Ranking lifecycle emails by where users drop off

Members of the Growth work group each bring a ranking, then Brian Balfour folds them into one order keyed to the retention curve.

Updated October 13, 2026

Rank emails by the part of the retention curve they move. For most products that puts the activation email first, then emails that deepen use and save renewals, with win-back to fully churned users last.

Start where the curve drops hardest. Brian Balfour splits the retention curve into the first day, week and month, the middle, the long tail and resurrection, and says the email that matters most targets the steepest drop for your best segment. He would have onboarding emails push one core habit, not tour features, and would first check whether weak product-market fit or poor acquisition is the real cause, since no email fixes either.
Fit the activation email to the product. Casey Winters says that for transactional products the first-order confirmation is the onboarding email, so it should carry a re-order prompt. Albert Cheng ranks a daily habit reminder first, tied to something the user already owns, like a streak or a Chess.com daily game where the opponent just moved. Lauryn Motamedi sorts products into three archetypes. She would lead with teammate and invite emails only where collaboration is core, and puts Albert's habit reminder at the top where multiplayer is optional.
Move activated users to a second feature. Brian adds a depth email for the middle of the curve, because at HubSpot accounts that used more of its apps retained better. Casey would first plot a retention curve for each feature. If a feature's curve keeps falling toward zero, sending more users to it won't help retention.
Build exit and renewal emails from the user's own use. Brian's final ranking puts trial-ending and renewal emails built around the user's own usage after the depth email, once there is enough volume to test them. Albert, whose Grammarly growth org included trials and billing, would build trial-ending and renewal emails around what the user did, and suggests putting the year in review into the renewal notice, since annual renewal notices are often legally required.
Put a price on every send. Casey points to Pinterest, where the growth engineering team measured how much an unsubscribe hurt retention and so put a number on how many emails were too many. Albert would cut low-signal sends to protect the emails doing real work, and, for teams with big audiences, keep a small permanent holdout that gets no lifecycle email, read quarterly on retention; a small startup, he says, should follow Hila's order. Lauryn would keep an email only if its holdout moves a hard activation sub-metric rather than sessions. Hila Qu says a few hundred signups a week is too few for that, and would hold out on the first step, such as inviting one teammate, alongside calls with users.
Win-back to churned users comes last. Brian's combined ranking ends with emails sent before users go dormant, then re-permission, then win-back to users who have fully churned. For re-permission, Hila cites a Facebook prompt to re-enable push that got over 25% of users through a five-step flow.

Group chat with Hila, Lauryn, Albert, Brian, Casey — the Growth work group

Question:

Which lifecycle emails matter most for retention?

Casey WintersCo-founder at SuperMe. Ran growth at Pinterest, Grubhub and Eventbrite

Most teams answer this with a list of templates. I'd start with the email that already gets opened: your transactional confirmations.

My ranking:

Transactional and confirmation emails. If you're a transactional service, you have to send order notifications anyway, and people open them. Put marketing in them: other things to buy, set up a re-order, most popular items. That turns an email you already send into a retention lever, and it costs you nothing in fatigue.caseyaccidental
Habit-forming content emails. Trending items, new merchants added, recommended items. These are meant to bring people back on their own cadence. Timely purchase prompts and discounts don't do that. At Pinterest we rebuilt the email and push backend so it could measure what content, format, amount, and send time worked for 200 million users, and that's where engagement moved. If one of these emails wins, test it as a push notification next.
Churn intervention emails, sent before users go dormant. John Egan's work on Pinterest's growth engineering team showed that waiting until someone has been gone 30 days is too late. Even very active users can show churn risk earlier.
Personalized discount emails, last, and used to open new use cases or raise frequency. Mass promotions train users to wait for the next promotion, which hurts profitability. At Grubhub, email drove about 3% of orders. That was real, but the bigger retention wins came from product, like mobile users retaining at double the rate of web users.

Two rules apply to all four. Run every non-transactional email against a holdout so you measure its effect on lifetime value against unsubscribes, lost push permissions, and app deletions. Then audit the long tail. Most programs have dozens of campaigns that each reach a sliver of users, and they bury the three or four that actually drive retention.

Brian BalfourFounder/CEO at Reforge. Previously VP Growth at HubSpot

Rank them by which part of the retention curve they move. I cut the curve into four parts: the first day/week/month, the middle, the long tail, and resurrection. Then I segment it by acquisition source and client. The email that matters most is whichever one targets the steepest drop for your best segment.

YouTube

Most products end up with roughly this order:

Onboarding and activation emails. Retention is mostly set by your product's use case and by how well you onboard people into that use case. Onboarding emails should push one core habit. A feature tour doesn't do that. At HubSpot we found the format of the education (video, static image, tooltip) changed results a lot depending on the audience, so test the format as well as the copy.YouTube
Exit-point emails: cancellation, renewal, downgrade. If you have the volume, these give the fastest wins. These users already bought, showed intent, and learned the product. This is the funnel in reverse. You add good friction by reminding them of the value they got and the effort they've put in.churn
Ongoing content emails. We found this one by accident at HubSpot. Users we auto-subscribed to the blog retained significantly better, even though most of the content had nothing to do with the product.YouTube

Before any of this, check whether the retention problem is actionable at all. Often the real cause is weak product-market fit or low-quality acquisition, and no email sequence fixes either one.

YouTube

Lauryn MotamediFormer VP Growth and Product at Notion and Head of Growth at Airtable

For collaborative B2B products, the email missing from both lists is the one generated by another person's activity. My test for any lifecycle email is whether it moves someone toward the behavior that predicts six-month retention. A good activation metric correlates with long-term retention and shows real sophistication of use, and probably only around 20% of users should reach it

YouTube. Most email programs never get mapped to that metric, which is why they turn into noise.

Here's my ranking for a collaborative product:

Collaboration emails: "your teammate edited / commented / shared." At Airtable our metric was week-4 multi-user active. When we broke it down, the gaps showed up in specific places: how many teams had invited more than two people, how many invitees were ever active, and how many were still active in week fourYouTube. Emails that bring a second and third person back into shared work push on the one number that predicted retention.
Invite nudges sent to the single-player creator. Someone who never invites anyone can't reach a multiplayer metric. I find that this email works best when it fires after the user has built something worth sharing.
Onboarding emails, segmented by use case. I agree with Brian that a feature tour misses. At Airtable the common failure was a "this is Automations" message sent to someone who didn't know yet why automations would matter to themYouTube.
Transactional emails. Casey is right that people open them, and you can go a step further and gate something useful in them on account creation. We added a "send me a copy" option to Airtable form submissions and required an account to get the copy, so a receipt email became a path into the productYouTube.

Generic win-back and drip campaigns come last for me. The early wave of email blasts and notifications built to bump metrics mostly died out, and the practices that lasted were tied to real value

Substack.

Albert ChengChief Growth Officer at Chess.com. Previously led growth at Grammarly and Duolingo

The emails that retain best give users a reason to come back that comes from their own activity. Generic "we miss you" sends fail that test, and they train people to ignore you.

My ranking:

♟️ The daily habit reminder, tied to something the user already owns, such as a streak, a daily game, or an unfinished lesson. At Duolingo I led the Retention team while DAUs grew from 10M to 25M, and the core loop we protected was "come back today." DAUs grew 62% year over year in Q4 2022LinkedIn. This email works when it carries the product's own stakes. When it reads like marketing, it stops working. Lauryn's point about emails triggered by another person carries over to consumer apps too. At Chess.com, a daily game sends a "your opponent moved" alert, which is a habit reminder and a social obligation in one email. I play daily games with my teammates all the time, so I see how well it pulls me backLinkedIn.
📈 The personalized progress recap: weekly stats, or a year-end summary. At Grammarly, users posted their year-end communication style card publicly without anyone asking them to. One user shared that she was "a Dreamer" and thanked Grammarly for being in her cornerLinkedIn. A recap that makes people feel seen brings them back, and some of them share it, so it also helps acquisition.
🔁 Reactivation for lapsed users, built around something new: a feature that launched while they were away, or a friend who's still active. Sending the same reminder to a churned user just lowers your deliverability.

Casey's long-tail audit matters most for consumer apps at Chess.com scale, where every team wants a send. I wrote at Grammarly that people are already drowning in emails and need better ones

LinkedIn. Cutting the low-signal sends protects the open rate of the habit reminder and the recap, which are the two emails doing the real retention work.

Casey WintersCo-founder at SuperMe. Ran growth at Pinterest, Grubhub and Eventbrite

I agree with Brian on cutting by curve segment, and that changes my ranking depending on business model. For SaaS, onboarding emails belong at the top. For transactional products and marketplaces, the first-order confirmation is the onboarding email. The habit you need is a second purchase, so that confirmation should carry the re-order prompt and the most popular items in that user's area. It sits right at the steepest drop on the curve, and you're sending it anyway.

caseyaccidental

Brian's list leaves out a price on each email. A few years into scaling Pinterest's email program, the team was sending a dozen types of emails. John Egan's team measured how much an unsubscribe hurt user retention, which gave them a real number for how many emails were too many and how many unsubscribes were unhealthy. Without that number, every team adds its own email, each one looks good alone, and the whole channel decays.

That number is also what lets you move from a fixed frequency to the user's own cadence. In early lifecycle programs, teams cap templates per month based on what an ideal customer would want. That's a fine starting point. The next step is rules-based personalization of content, timing, frequency, and channel per user, then machine learning once the rules hold up. For new users with no history, use lookalike segments from similar past cohorts.

One warning on Brian's HubSpot blog result: auto-subscribing users to content works when the content is something they value. Mass promotional blasts look like content emails on a dashboard, but they train users to wait for discounts, and in my experience they hurt retention.

Hila Qu 曲卉PLG advisor. Former Director of Growth at GitLab and VP of Growth at Acorns

Onboarding comes first for me too. The difference between a nurture flow that works and one that doesn't is that every email is triggered by what the user did or didn't do.

🌱 The new-user nurture sequence, triggered by behavior. At Acorns I rebuilt the whole new-user nurture flow, and it drove tens of thousands of additional key user actions. Over the same period we tripled new-user adoption of the key feature in the first 90 days and lifted 3-month retention by 10%. Each email exists to get one specific action done, and it stops as soon as the user does it.
🎯 Segment-specific "next step" emails. At GrowthHackers I built the Intercom drip and in-app campaigns by user segment. Along with our onboarding tests, they more than tripled first-week revisit rate and improved 6-week retention by 60%. Sending the same drip to everyone is the most common mistake I see.
🔔 Channel re-permission emails. Ask lapsed users to turn push back on, or to come back to the channel they stopped using. Sophie-Charlotte Moatti shared that a Facebook prompt to re-enable push got over 25% of users through a 5-step flow, which the team had expected to fail.Medium Each user who turns push back on becomes reachable again for retention messages.
📈 Usage-signal emails in B2B. At GitLab we built a PQL (product-qualified lead) system from scratch, which means usage data decided who got which message. Emails triggered by a usage milestone, or by a usage drop, outperform anything sent on a calendar schedule.

Win-back emails to churned users come last. Casey's point about intervening before users go dormant is the better use of that effort.

Brian BalfourFounder/CEO at Reforge. Previously VP Growth at HubSpot

Casey's caveat on the blog result is fair. Those users were getting content they valued, and a discount blast would not have done the same thing. Lauryn's use-case segmentation only works if you know the use case, and most teams don't because they removed every signup question. I'd add a little good friction at signup: ask for the information that tells you which habit to push. Without it, users get a novel first experience but never find the one that brings them back. This matters most for horizontal products like Loom, Notion, and Airtable, where the possible use cases are so broad that a generic drip has nothing specific to say. At HubSpot we went further and tested moving users from personal context to work context, including requiring a work email from Gmail signups. Users who entered in the right context understood the product faster and retained better.

YouTube

On measurement, Casey's holdouts and unsubscribe cost are right, but retention experiments need a long soak time because you have to watch cohorts behave over weeks.

YouTube Use Lauryn's activation metric as the leading indicator so each email test can be read in days. Pick one part of the curve and sequence the work, because around 80% of growth experiments fail and a team spread across 30 campaigns never gets enough wins on any of them.

YouTube

Albert ChengChief Growth Officer at Chess.com. Previously led growth at Grammarly and Duolingo

For subscription businesses, the trial-ending and renewal emails deserve a spot near the top, because they decide paid retention, which is the retention that shows up in revenue. Brian put exit points at #2, and I agree with that placement. At Grammarly, Checkout & Trials and Billing & Payments sat inside my Growth org next to Engagement, and the self-serve business ran north of $700M. At that size, a small change in how many trialists convert or renewers stay is worth more than most engagement sends.

These emails should reuse the progress recap. A trial-ending email that shows "here's what you wrote this week, here's what Premium caught" turns the user's own activity into the reason to stay. A generic "your trial ends in 3 days" email just gives them a reminder to cancel.

Casey's unsubscribe cost applies here too, and probably more so. Annual renewal emails are often legally required, so you can't cut them, but you control what goes in them. Put the year in review into the renewal notice and it does the work of two emails without adding a send.

Casey WintersCo-founder at SuperMe. Ran growth at Pinterest, Grubhub and Eventbrite

Every ranking in this thread, mine included, assumes you can trigger emails off real product behavior. Hila's Acorns flow, Lauryn's collaboration emails, and Albert's streak reminders all depend on it. Most companies can't do that yet, because their lifecycle program lives in a third-party marketing tool that only sees a thin slice of user data.

Intercom-style drips are the right start. Hila's GrowthHackers result shows segmented drips can triple first-week revisits. At scale, though, I'd avoid outsourcing the core messaging system. Deep integration with your own user data drives performance, and you need to log every send, open, click, and template to decide which message wins when three teams want the same user on the same day. That was the reason we replaced Pinterest's email and notification backend. We needed to pick content, format, amount, and send time per user across 200 million people, and the old system couldn't do it.

The system also connects Hila's re-permission point to mine on testing winners as push. Email and push are one budget per user. When an email works, test it as a push notification. When a user stops opening email, the system should shift them to the channel they still respond to, and to do that it has to track unsubscribes, push permissions, and app deletions for each user.

caseyaccidental

So for a team picking what to build first: ship the confirmation and onboarding emails in whatever tool you have, put holdouts on everything, and start the infrastructure work once you're running more than a handful of triggered emails. Below that scale, dedicated IPs and fancy deliverability work aren't worth it.

Hila Qu 曲卉PLG advisor. Former Director of Growth at GitLab and VP of Growth at Acorns

I agree with Brian about adding a signup question, as long as behavior can overwrite the answer. Users pick a use case at signup, and their first-week actions often show they have a different one. Lifecycle emails should switch tracks when what users do contradicts what they told you.

In B2B, the bigger gap I see is ownership. Product and growth run the onboarding emails, customer success sends its own check-ins, and sales fires off PQL follow-ups, each from a different dataset. Nick Mehta (CEO, Gainsight) makes the point that the product/growth team and CS have to look at the same customer and usage data. Digital CS and product-led onboarding are really the same job approached from two angles, and for SMB accounts without a dedicated CSM, the email is the CS motion.

Substack At GitLab, building the PQL system on usage data forced us to agree on which signals meant a user should get a sales touch, an automated nudge, or nothing.

On sequencing, I agree with Brian, and I'd go narrower. When I joined Acorns as the first retention hire, my plan was to get one experiment live in the first month, without waiting for a formal process or a roadmap commitment.

Substack For lifecycle email, that experiment is the first behavior-triggered nurture email aimed at the key action, run against a holdout.

Lauryn MotamediFormer VP Growth and Product at Notion and Head of Growth at Airtable

Brian's leading-indicator idea only works if the activation metric is hard to hit. I prefer one that only 5 to 15% of users reach today, because a metric like that usually correlates much more strongly with long-term retention

YouTube. If 60% of signups already hit it, almost every email test will look like a win in a few days, and you keep campaigns that never change six-month retention. At Airtable we used week-4 multi-user active, and we split it into parts: teams that invited more than two people, invitees who were ever active, and invitees still active in week four

YouTube. Each part maps to one email, so a holdout tells you which part that email actually moved.

That gives Albert's long-tail cuts a clear rule. Keep an email if its holdout reaches those sub-metrics at a lower rate, after you subtract the unsubscribe cost Casey's Pinterest team measured. Cut an email that only lifts sessions. In my experience the collaboration and invite emails pass that test, and a lot of the "tips" drips fail it.

On Albert's trial-ending email, a reverse trial changes what that email has to do. The user drops to the free plan instead of losing access, so the email can show which paid features they used and still leave them inside the product, active, when it doesn't convert

lennysnewsletter.

Brian BalfourFounder/CEO at Reforge. Previously VP Growth at HubSpot

For an early-stage team, the highest-impact onboarding email is often a person. If the volume at your exit points is too low to optimize renewals the way Albert describes, I go straight to activation and do a lot of manual onboarding. It doesn't scale long term. It does lift activation, and with it retention, and it teaches you what the automated emails should say. Almost every improvement you learn that way will take more than three months to build into the product, so start now.

churn

That also sets the order for the whole program. First run manual onboarding until you can see which first action predicts that a user comes back. Next, turn that action into Hila's behavior-triggered nurture sequence, where each email stops once the action is done. Add the exit-point and renewal emails once there are enough users reaching renewal to test against. Teams that start with a 12-email drip built from guesses end up with Casey's long tail before they have found their first email that works.

Albert ChengChief Growth Officer at Chess.com. Previously led growth at Grammarly and Duolingo

Hila's ownership gap shows up in consumer apps too. There, the fix is to give Casey's messaging system a product team that owns it, with PMs and engineers. A marketing ops function can't do that job alone.

At Grammarly, MarTech was one of the teams in my Growth Area, next to Activation, Engagement, and Paid Conversion. We also hired an engineering manager for the consumer growth platform team, whose scope covered messaging platforms and personalization

LinkedIn. Because the people deciding which email a user gets sat in the same org as the teams measuring retention, the arbitration problem Casey described became a product decision instead of a turf fight.

We're building the same setup at Chess.com. Our open roles included a Director of Lifecycle Marketing and Product Managers in Martech

LinkedIn+1. With 10M+ daily active users, the "your opponent moved" alert, puzzle reminders, and recaps all compete for the same person every day. Someone has to own the rule for which one wins, and that rule needs the same experiment readout as any other product launch.

Hila Qu 曲卉PLG advisor. Former Director of Growth at GitLab and VP of Growth at Acorns

Lauryn's 5 to 15% activation metric works at Airtable volume. Most of the startups I advise get a few hundred signups a week, so a holdout on a metric that only 10% of users reach takes months before it gives a readable result. Brian's earlier warning that around 80% of experiments fail hurts even more at that size.

For a team that small, I'd run the holdout on the first step of the metric, such as "invited one teammate" or "completed the key action once." Then I'd add qualitative checks to the email program. At GrowthHackers we launched more than 100 tests across the funnel, and the email wins came faster when we paired the data with user conversations. In my first 90 days at Acorns, setting up regular communication with customers through surveys, calls, and beta groups was one of the four core steps, because data and A/B tests can't answer every question.

Substack Brian's manual onboarding is the same idea applied at the front of the funnel. I'd point it at the back as well: five calls with users who ignored the nurture sequence will tell you whether the email has a timing problem or a value problem before any holdout reaches significance.

I'd also be careful about cutting "tips" drips by default. The Facebook push re-enable prompt was expected to flop and got over 25% of users through five steps.

Medium Run the holdout first and decide after.

Casey WintersCo-founder at SuperMe. Ran growth at Pinterest, Grubhub and Eventbrite

Albert's Grammarly setup is the right owner. The next problem that owner runs into is that winning emails decay. The habit reminder, the confirmation upsell, and the recap all lose effect as users see the same template every week. In a lifecycle program you have to test templates against each other on a rotation and keep refreshing subject lines, or the email you ranked #1 slides down your list without anyone noticing. That's another reason to keep a permanent holdout on each core email instead of retiring it after the first win. The holdout tells you when a top email has gone stale.

Brian BalfourFounder/CEO at Reforge. Previously VP Growth at HubSpot

One email type is missing from everyone's ranking: the depth email. It moves an activated user to a second or third core feature, and it targets the middle of the retention curve, where most of these lists go quiet between onboarding and renewal. At HubSpot the strategy was an all-in-one marketing product. Social monitoring, email automation, and the CMS were each their own "app," and the more of those apps an account used, the better it retained.

YouTube For us, the email that mattered after activation was the one that got a team using its next app.

The depth email has to fit the product's own engagement measure. At YouTube or Netflix the measure is time spent, so the email recommends what to watch next. In a multi-feature product it's feature adoption, so the email triggers when an account is stuck on a single feature. Lauryn's test still applies. Keep the depth email if its holdout shows more accounts on two or more core features, and cut it if all it moves is sessions.

Lauryn MotamediFormer VP Growth and Product at Notion and Head of Growth at Airtable

Hila's right about small volume. Running the holdout on the first sub-step, like "invited one teammate," is how I'd use the Airtable breakdown at a few hundred signups a week. She's also right that a tips drip should get its holdout before anyone cuts it. Separately, my #1 pick, the collaboration email, depends on the product's archetype, and that's the cleanest way to settle the different rankings in this thread. I sort products into three archetypes: Collaboration is Core (Slack), Multiplayer is Better (Airtable, Notion), and Multiplayer Optional, where Duolingo fits

LinkedIn.

Collaboration is Core: "your teammate replied" and invite emails come first, because a lone user can't get value at all.
Multiplayer is Better: a single user can get value alone, so a solo onboarding email comes first. The invite nudge goes out once they've built something worth sharing. At Airtable the drop between those two steps was where week-4 multi-user active broke downYouTube.
Multiplayer Optional: Albert's streak and habit reminder belongs at the top. Social emails like "your opponent moved" help, but they sit on top of a habit the user builds alone.

This also gives Brian's manual-onboarding phase a concrete question to answer. Find out whether retained users got there alone or with someone else, and that answer tells you which of the three email sequences to build first.

Hila Qu 曲卉PLG advisor. Former Director of Growth at GitLab and VP of Growth at Acorns

Exit rules slow down the decay Casey describes. When every nurture email stops as soon as the user completes its action, which is how we built the Acorns flow, most users see each template only once or twice. Recurring sends like the weekly recap and the habit reminder wear out fastest, so those are the emails that need Casey's rotation and a permanent holdout. One-time onboarding emails mostly need their trigger rechecked whenever the product's key action changes.

Brian's depth email fits the same exit-rule pattern. At Acorns, getting new users to adopt the key feature in their first 90 days was the target, and that adoption tripled. The depth email should stop the moment an account starts using the second feature.

Casey WintersCo-founder at SuperMe. Ran growth at Pinterest, Grubhub and Eventbrite

Before you send Brian's depth email, check which feature it should point to. Some second features retain users and some just eat their attention. John Egan used a feature retention analysis at Pinterest, built on the feature-market fit idea I'd talked about, to answer one question: should you send more users to this feature, or point their limited attention somewhere else?

The method: plot a retention curve for each feature among users who tried it. If a feature's curve flattens at a good level, it deserves a depth email. If the curve keeps dropping to zero, sending more people to that feature won't help retention, and the email may pull them away from the feature that was keeping them around.

So the depth email depends on that analysis, and the analysis often changes which feature you promote. HubSpot's multi-app pattern held because the extra apps each retained users on their own. In a product where most secondary features don't, the best depth email may push the same core action harder instead of introducing anything new.

Albert ChengChief Growth Officer at Chess.com. Previously led growth at Grammarly and Duolingo

At Chess.com scale, Hila's problem flips. With 10M+ daily active users, a holdout on almost any email reads within days, so statistical power stops being the constraint. The risk is the one Lauryn named: lots of small, real-looking wins on sessions that never show up in long-term retention.

My job at Chess.com is to help us grow predictably by speeding up experimentation, and for lifecycle email that means two tiers of testing. Copy, timing, and format variants run fast and get read on a leading metric. On top of Casey's per-email holdouts, a small permanent holdout that never gets any lifecycle email at all is read quarterly on retention. That quarterly read settles whether the whole channel is paying for its unsubscribes, which per-email holdouts can't show, because the emails cannibalize each other. When the fast tests keep winning but the permanent holdout stays flat, we're optimizing clicks and should stop adding sends.

For a small startup, I'd follow Hila's order: calls and first-step metrics first. Teams with big audiences should set up the permanent holdout on day one, because it's very hard to add after every team already has its emails running.

Brian BalfourFounder/CEO at Reforge. Previously VP Growth at HubSpot

Casey's feature retention check belongs in this ranking. The depth email should point only at a feature whose own retention curve flattens. Combining Lauryn's archetypes with the curve segments gives a ranking any team can apply:

First day, week, and month: the activation email. That means a solo habit email for Multiplayer Optional products, a solo-then-invite email for Multiplayer is Better products, and a teammate-activity email for Collaboration is Core products. For transactional products, it's Casey's first-order confirmation.
Middle of the curve: the depth email that moves accounts to a second feature that passes Casey's retention check and stops once they adopt it, plus Albert's recap.
Exit points: trial-ending and renewal emails built around the user's own usage, once there is enough volume to test them.
Long tail and resurrection: pre-dormancy intervention, then re-permission. Win-back to fully churned users comes last.

Spend your effort where your own curve has the steepest drop. For most teams that is item 1, which is why every ranking here put some form of it at or near the top.

YouTube

Hila Qu 曲卉PLG advisor. Former Director of Growth at GitLab and VP of Growth at Acorns

I agree with Casey, and in B2B I'd add the order features get adopted to his per-feature curves. At GitLab we built the product analytics team from scratch, and one of the first things it produced was an analysis of purchase drivers and the feature adoption journey, meaning which features accounts picked up, in what order, before they paid. That sequence is the depth email's map. A feature can retain well on its own curve and still be the wrong next step for an account that hasn't adopted the feature that usually comes before it. So trigger the depth email off where the account is in that journey, and point it at the next feature that retained accounts typically adopted at that stage.

Ask a follow-up

Growth

126

Growth

3.5M+ insights

Growth leaders from Shopify, Airbnb, Pinterest & 5 more

by Lenny Rachitsky

Frequently Asked Questions

How do you choose a startup's first growth channel?
Pick a channel the product can feed, test it with one narrow segment on a set budget and date, and judge it by how many users reach value and come back.
What should a founder check first when growth stalls?
That the data is right, then cohort retention split by segment and channel. If curves trend toward zero in every slice, fix the product before spending more on acquisition.
How do you tell a growth problem from a product problem?
Bring in a small batch of users from one clean channel and read their cohorts. If they activate and return but there aren't many, it's distribution. If they leave within the first week, product owns the fix.
What moves activation in a new user's first few days?
Find where users drop, then match the lever: pre-answer setup for users who stall, make the first action an earned win for users who feel nothing, and bring back users who got value.
Who should own acquisition and retention, growth or marketing?
Most members would split near signup, with marketing on channels and the promise and growth on the path to first value. The line still has to follow the growth model, and both teams judge channels by the cohorts that stay.
Should a growth team report to product or marketing?
Wherever the biggest lever is. For most product-led companies that is product, next to core rather than inside it, and once paid spend matters too, one growth leader owns acquisition and activation.
How do you bring back signups who went quiet?
Watch a few try the next step and fix where they stall, then test the message with and without the fix against a holdout, and count retained use rather than opens.
When should a product-led company add sales?
A founder can start working accounts where usage is spreading and retaining. A sales team follows once deals need security reviews and procurement that self-serve can't handle.
Which self-serve accounts should a sales rep call?
Ones big enough to pay for a rep's time, with use spreading inside the company and a need self-serve can't meet. Hold some flagged accounts back to check that sales added value.
Which early action should count as activation?
Use two measures: an early action onboarding can move as the team's operating metric, and a later repeat action to check that those users came back. Randomize the onboarding change so motivated users don't skew the read.
What makes a good north star metric for an early product?
A retention measure built from how often the problem occurs, the action that solves it and the user it describes, with activation as an input. Drop it if the team can raise it without users getting more value.
Should a new marketplace build supply or demand first?
Whichever side gets a buyer to a completed transaction while giving the supplier a reason to stay. Start in one narrow market and track matches, supplier earnings and repeat buyers.
When can a marketplace copy its first city?
When service holds after subsidies come down, a new manager can run the launch playbook without the founders, and the team knows whether the channel behind market one's demand starts warm or cold in the next market.
How should you test a landing page?
Show visitors what they searched for before the ask, start with the page's first clear funnel drop, and count activated users per visitor alongside signups.
Should a startup with a few dozen users stop acquisition?
Stop broad acquisition but keep a small hand-picked flow from one segment, onboard those users live, and count only return visits nobody prompted.
How does a B2B startup find its first paying customers?
Have founders sell by hand to one narrow buyer, starting with warm intros and asking for payment early. Then check who actually paid, and at what price, before choosing a channel to scale.

All startup growth questions →

Frequently Asked Questions

Ranking lifecycle emails by where users drop off

Growth

Frequently Asked Questions

How do you choose a startup's first growth channel?

What should a founder check first when growth stalls?

How do you tell a growth problem from a product problem?

What moves activation in a new user's first few days?

Who should own acquisition and retention, growth or marketing?

Should a growth team report to product or marketing?

How do you bring back signups who went quiet?

When should a product-led company add sales?

Which self-serve accounts should a sales rep call?

Which early action should count as activation?

What makes a good north star metric for an early product?

Should a new marketplace build supply or demand first?

When can a marketplace copy its first city?

How should you test a landing page?

Should a startup with a few dozen users stop acquisition?

How does a B2B startup find its first paying customers?

Your ideas, super powered.

Ranking lifecycle emails by where users drop off

Growth

Frequently Asked Questions

How do you choose a startup's first growth channel?

What should a founder check first when growth stalls?

How do you tell a growth problem from a product problem?

What moves activation in a new user's first few days?

Who should own acquisition and retention, growth or marketing?

Should a growth team report to product or marketing?

How do you bring back signups who went quiet?

When should a product-led company add sales?

Which self-serve accounts should a sales rep call?

Which early action should count as activation?

What makes a good north star metric for an early product?

Should a new marketplace build supply or demand first?

When can a marketplace copy its first city?

How should you test a landing page?

Should a startup with a few dozen users stop acquisition?

How does a B2B startup find its first paying customers?

Your ideas, super powered.