Lenny's Written Position
The best AI evals combine automated metrics with human judgment — neither alone is sufficient for measuring AI product quality.
The 'current user retention rate' (CURR) metric — measuring the percentage of current users retained over time — is more actionable than traditional cohort analysis for mature products.
AI adoption should be tracked as both inputs (who is using AI) and outputs (business value created), and included in performance reviews to create accountability.
Analytical thinking interviews should follow a five-step linear flow: assumptions and game plan, product rationale, metric framework, goal-setting, and tradeoff evaluation.
Never use averages or ratios as North Star metrics because if your NSM increases while your ecosystem actually shrinks, you get a false positive — a metric that looks great even as the product dies.
The critical 'altitude shift' from company-level metrics to team-level goals is where most candidates stumble, because tracking a metric is fundamentally different from executing against a specific goal.
If you cannot describe a metric to a data scientist in a way that they could run a query with, it is not a useful metric — always define metrics so specifically that someone could implement them tomorrow.
Almost half (44.67%) of tech workers are currently experiencing significant burnout, and burnout strongly correlates with quitting intentions and low engagement.
Mid-career tech professionals (7-14 years experience) are struggling the most, showing higher burnout, higher quitting intentions, and lower job enjoyment than both early and late-career workers.
Opportunity areas should be scored along four dimensions: expected impact, certainty of impact, clarity of levers, and uniqueness of levers, with a simple sum and sort to identify the top three strategic pillars.
Roughly 4.5% of YC companies become unicorns compared to the 2.5% rate for other venture-backed seed-stage startups, and around 45% go on to raise a Series A versus the 33% industry average.
The top four YC companies (Airbnb, DoorDash, Coinbase, Instacart) account for more than 84% of all market value created, demonstrating extreme power law distribution in startup outcomes.
More than 50% of YC companies are still alive after 10 years, compared to the average startup lifetime of about five years, with only 13% having gone out of business.
99% of YC's financial returns have come from U.S.-founded companies despite accepting startups from many countries, with the most valuable European YC company (GoCardless) valued at only $2.1 billion.
Core 4 is a unified developer productivity framework with four dimensions — Speed, Effectiveness, Quality, and Impact — that unifies the principles behind DORA, SPACE, and DevEx.
Product velocity is about speed and direction; when teams focus only on moving fast without alignment on what to work on, trying to move faster can actually slow you down.
Improving the developer experience score (DXI) by one point saves 13 minutes per week per developer, equivalent to 10 hours annually per developer.
The three most common friction points slowing developer teams down are poor build and test processes, lack of time for deep work, and poor support for production debugging.
PR throughput per engineer is a controversial but useful metric when used as a system health metric alongside other metrics, as validated by companies like Meta, Microsoft, and Uber.
Core 4 balances qualitative and quantitative measurements because quantitative data shows what is happening while qualitative insights reveal why, which is necessary to actually change team behavior.
Even top-quartile consumer subscription apps only convert roughly 1 in 20 installs into a paid subscription and lose more than half of annual subscribers after the first year.
The best consumer subscription apps grow by building on a core value promise that drives a compounding Subscription Value Loop across three steps: Value Creation, Value Delivery, and Value Capture.
Pricing and packaging optimizations consistently generate 5% to 15% lift on subscription revenue, yet many startups set prices once and never revisit them for years.
Web conversion flows can boost subscriber LTV by 30% to 50% by avoiding app store fees and reducing involuntary churn from payment failures.
Twenty to forty percent of churn in consumer subscription businesses is involuntary churn caused by payment failures, expired cards, and processing errors rather than deliberate cancellation.
Consumer subscription apps should target a Day 1 Paywall View Rate at or above 80% to maximize conversion opportunities.
Direct monetization of AI features is generally the better long-term strategy because indirect monetization makes it very hard to track and accurately attribute value from retention and upsell.
The best AI features may reduce metrics like time spent in-app because generative AI finds the best product-market fit with products that increase productivity and give time back to people.
The closer your North Star metric is to revenue, the more suited your organization is to a GM model; companies with metrics removed from direct revenue (like ad-based businesses) work better with functional structures.
Only 20-30% of signups actually matter in most products; hyperfocusing on the segment driving revenue rather than growing overall signup volume can drive non-linear business growth.
As a rule of thumb, 50% of signups will not activate, and 40% of users who activate for the first time will drop within seven days; focusing on that 40% churn first yields the fastest results.
Product teams can frequently drive more growth by optimizing engagement with existing key features than by launching new ones, because most new features target only a subset of users.
When Google focused on raising awareness of frequently-changing search types (sports scores, weather, movie times) rather than building new features, they nearly doubled the number of users doing these searches and generated millions in incremental ad revenue.
A sense of urgency in launching experiments leads to compounding growth because improvements stack multiplicatively over time, especially when combined with word-of-mouth effects.
When prioritizing experiments, you should consider the total size of the user base affected (the whole pie), not just the percentage improvement; opt-out features affect far more users than opt-in features.
Pricing is the most under-leveraged growth lever: a 1% improvement in pricing can increase profits by up to 11%, yet 50% of software companies have never run a pricing study.
The Van Westendorp method for pricing research suffers from significant hypothetical bias where people state higher valuations than their actual willingness to pay, and should be used with caution unless combined with incentive-compatible elements.
For unfamiliar or high-ticket products, choice-based (discrete choice) pricing methods outperform direct methods because people struggle to evaluate pricing for products they have never used.
Every business can be distilled into a simple equation, and until you can express your business as one, you do not fully understand it because the equation reveals which levers have the most impact.
Customer acquisition payback period is a better measure than LTV/CAC because it tells you how quickly you can reinvest in driving more growth.
In consumer, if product-market fit is not immediate, it usually takes 6 to 18 months; in B2B, the median time is two years, extended further for products requiring network effects.
By picking even slightly better projects through improved prioritization, a PM can double their team's impact without increasing build speed.
DRICE (Detailed RICE) invests 30 minutes per idea to create bottom-up financial estimates, and teams at Dropbox that used it had twice the impact rate of teams using simpler prioritization.
If your experiment win rate is above 70%, you can delay investing in formal prioritization processes because you are still in the easy-wins phase.
Top B2B startups took on average about 2 years from founding to hit $1M ARR, and roughly 1.5 years after closing their first customer.
Linear uses no A/B tests and no metrics-based goals for individual projects; decisions are based on taste and opinions, validated through beta testing and conviction rather than data.
The median time from idea to feeling product-market fit in B2B is roughly 2 years, with 9-18 months of iteration after launching a working product.
90% of the time, acquisition is the first goal new growth teams should take on, because there is more volume for experimentation and the biggest drop-offs occur in the acquisition stage.
A good free-to-paid conversion rate for freemium self-serve B2B products is 3-5% and great is 6-8%; for freemium with sales-assist it is 5-7% good and 10-15% great; for free trial products it is 8-12% good and 15-25% great.
Reverse trials, which give users full premium access then downgrade to free, convert at twice the rate of classic freemium while maintaining similar sign-up rates.
Having sales reach out to free accounts is the single biggest driver of increasing free-to-paid conversion; nearly half of free-trial companies have sales contact more than half of sign-ups.
Shopify avoids formal OKRs because metrics-driven optimization can lead to local maxima where the product feels incohesive, and they will approve investments even without measurable metrics if the outcome improves craft and quality.
Over 40% of top consumer brands use at least two marketing measurement methods (MTA, MMM, CLS) together, and about 20% use all three to triangulate the truth of what is working.
Marketing mix modeling (MMM) has surged in popularity after iOS 14 because it only needs aggregate data and has no privacy concerns, with 82% of the top brands studied talking about MMM in case studies.
The recommended triangulation approach is to use MTA for daily ad optimizations, MMM for channel budget allocation forecasting, and conversion lift studies every few months to calibrate the MMM model with causal ground truth.
Snowflake has zero tolerance for organizational politics and embeds data scientists directly into each product team so that data-driven decisions are part of the DNA rather than an afterthought.
OKRs should not be used for performance management because failing to hit a cross-functional OKR doesn't mean everyone tied to that goal underperformed.
Having OKRs at every organizational level creates overlapping initiatives, excessive process time, and too many priorities, so reducing OKR layers and frequency improves clarity.
Always run a correlation analysis before running a linear regression, because if variables aren't correlated, the regression will produce inconclusive patterns.
Neither correlation nor linear regression proves causation, but correlation confirms the relationship between variables while regression shows how much one variable affects another and can predict behavior.
Product analytics tools like Amplitude Compass and Mixpanel Signal can quickly identify which features correlate most strongly with user retention.
For PLG products, the target marketing KPI should be product activation (activated signups), not website traffic or monthly signups.
PLG companies need to keep customer acquisition cost below $1 per unique visitor because the average PLG product generates only $1-2 in first-year revenue per visitor.
Duolingo runs over 200 A/B tests at any given time, testing every product change as an experiment before rolling it out.
Feature-based teams without clear success metrics can measure progress through a combination of internal dogfooding buzz, user research, public forum sentiment, and long-term holdout experiments.
Product teams should distinguish between input metrics they can control and output metrics like activation that are lagging indicators, and goal themselves on input metrics.
Product-led growth is essentially data-led growth because without usage data infrastructure, you cannot activate or convert free users.
Companies commonly skip buying an experimentation platform and jump to building their own, which is a mistake because it requires not just engineering but also data science and statistical expertise.
For B2B PLG products, there are four distinct aha moments for different personas: user, team, buyer, and paid-customer, and most companies stop strategizing after the first.
Figma deprecated traditional OKRs in favor of headlines, which are claims teams want to make by end of a time period, evaluated by a mix of quantitative and qualitative signals.
The three-step process for finding your activation metric is: brainstorm potential aha moments, run regression to find correlation with retention, then experiment to prove causation.
Figma's core activation metric is collaboration in the same file with someone else within 24 hours, which allowed them to focus on winning design teams rather than solo users.
Slack's early activation metric was a team with 2+ users having 50+ messages within the first seven days, focusing on team success rather than individual user actions.
The average activation rate across 500+ products is 34% with a median of 25%, and for SaaS products specifically the average is 36% with a median of 30%.
The most common mistake when setting activation milestones is setting them too early (just completing signup) or too late (multiple purchases), when the milestone should be a leading indicator of habit formation.
A good activation metric should be predictive of long-term retention at a rate at least 2x better for users who hit it versus those who don't.
You need goals before you can align around a roadmap because when prioritizing ideas you look at ROI, and the impact you're prioritizing against is based on how much a project will impact your goals.
A feature with low measurable impact can still be worth keeping if it supports the company's long-term strategy, as was the case with Airbnb's Superhost program which initially showed little measurable impact but was aligned with strategy.
The recommended event to use for defining 'active users' in retention measurement is the main user action rather than visits, sessions, or logins, because it gives the cleanest sample of truly active users.
SaaS companies commonly make the mistake of reporting one blended retention metric mixing free and paid users, which is misleading because paid users use the product far more than free users.
Retention is an output metric and should not be used as a baseline for A/B tests because user activity is only one component of retention, and changes in activity often do not lead to measurable changes in retention.
X-day (bounded) retention is more appropriate for SaaS products tied to specific time bounds like trial or subscription length, while unbounded retention is better for B2C consumer transactional or social businesses with chaotic engagement patterns.
Good six-month user retention benchmarks for consumer products are approximately 25% for social, 30% for transactional, and 40% for subscription, with great being roughly 45%, 50%, and 70% respectively.
Noom targeted one-month payback on paid acquisition which allowed immediate reinvestment into their performance marketing engine, making them less reliant on raising money to grow.
Companies underestimate the power of experimentation volume; Noom ran up to six tests per week per PM, with 90% failing, because high velocity of learning beats high effort per test.
For early-stage B2B businesses, investors focus on how quickly they ramp to $1M ARR after going live rather than month-over-month growth percentages, with reaching $1M ARR within 12 months considered good and 9 months considered great.
Best-in-class SaaS companies follow a triple-triple-double-double-double ARR growth pattern after reaching $1M in revenue.
For early-stage consumer businesses, investors focus on intensity of engagement, virality, and retention rather than revenue, with greater than 50% DAU/MAU or greater than 50% D30 retention considered excellent.
The four most important marketplace metrics are fill rate (percentage of intentful sessions converting), bookings growth (completed transactions), supply growth (new active supply), and GMV growth (dollars through the system).
Fill rate is the ultimate measure of marketplace health because it bakes in supply quality, availability, and booking conversion, measuring whether people can consistently find what they want.
80% of marketplaces start out supply-constrained and about half stay that way, making supply growth a major priority for most marketplace businesses.
The most important metrics to track vary significantly by consumer business type: trial-based subscriptions, freemium subscriptions, ad-based, marketplaces, and DTC each require different metric sets.
Greater focus on fewer metrics leads to greater impact; teams should narrow to the two or three metrics that most directly drive business success.
So-called 'design goals' like consistency, usability, and accessibility are actually business goals because they directly drive conversion, loyalty, and efficiency.
Good and great 6-month user retention benchmarks vary dramatically by business type: consumer social (25%/45%), consumer transactional (30%/50%), consumer SaaS (40%/70%), SMB SaaS (60%/80%), enterprise SaaS (70%/90%).
If you plot the percentage of active users over time for various cohorts and the retention curve flattens off at some point, you have probably found product-market fit.
If you survey users asking how they would feel if they could no longer use the product and over 40% say 'very disappointed,' you have product-market fit.
Designers and engineers are most engaged with metrics when they are involved in goal setting, metric definition, regular metric reviews, and post-launch reflections.
Podcast Moments
“EV is enterprise value, TV is your team's value, MEV is your value. Immature managers solve for TV over EV. We always put on the wall: solve for CV, then EV, then TV, then MEV. Customers first, then company, then team, then yourself.”
Sequoia CEO coach: Why it’s never been easier to start a company, and never been harder to scale one | Brian Halligan (co-founder, HubSpot) · Brian Halligan
“When growth stalls, 90% of the time it's a retention problem, not an acquisition problem. But teams always want to fix acquisition first because it's more fun.”
5 questions to ask when your product stops growing | Jason Cohen (2x unicorn founder) · Jason Cohen
“Logo retention is the most honest metric in B2B. Net revenue retention can mask churn with expansion. Logo retention tells you if customers are actually staying.”
5 questions to ask when your product stops growing | Jason Cohen (2x unicorn founder) · Jason Cohen
“In sales, there's rules of eight. Eight SDRs need one manager, eight AEs need a director above them. You can build your whole org with eights.”
We replaced our sales team with 20 AI agents—here’s what happened | Jason Lemkin (SaaStr) · Jason M Lemkin
“The best growth teams I've seen all have one thing in common: they measure retention before acquisition. If your bucket is leaky, pouring more water in doesn't help.”
Elena Verna 4.0 · Elena Verna 4.0
“Instead of building AI that will actually advance us as a species, we are optimizing for AI slop instead. We're basically teaching our models to chase dopamine instead of truth.”
The 100-person AI lab that became Anthropic and Google's secret weapon | Edwin Chen (Surge AI) · Edwin Chen
“Everyone says they have a bottom of funnel problem. It's never a bottom of funnel problem. It's always qualification.”
"Sell the alpha, not the feature": The enterprise sales playbook for $1M to $10M ARR | Jen Abel · Jen Abel
“The biggest mistake in measuring AI productivity is looking at lines of code or PRs per day. You need to measure outcomes: cycle time, change failure rate, deployment frequency.”
How to measure AI developer productivity in 2025 | Nicole Forsgren · Nicole Forsgren
“DORA metrics still work in the AI era, but you need to add new dimensions. We're adding 'AI augmentation rate' — what percentage of work is AI-assisted vs. purely human.”
How to measure AI developer productivity in 2025 | Nicole Forsgren · Nicole Forsgren
“If I wrote a book on building great products, it would have three chapters: jobs-to-be-done, analytical rigor in root cause analysis, and designing for clarity instead of cleverness.”
Inside Google's AI turnaround: The rise of AI Mode, strategy behind AI Overviews, and their vision for AI-powered search | Robby Stein (VP of Product, Google Search) · Robby Stein
“Close Friends totally failed originally. We found it worked for people who added 20 to 30 people because two would reply via DM.”
Inside Google's AI turnaround: The rise of AI Mode, strategy behind AI Overviews, and their vision for AI-powered search | Robby Stein (VP of Product, Google Search) · Robby Stein
“If you have a human process that is 10 or 20% accurate, AI is awesome. If you have one at 98% accurate and expect the remaining 2%, not totally there yet.”
First interview with Scale AI’s CEO: $14B Meta deal, what’s working in enterprise AI, and what frontier labs are building next | Jason Droege · Jason Droege
“Don't just look at your retention curves. Look at what your most retained users do differently in their first week. That behavioral signature is your growth playbook.”
How to find hidden growth opportunities in your product | Albert Cheng (Duolingo, Grammarly, Chess.com) · Albert Cheng
“The biggest mistake teams make is trying to automate all evals. You need human judgment in the loop, especially for anything subjective. The best systems combine both.”
Why AI evals are the hottest new skill for product builders | Hamel Husain & Shreya Shankar (creators of the #1 eval course) · Hamel Husain & Shreya Shankar
“LLM leads are significantly more valuable. Webflow saw a 6X conversion rate difference between LLM traffic and Google Search traffic.”
The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite) · Ethan Smith
“One out of 20 landing pages drive roughly 85% of all your traffic. If you knew the few things that would work, you could push all that money to that one page.”
The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite) · Ethan Smith
“ChatGPT and Google Search citation overlap was around 35%. Perplexity was around 70%.”
The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite) · Ethan Smith
“One of the first things I did was always to build a growth team. When you build a growth team and you hire the right growth leader, they start asking all the right questions because the archetype of person who is a growth PM will be like, 'Well, wait. Why is this happening? Let's get the data.' That's when you realize you don't have X, Y, and Z thing logged.”
From ChatGPT to Instagram to Uber: The quiet architect behind the world’s most popular products | Peter Deng · Peter Deng
“I think you know when your product is really serving people. So much of when you get really metrics obsessed is when you're trying to convince yourself that it is when it's not. That's our north star: do we repeatedly hear from people that Claude is unlocking their own creativity?”
Anthropic’s CPO on what comes next | Mike Krieger (co-founder of Instagram) · Mike Krieger
“A lot of marketing metrics tend to be vanity metrics about the number of clicks, views, and impressions. I think those are all bullshit numbers. What is that experience that you want your customers to come away with when they interact with your brand?”
Growth tactics from OpenAI and Stripe’s first marketer | Krithika Shankarraman · Krithika Shankarraman
“Strategy is a little bit overrated for product. For most product managers, your strategy should be, 'How fast can I go from hypothesis to data?'”
Unconventional product lessons from Binance, N26, Google, more | Mayur Kamat (CPO at N26, ex-Binance Head of Product) · Mayur Kamat
“We really have an approach of very radical transparency about everything. Before we went public, we actually shared every bit of information with our employees. Instead of demoralizing people, it gives them a sense of deep partnership.”
Inside monday.com’s transformation: radical transparency, impact over output, and their path to $1B ARR | Daniel Lereya (Chief Product and Technology Officer) · Daniel Lereya
“We had the daily numbers concept. You buy a TV, put it on the wall, and when you had a new paying account, you had the Simpson saying the same with the $1 million. For new signups, you had a tick. Suddenly, everyone is living it.”
Inside monday.com’s transformation: radical transparency, impact over output, and their path to $1B ARR | Daniel Lereya (Chief Product and Technology Officer) · Daniel Lereya
“If you haven't tried Superhuman, then gosh, what are you doing? Getting through your email twice as fast, responding one to two days sooner, saving four hours or more every single week, they're all real.”
Superhuman's secret to success: Ignoring most customer feedback, manually onboarding every new user, obsessing over every detail, and positioning around a single attribute: speed | Rahul Vohra (CEO) · Rahul Vohra
“The company was on the verge of going under when we launched Bolt, and what ended up happening is, in the first two months it went from zero to 20 million of ARR. And we've already crossed 30 million of ARR, with the current rate we're on, our forecast for the year is we want to get to 100 million of ARR.”
Inside Bolt: From near-death to ~$40m ARR in 5 months—one of the fastest-growing products in history | Eric Simons (founder and CEO of StackBlitz) · Eric Simons
“Test everything. We've run in the last four years over 600 experiments on the streaks, so every other day. We've actually set up really good infrastructure for copy testing.”
Behind the product: Duolingo streaks | Jackson Shuttleworth (Group PM, Retention Team) · Jackson Shuttleworth
“When you have teams naturally break up the world into different funnel stages, it gets very seductive to look at my part of the funnel and what's my conversion rate. But in practice, it's actually almost always easier to just make it harder to do the thing right before your step in the funnel to increase your conversion rate.”
Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, prioritizes intuition, and builds toward a 100-year vision | Archie Abrams (VP Product, Head of Growth at Shopify) · Archie Abrams
“The best way to get more people to get to a step is just get more people in the door in the first place. That will always hurt your conversion rate, but it may actually give you more people on the outside.”
Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, prioritizes intuition, and builds toward a 100-year vision | Archie Abrams (VP Product, Head of Growth at Shopify) · Archie Abrams
“Shopify doesn't have KPIs in the traditional sense. We have a north star which is GMS, gross merchandise sales, and everything else is context. We trust people to make good decisions.”
Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, prioritizes intuition, and builds toward a 100-year vision | Archie Abrams (VP Product, Head of Growth at Shopify) · Archie Abrams
“The biggest prioritization mistake is using impact-effort matrices. They give you a false sense of rigor. In reality, the highest-impact items are almost never low effort, and the low-effort items are almost never high impact.”
4 questions Shreyas Doshi wishes he’d asked himself sooner | Former PM leader at Stripe, Twitter, Google · Shreyas Doshi
“The question is, how would you feel if you could no longer use this product? Once you got a high enough percentage of users saying they'd be very disappointed, most of those products did pretty well. If you felt too low, those products tended to suffer.”
The original growth hacker reveals his secrets | Sean Ellis (author of “Hacking Growth”) · Sean Ellis
“Just ignore the people who say they'd be somewhat disappointed. They're telling you it's a nice to have. If you start paying attention to what your somewhat disappointed users are telling you and then you start tweaking onboarding and product based on their feedback, maybe you're going to dilute it for your must have users.”
The original growth hacker reveals his secrets | Sean Ellis (author of “Hacking Growth”) · Sean Ellis
“I start with the value that's uncovered through the test. So with a company, I'll say, 'This is what the must have value is according to our most passionate customers, and we want to think about a metric that reflects us delivering that value.'”
The original growth hacker reveals his secrets | Sean Ellis (author of “Hacking Growth”) · Sean Ellis
“We all talk about product sense. To me, it's just a fancy way of saying you can make good decisions with insufficient data. PMs need as many reps as possible in making decisions, documenting the rationale behind those decisions, and then crucially seeing the outcome of them.”
Unorthodox PM wisdom: Automating user insights, unselling job candidates, logging every decision, more | Kevin Yien (Stripe, Square, Mutiny) · Kevin Yien
“Instead of thinking about being on top of the page, and that's like ego marketing, I want to be number one. I want to be there all the time. It's about showing to the right person as often as possible.”
The ultimate guide to performance marketing | Timothy Davis (Shopify) · Timothy Davis
“For me, analytics is a business impact driving function and not purely a service function, not just answering the why, but answering the, 'What do we do now that we know this?'”
Building a world-class data org | Jessica Lachs (VP of Analytics and Data Science at DoorDash) · Jess Lachs
“Retention is a terrible thing to goal on. It's almost impossible to drive in a meaningful way in a short term. Ultimately, you want to find a short-term metric you can measure that drives a long-term output.”
Building a world-class data org | Jessica Lachs (VP of Analytics and Data Science at DoorDash) · Jess Lachs
“Yes, you are a data scientist, but your goal is to figure out what's happening. And if that means that you're going to pick up the phone and call customers, then that is what you're going to do to roll up your sleeves.”
Building a world-class data org | Jessica Lachs (VP of Analytics and Data Science at DoorDash) · Jess Lachs
“What was the value that we're trying to produce for the customer and can we measure it from their perspective? And okay, how do you know you have product market fit? Charts that showcase things are going up into the right on one hand and then tweets on the other.”
Building product at Stripe: craft, metrics, and customer obsession | Jeff Weinstein (Product lead) · Jeff Weinstein
“The Safety Funnel is amazing. You basically put a hard stop and you limit the number of people who had bad experiences. And you do that for a while, up until you can prove it's amazing and then you invite more people.”
Hard-won lessons building 0 to 1 inside Atlassian | Tanguy Crusson (Head of Jira Product Discovery) · Tanguy Crusson
“What I call the anti-pattern of what we want to do. Someone says, 'Hey, you know what? This would be great to build.' And you go pull data to go justify why that would be great to build. Call that identify, justify, execute. First you have to really understand from first principles what is actually going on. So understand, identify, execute.”
Unorthodox frameworks for growing your product, career, and impact | Bangaly Kaba (YouTube, Instagram, Facebook, Instacart) · Bangaly Kaba
“We were losing 10, 12 million people a year from what we called account access churn. So we worked on this problem... just being really thoughtful around what is actually the core job that people are trying to do.”
Unorthodox frameworks for growing your product, career, and impact | Bangaly Kaba (YouTube, Instagram, Facebook, Instacart) · Bangaly Kaba
“We built a lending product, we built an investment product, we built an insurance product, we built a series of small business products. We rarely scale a project until we know the Sean Ellis score hit a threshold that we find really compelling.”
Be fundamentally different, not incrementally better | Jag Duggal (Nubank, Facebook, Google, Quantcast) · Jag Duggal
“We've published dozens of articles on the First Round Review, and we have found a very consistent set of patterns, demand satisfaction, and efficiency. But the interesting thing is that you don't go for all three of them from the very beginning.”
A framework for finding product-market fit | Todd Jackson (First Round Capital) · Todd Jackson
“Roughly, 60% are never going to get past L2.”
A framework for finding product-market fit | Todd Jackson (First Round Capital) · Todd Jackson
“I remember in a board meeting, the new model really started to show legs and one of the board members, Brian Schreier at Sequoia, said it was the prettiest smile graph that he had ever seen.”
The happiness and pain of product management | Noam Lovinsky (Grammarly, Facebook, YouTube, Thumbtack) · Noam Lovinsky
“Humanizing the product development process means remembering that there are real people on the other side of every feature you ship, and that the metrics you're optimizing for should ultimately map back to improving someone's life.”
Humanizing product development | Adriel Frederick (Reddit, Lyft, Facebook) · Adriel Frederick
“The problem with funnels and pirate metrics and the favorites that I love to pick on are MQLs and SQLs is that nobody knows what those mean. It puts every customer in the same sort of buckets. It assumes that all customers and all products are the same.”
Customer-led growth | Georgiana Laudi (Forget The Funnel) · Gia Laudi
Cutting Room Floor
Guest insights on this topic that Lenny hasn't (yet) written about in his newsletters. Potential material for future posts.
“In sales, there's rules of eight. Eight SDRs need one manager, eight AEs need a director above them. You can build your whole org with eights.”
We replaced our sales team with 20 AI agents—here’s what happened | Jason Lemkin (SaaStr) · Jason M Lemkin
“If you have a human process that is 10 or 20% accurate, AI is awesome. If you have one at 98% accurate and expect the remaining 2%, not totally there yet.”
First interview with Scale AI’s CEO: $14B Meta deal, what’s working in enterprise AI, and what frontier labs are building next | Jason Droege · Jason Droege
“ChatGPT and Google Search citation overlap was around 35%. Perplexity was around 70%.”
The ultimate guide to AEO: How to get ChatGPT to recommend your product | Ethan Smith (Graphite) · Ethan Smith
“Strategy is a little bit overrated for product. For most product managers, your strategy should be, 'How fast can I go from hypothesis to data?'”
Unconventional product lessons from Binance, N26, Google, more | Mayur Kamat (CPO at N26, ex-Binance Head of Product) · Mayur Kamat
“Test everything. We've run in the last four years over 600 experiments on the streaks, so every other day. We've actually set up really good infrastructure for copy testing.”
Behind the product: Duolingo streaks | Jackson Shuttleworth (Group PM, Retention Team) · Jackson Shuttleworth
“When you have teams naturally break up the world into different funnel stages, it gets very seductive to look at my part of the funnel and what's my conversion rate. But in practice, it's actually almost always easier to just make it harder to do the thing right before your step in the funnel to increase your conversion rate.”
Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, prioritizes intuition, and builds toward a 100-year vision | Archie Abrams (VP Product, Head of Growth at Shopify) · Archie Abrams
“The best way to get more people to get to a step is just get more people in the door in the first place. That will always hurt your conversion rate, but it may actually give you more people on the outside.”
Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, prioritizes intuition, and builds toward a 100-year vision | Archie Abrams (VP Product, Head of Growth at Shopify) · Archie Abrams
“Instead of thinking about being on top of the page, and that's like ego marketing, I want to be number one. I want to be there all the time. It's about showing to the right person as often as possible.”
The ultimate guide to performance marketing | Timothy Davis (Shopify) · Timothy Davis
“For me, analytics is a business impact driving function and not purely a service function, not just answering the why, but answering the, 'What do we do now that we know this?'”
Building a world-class data org | Jessica Lachs (VP of Analytics and Data Science at DoorDash) · Jess Lachs
“Yes, you are a data scientist, but your goal is to figure out what's happening. And if that means that you're going to pick up the phone and call customers, then that is what you're going to do to roll up your sleeves.”
Building a world-class data org | Jessica Lachs (VP of Analytics and Data Science at DoorDash) · Jess Lachs
“What was the value that we're trying to produce for the customer and can we measure it from their perspective? And okay, how do you know you have product market fit? Charts that showcase things are going up into the right on one hand and then tweets on the other.”
Building product at Stripe: craft, metrics, and customer obsession | Jeff Weinstein (Product lead) · Jeff Weinstein
“We've published dozens of articles on the First Round Review, and we have found a very consistent set of patterns, demand satisfaction, and efficiency. But the interesting thing is that you don't go for all three of them from the very beginning.”
A framework for finding product-market fit | Todd Jackson (First Round Capital) · Todd Jackson
“Roughly, 60% are never going to get past L2.”
A framework for finding product-market fit | Todd Jackson (First Round Capital) · Todd Jackson
“I remember in a board meeting, the new model really started to show legs and one of the board members, Brian Schreier at Sequoia, said it was the prettiest smile graph that he had ever seen.”
The happiness and pain of product management | Noam Lovinsky (Grammarly, Facebook, YouTube, Thumbtack) · Noam Lovinsky
“Humanizing the product development process means remembering that there are real people on the other side of every feature you ship, and that the metrics you're optimizing for should ultimately map back to improving someone's life.”
Humanizing product development | Adriel Frederick (Reddit, Lyft, Facebook) · Adriel Frederick
“The problem with funnels and pirate metrics and the favorites that I love to pick on are MQLs and SQLs is that nobody knows what those mean. It puts every customer in the same sort of buckets. It assumes that all customers and all products are the same.”
Customer-led growth | Georgiana Laudi (Forget The Funnel) · Gia Laudi