is we don't know what the probability is. We don't know what the probability curves look like. We actually can't use expected value. And anyway, even if you could have expected value, I am a human being. This is my life, and I either sell the company or I don't. And so all this stuff about probability, like, that doesn't apply to me. I need other ways of sorting this out.
So I guess I would just say briefly, probability is not going to work for these decisions. So that doesn't say what is right.
But it's not that. And so, which is nice because you can put those tools down. I'll do some market research to see if I should sell my company. Nope. That's not where the answers are.
More questions than answers on that one.
Yeah.
Speaking of the book, let's give you a chance to share what you're working on and when this is coming out and where folks can find it.
Sure. So, the book is called Hidden Multipliers, and you can pre-order it at hiddenmultipliers.com, or I guess if this is out long enough, you order it, I guess, depending on when you're listening to this. And it's a lot of stuff kind of like we were talking about today. These questions of... it's called multipliers because the idea is little things that you can do or little decisions you can make that have a huge impact. And like moving the cancellation rate from 5 to 4% sounds small, has a huge thing. Onboarding as opposed to later use, huge thing. So those are some examples, but the book is of course full of different kinds of topics, but all of this idea of this stuff that has such a big impact on things like revenue and profit. And just as you said earlier, either maybe you have never thought of it that way, so you weren't thinking about it right, or, yeah, you've heard that, you say, "I know," but your actions don't reflect it. And so if we go deep enough with examples and specific things to do, then you can actually act on that supposed knowledge and realize those multipliers.
And just to remind people, the URL, hiddenmultipliers.com.
Yeah.
And there's an S at the end. Multipliers.
Right? There's more than one.
Many. There's more than one.
Yeah. Yeah.
Jason, I had other things I wanted to talk about, but I feel like this episode's actually going to be stronger if we just focus on the thing that we've been talking about, which is unstalling growth. So, if we do that, is there anything else you want to mention or leave listeners with before we get to a couple corners and then the lightning round?
I think if you tried to find a common thread throughout all this stuff about growth, it comes back to the customer getting value. And I know we already talked about that, but I think if there could be one thing where it would help solve kind of all of it, it would be that they really are getting value. Your product actually promises the right thing, and then it actually delivers on that thing, and the customers can onboard so that they can do the thing, and the customers know, they realize that they're getting the thing, and you're measuring the thing so you know it's increasing.
That is probably, if I was an LLM, I'd probably say that's the common thread, or, you know, there's many ways that manifests, of course, but if that's your north star, is how are we actually creating value in the way the customer values it, and their language and their way of understanding it. I wouldn't say all the pieces magically fit into place, but certainly isn't that sort of the root thing that is going to make all the stuff work. Then there will be a good way to do pricing, and they will stay as long as possible, and, you know, these things will probably be right. So this idea of creating value for the customer and then figuring how to split with them is probably the root idea. And of course I hesitate because platitudes like that are actually not very actionable, like, "All right, well, I'll move on with my day," and that's why Twitter's not so useful. But given that we've gone into so much detail, perhaps that's a nice way of summarizing it.
I think that's such an important point. I think what's also interesting is some of your advice is the value may... you may be picking the wrong customer, the wrong market. You may be positioning it wrong. So the value may be there. You're just trying to convince the wrong people about it.
Yeah, there's so many ways to get it wrong, right? Because, like we said, all these things have to be right. You just said another one, which is you have to say it in a way that when this person hits the homepage, they know it. It's true. But do they know that?
Just so many things have to go right.
What a tough job we've got over here, just solving people's problems. Come on.
Well, thanks for growing.
So now we will be after this conversation.
Okay. So I'm going to take us to a recurring corner, a recurring segment on the podcast that I call AI corner. What's one way that you have discovered using AI in your work or in your life that might be helpful for folks to hear?
There's a lot of data on the internet, and it's often in things like images, which makes it hard to do your own analysis or plug it in, or come up with your own models or apply it. But I found that AI is really good, actually, especially Gemini, at... say you just give a chart to it and say, like, make this into a table that I can paste, like literally say that, that I can paste into Google Sheets, and it will do it in a way that literally you can copy, and it will actually paste correctly into Google Sheets, and then you can do stuff. So especially with the book and my articles, I love to use real data whenever I can, of course, so I do that all the time. So I think that kind of interpretation is very useful. And so all of a sudden you can get 10 examples of something and test a theory, where before it was kind of just too hard and you didn't.
That's an awesome tip, because people know you can generate all these infographics, especially with Gemini and Nano Banana and all these things. That's really cool to know. You can just feed it, here's a chart, make it text.
Yeah.
Okay. I'm going to now take us to contrarian corner. The question here is, what's something that you believe that most other people don't?
A/B testing doesn't work very well, and it doesn't work on most things. It won't work on strategy or vision or insights, like nothing actually important to the success of the company. You don't A/B test whether Uber is a good idea. And then even when you do A/B test the details, where I agree sometimes that can work, what happens is people will try things like, oh, I'll just try this verb and that verb and this ad and that ad, and then, oh, the seventh or eighth one I got a positive result, that must be good. And what happens is you keep doing that, you pick the best one, and then you go on and you find another one and pick the best one, and then a year later you look back and you should be like 50 or 100% better because you've stacked these things, and you look back and nothing's different. The conversion rates are the same as they've always been. You say, what the hell happened? I thought I picked the winner.
And the answer is in a combination of the tools not being statistically accurate, which they're not, and the fact that you will get false positives even if the tool is statistically accurate, means that most of them are false positives. Even if the tool is 95% accurate, when the thing you're looking for is rare, which it is in the case of A/B testing, the false positives happen more often than the actual thing happens. And so most of the results you get are false positives anyway. So, as a result, this isn't true of all A/B testing, but for what most people do when they just do the sort of mundane A/B testing, you can't A/B test the important things, and the details are mostly false positives. So it's an enormous waste of time unless you're incredibly sophisticated. I know there's special groups that actually are very sophisticated. Fine. If you're not doing that, it's sort of like the poker table. If you don't know who the patsy is, it's you, right? If you don't have all of this information and knowledge about A/B testing, then you're the patsy.
Bam. All right. I will just say that I have found A/B testing useful in my career. I think it's maybe at a certain scale when you're just kind of trying to optimize and continue to grow, you know, where it's like millions of users, like a percentage gain is like millions of dollars.
True.
Most people are not working at that scale. Most people, right?
Yeah. So, just wanted to, for folks that find it valuable. That's true.
But I love it.
However, even so, on your own podcast, when you were interviewing the guy from Shopify, and he was saying how maybe a third of the things that they found with their systems just disappear. They just magically disappear.
And they have a team of 100 people and they're really good at it.
And their effects disappear all the time, so they double check later whether the immediate effect goes away, because even then.
Right. I think that was the CTO of Shopify conversation.
Okay. Yeah.
Yeah. Sweet. Okay. Worth of memory, cuz I listen to a lot of episodes, but I don't...
Such a good one. I love that. Yeah. Where they leave like a holdout group essentially, and then they just look back: was the effect something that lasted? And most times it didn't.
Yeah.
There you go. And that's with a lot of... So that's what I mean. If you're doing that level of stuff,
good for you. But if you're not, I don't know, man.
There we go. Well, Jason, it's always a really good sign when I'm just like, I can't wait to get this conversation out the door and into people's minds, because there's so much value here. I'm already anticipating all the people that are going to reply and just be like, I had so many ideas for what to do with my product, which is exactly the goal. And with that, we have reached our very exciting lightning round. I've got five questions for you. Are you ready?
Boy. Yeah. I mean, I don't like talking a long time anyway. So lightning is great.
Here we go. What are two or three books that you find yourself recommending most to other people?
For writing, On Writing Well by William Zinsser. I know I'm not the only one, but that's kind of the point. On my best day, I write like that. And then for product, I actually like Crossing the Chasm, which of course everyone's heard of, but what I find is no one's read it. So you know the little picture and you think you know what the chasm is. What I find is very quickly I realize, oh, you haven't read the book, you saw a blog post. And there's so much good stuff in there: how to define a market, and really what to do with this model. It's fantastic. So I highly recommend reading the book.
I've had Geoffrey Moore on the podcast. We dove into a lot of this stuff. One of the things that always stuck with me is when early companies are looking for someone like them to adopt the thing. That's something that really stuck with me. It's not like they're looking for an early adopter to be like, "Oh, this is awesome." They're looking for someone that feels like them to say, "This is great." And so the early adopters are just going to spread to other early adopters, and there's work to do.
Yeah. Part of that's because he defines a market, among other things, as: the people in the market respect the opinions of the other people in the market.
Exactly.
And that's when you realize, oh, so jumping to a different market, it's not impossible. It's just, like, case studies aren't going to work.
Yeah. And then you think, okay. And then On Writing Well, such a huge fan of the book. That's like the book that most helped me write. And if you summarize the book for me, it's just cut. Cut more and more of your stuff. There's always more to cut. I love this phrase where he's on a panel with this guy who is like an amateur writer, and the guy told him, "I never knew writing could be hard," and Zinsser says, "I never knew writing could be easy." I think both of those kind of summarize the turmoil of being a writer.
Yeah. The classic, maybe Hemingway, maybe not, quote: writing is easy, I just sit at the typewriter and bleed.
And bleed. Yeah.
So good. Okay, moving on. Favorite recent movie or TV show?
ER, from 1994. 15 seasons. Why do I say that? Besides the fact I think it's good, I have a 16-year-old daughter, and we're now on season, I think, 13, watching this whole thing. She says it holds up after 25 years, and being, you know, Gen Alpha or whatever, I don't even know what it is. And so if this is an era when, you know, shows were an hour long and seasons were forever, and she says it's great TV, it must be great TV.
Wow. I've not had that on this podcast yet. Also The Pit. I don't know, if you enjoy ER, you'll enjoy The Pit, which just won all these awards.
Yep. If it's good.
On Netflix. Fun fact, my cousin was in ER, not as a recurring character, but she was like a young girl patient, and now she's a fancy actress in the world.
Oh, cool.
That was her start.
Okay, next question. Favorite product you've recently discovered that you really love?
This is probably not unique, but Wispr Flow for dictation. It's really good. I like the keyboard shortcuts because, you know, I just use it all the time in all the software. And anything Anker makes. You know, they have like power stations and docks and recharging.
Over the K.
Yeah, A-N-K, right? And just all their stuff is super high quality and works really well. Everything seems to charge twice as fast when plugged into an Anker thing. So, I don't know, whatever it is, it's really good.
I got a new Anker charger. I also love Anker. It has like a display on the side when you plug in stuff, and it's got multiple ports, and it shows you the percentage it's charging and the wattage per outlet. I love it. They're just like, "How do we make this more fancy and fun and charge more?"
Yeah.
I love it. Yeah. And then Wispr Flow, quick shout out. You get a year free of Wispr Flow by becoming an insider, I think even just an annual subscriber of my newsletter, as part of the product pass. And so check it out, lennysproductpass.com. You can also check out...
I did not know that. I'm not a shill for them.
No, I love that. I love when people recommend products in the product pass. A whole...
Because I've been a subscriber for long enough that I didn't get that.
You missed out. You missed out. There's 19 products in there right now. By the time this comes out, there will be even more.
Okay, two more questions. Do you have a favorite life motto that you find yourself coming back to in work or in life?
Yes. Be yourself; everyone else is taken. And it's attributed to Oscar Wilde, but I've tried to look into that as I tried to get all my annotations correct for the book, and there's no evidence that he said it. But there's also no evidence who said it. So, let's say it's Oscar Wilde, because he said lots of things like that.
I love that, and it's such a deep point. You know, it's easy to hear and be like, yeah, yeah, yeah, but it's something I've learned to be more and more true every time, especially as you see people online doing their thing and you're just like, oh, I want to be like that, and then you realize, no, you've got to be yourself.
No, and the people who love you or like what you do also want you to be yourself, because that's what they love, and if you changed, then they wouldn't love that. So...
Final question. You have this fancy award behind you on your desk. I'm curious, what's the story there?
That's the Ernst & Young Entrepreneur of the Year award for 2017 for Central Texas, which I co-won with the CEO of WP Engine, Heather Brunner. Which is awesome, because I often call Heather our late-joining co-founder, because that's what Reid Hoffman called Jeff Weiner at LinkedIn, because, you know, Jeff was like four years in but was so impactful
to everything, the success of the company, the culture, that, like, basically is a co-founder. And that's exactly what Heather is like at WP Engine. It's now been 11 years since she became the CEO. So this is a, you know, there's lots of data to back this up.
And I used to say to people at WP Engine, like, if I just told you that Heather was a co-founder, you'd say, "Yeah, no kidding." I'm like, "Right, that's why I think of it that way, because so do you. So did anyone, because that's the impact she's had."
So we co-won that award, which is nice because you almost never have co-winners. In fact, I can't remember another one. I mean, I know there are others, but it's rare enough I can't think of another one. So it's really cool that we co-won that entrepreneur award.
Jason, this was so awesome. I really appreciate you making time. I really appreciate you sharing so much wisdom with us. Two final questions. Where can folks find you online? Point them to your book, your website, and how can listeners be useful to you?
Yeah, I mean, to be useful, order the book, [laughter] hiddenmultipliers.com. Or, of course, you don't have to. I have all these articles online for free. So you can go to asmartbear.com, and I'm on Twitter and other stuff that's all linked off of that website.
And the articles, they're free. I don't have ads. I don't sell courses. I don't sell anything. So that's very, very non-commercial. And so, in fact, the one thing I've ever done with writing that costs money is the book, because, you know, it's a physical book. I've got to charge something so I can ship it and everything.
But I think Hidden Multipliers is certainly my best work. So I'm very proud of that. But you don't have to buy it. It's okay.
This is our chance to repay you for all the free content you've put out over time, right?
And so I'm going to order a number of them.
Jason, thank you so much for being here.
Thank you. This is fun.
So fun. Bye, everyone.
Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review, as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show [music] at lennyspodcast.com. See you in the next episode.
Article published · Updated