Which activation lever to pull in a new user's first days

Members of the Growth work group argue over cutting steps versus adding them, then Brian Balfour maps each lever to the moment users drop.

Updated October 13, 2026

Find where new users drop before picking a lever: pre-answer setup for users who stall, make the first real action a win they earn, and give users who got value a reason to return.

Count the decisions a new user faces. Casey Winters says removing things moves activation more than adding them: Pinterest's activation experience hid advanced features until people understood the core value. Lauryn Motamedi partly disagrees, since a new guided wizard at Airtable raised activation by 20%. Brian Balfour reads both as cutting the decisions a new user makes before value, and would count the open-ended decisions between signup and first output, then remove or pre-answer each one. Casey agrees with that reading.
Ask for information before the aha, commitment after it. Brian would keep setup questions that route a user to a use case, because they tell the product what to show, and save asks like downloading the app for after the value lands. Casey describes Pinterest's early emails changing from feature lessons to delivering the value the first session promised, and activation going up.
Let the user take the first action and win it. Hila Qu points to Jira, which pre-created three tickets for new users and moved neither retention nor conversion. She would make the real action feel small instead, as Acorns did by framing savings as $5 a day rather than $150 a month. Albert Cheng would make the first action an early win: Chess.com built Play Coach, a practice mode, so beginners are not crushed in their first rated games.
Set the window by how often the use case recurs. Brian would judge a daily product in week one and give a weekly or monthly one longer, then check that activated users retain better than the rest. Hila would also give the team a leading indicator inside the first few days, validated against 30 to 90 day retention.
Fix the earliest broken moment first. Brian maps the levers: users who stall in setup need pre-answered decisions, users who produce output and feel nothing need an earned win, and users who got value but don't return need emails, a metagame or, on horizontal products, a second use case. For B2B, Hila adds a row for the team and the buyer, so the first output should be easy to share.

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

Question:

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

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

The biggest lever is getting the user to their first habit-forming use case. Shaving steps off signup comes later. When I wrote the Reforge Retention + Engagement program with Casey and Shaun Clowes, we split activation into three moments: setup, aha, and habit. Most teams over-invest in setup because it's the easiest to measure.

Two things decide what works in the first few days.

1️⃣ The type of habit you're asking for. Workout, diet, and to-do apps have low activation rates because they ask users to build a new behavior from scratch. A product that slots into an existing habit activates much more easily. The habit type tells you which metric to watch and how much hand-holding the first days need.

churn

2️⃣ Good friction. Horizontal products like Loom, Notion, and Airtable, and most AI tools, give users a novel first experience with no clear reason to come back. A few setup questions that route someone to a specific use case are worth the drop-off they cause, because that information lets you deliver the habit-forming experience. A first use case can also fail to stick, so the next job is getting them to a second one.

Laura Schaffer's Amplitude case is the best concrete example I know. Her team cut setup from five steps to fewer and made templates prescriptive. Chart saves got people engaged early. The templates were sticky enough that they later shaped the homepage, and higher activation fed directly into free-to-paid conversion and quarter-over-quarter self-serve acceleration.

YouTube

On execution, about 80% of growth experiments fail. Pick one activation moment and sequence your work against it. Spreading experiments across the whole funnel means none of them move the metric.

YouTube

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

Removing things moves activation more than adding them.

Most teams build onboarding for the users they already have. New users are still trying to learn the basics, and the features you built for power users make those basics harder to find. At Pinterest we doubled activation rate, and a big part of that was an activation experience that stripped out advanced features until people understood the core value: interesting content across several topics they care about, beyond the one topic they searched on Google.

Medium

The second lever is the first few days after the session ends. Our early new-user emails at Pinterest taught people about features. Trevor Pels changed them to deliver the value we promised in the first experience, and activation went up. That is the timing behind Brian's good friction: ask for more once the value is proven. We added friction on purpose only after the value was delivered. Web signups who clicked that content on mobile got asked to download the app, and they did because they already trusted the value. An email that only asked them to get the app drove more unsubscribes than downloads.

caseyaccidental

On Brian's habit moment, be careful about judging it only in the first week. One company I worked with found that many users who skipped the core action in week one still came back later, and extending the window to 35 days showed much more clearly who had actually formed a habit.

There is also a failure mode. When we added a new value proposition to one company's onboarding flow, metrics dropped because we showed new users every part of the program at once when only some of it was relevant to them.

caseyaccidental

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

Casey's timing point and my good friction point fit together once you sort friction by what it buys. Friction that collects information you need to deliver the first value belongs before the aha moment. Friction that asks for commitment, like downloading the app, belongs after it, which is what Pinterest did with the mobile prompt.

Horizontal and AI products make the distinction matter. A blank prompt box has no setup friction, but the user has to invent their own use case, and most of them land on something novel that never becomes a habit. Two or three questions about role or job-to-be-done cost some drop-off. They let you pick the use case, the template, and the follow-up email content, which is the information Casey's emails needed to deliver value instead of feature tours.

I agree with the 35-day window, and the habit type sets the window. A product with a daily natural frequency can be judged in week one. A weekly or monthly use case, like a reporting tool or a planning ritual, will look like it's failing in week one when users are on schedule. Set the activation window by how often the use case actually recurs, then check that the activated cohort retains better than the rest. If it doesn't, you picked the wrong activation metric, and every onboarding experiment built on it is optimizing the wrong thing.

churn

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

I partly disagree with Casey's "removing beats adding." At Airtable, the change that moved activation was a new step. Our rebuilt onboarding raised activation rate by 20%

lennysnewsletter, and the main reason was a guided, immersive wizard. I'm usually skeptical of wizards. Airtable users had a specific problem, though: building the basic scaffolding of a base took too much effort, and they needed guardrails to get through it. The wizard worked because Airtable is a complex product where figuring out how to start has a high cognitive load

YouTube. So the useful test is how hard the first build is. If a user can't make something real on their own in the first session, add structure. If they can, cut steps the way Casey did at Pinterest.

I agree with Casey on what to cut. Tooltips like "this is automations" announce a feature the team is proud of, and they don't tell a new user why it matters for their use case

YouTube.

The other big lever is the activation metric itself, because it decides which of Brian's moments you can even see. Before we rebuilt onboarding, we broke our metric, week-4 multi-user active, into its parts:

Team vs. individual signups.
Accounts with more than two people invited.
Accounts with more than two people ever active.
Accounts with more than two people active in week 4.

We also tracked individual week-2 and week-4 retention, plus a "Build" sophistication score. That breakdown is what showed us which levers existed

YouTube. I find broad definitions work better than a narrow checklist of five to seven actions, because a narrow checklist measures what the company thinks value is and return visits measure whether the user found it. My targets: roughly 20% of signups should hit the metric, and it has to correlate with six-month retention

YouTube. Brian covered the window. What I'd add is that, for a collaborative product, the multi-user parts of the breakdown usually show the biggest gap, so look there first.

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

Brian's setup/aha/habit split and Casey's point about subtracting both hold up in my experience. The failure I'd flag is teams hitting the activation metric without the user ever getting the value.

My working formula for the aha moment is "someone does X action Y times in the first Z days." Facebook's 10 friends in 7 days and Slack's team sending 2,000 messages in 30 days are the classic examples. The metric is only a proxy for the moment the user first feels the value, and teams forget that. Jira identified "create 3 tickets" as its aha moment and pre-created 3 tickets for new users. Neither retention nor conversion moved, because the user never did anything or felt anything. You can guide users and remove friction, but they have to take the action themselves.

So the lever I'd pull is making the real action feel small, and I'd leave the action itself alone. At Acorns we worked with Shlomo Benartzi (behavioral economist) on reframing the savings ask: $5 a day instead of $150 a month. The money was the same, but the commitment felt easy, and the reframe eliminated the income gap in savings behavior.

Brian's point about setting the window by natural frequency is right for the habit metric. I'd still give the team a separate leading indicator inside the first few days that it can run experiments against every week, for example the first real deposit at Acorns. Validate that indicator against 30 to 90 day retention, then optimize the % of new users who reach it.

That combination of lifecycle email, in-app messaging, and product tests focused on one key action is how we tripled new user key feature adoption in the first 90 days at Acorns and lifted 3-month retention by 10%.

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

Lauryn's Airtable wizard and Casey's Pinterest cuts are the same move. Both reduced the number of decisions a new user has to make before reaching value. Step count is a poor proxy for that. A wizard adds screens and removes the hardest decision on a horizontal product, which is what to build first. Hiding power-user features removes choices the user can't evaluate yet.

Laura's Amplitude rebuild did both at once. Her team cut setup from five steps to fewer and also added prescriptive templates, and the templates turned out to be the stickiest part. They later shaped the homepage.

YouTube

So the test I'd give a team is to count the open-ended decisions between signup and the first real output, then remove or pre-answer each one. Pre-answering usually means a template, a default, or a short wizard driven by the two or three setup questions. Hila's Jira example marks the limit. A template pre-answers what to build, and the user still builds it. Pre-creating three tickets did the action itself, so nobody felt anything. Teams that only count steps make the opposite mistake and delete the questions that would have told them which template to show.

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

An early win in the core loop moves activation more than anything else. In the first few days, a new user decides whether they're good at your product. A user who feels clumsy or beaten in session one rarely gives you a session three.

♟️ At Chess.com, the risk for a beginner is getting crushed in their first rated games. So we built Play Coach, a practice mode where beginners learn fundamentals in a non-competitive setting, with coaching voiced in the style of Magnus Carlsen, Hikaru Nakamura, and Levy Rozman.

LinkedIn At a gaming panel I spoke on last year, one of the takeaways the organizer shared was "people like to win," so they don't always want AI as a strong opponent. Play Coach fits Lauryn's test: for a beginner, playing a full game alone is the hard first build, so we added structure. And per Hila's Jira story, the player still makes every move themselves, so the win is real.

🚀 The second lever is showing the full product during the first days, before a user has to pay or commit. At Grammarly, showing premium features to free users doubled upgrades to paid plans. The same thinking is why I prefer reverse trials to time-based trials: the user experiences the best version of the product while they're still forming habits.

LinkedIn This can look like it conflicts with Casey's point about stripping out advanced features. I think both hold: the premium value should show up inside the one core action the user is already doing, like a better suggestion on the sentence they're writing in Grammarly, and stay out of a feature tour.

Third, I put a metagame and a profile on top of the core loop early. Those are two of my three pillars of gamification, along with the core loop itself. At Duolingo, streaks and XP gave day-two users a reason to come back on top of the lesson itself. That works because language learning and chess both have a daily natural frequency, which fits Brian's point that a daily product can be judged in week one.

The order matters. A metagame bolted onto a core loop where users are losing gives them a reason to return to something they already find frustrating, so I fix the early win first.

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

Brian's decision count settles the Airtable question for me. I also argued that an added step raises activation when users see why you're asking.

Medium Brian's setup questions are one way to learn what a user wants. The channel they came from is often a cheaper way, and teams ignore it.

At Pinterest, a user arriving from Google had already told us their topic through the search. Our best results came from sending first-time visitors straight to relevant content and holding back feature education. We also learned that delaying an aggressive signup prompt let people explore and build intent, and that tailoring landing experiences by category matched intent better than one generic flow. So you can sometimes skip questions you already know the answer to. Use them to cover what the channel doesn't tell you. For an SEO visitor, that was the second and third topic, because one topic from Google was rarely enough to form a habit.

Channel also explains why activation can drop when growth seems to be working. When a growth team opens a new channel or segment, like organic search, paid social, or an upmarket push, signups go up first and every metric below signup falls. Teams usually blame the onboarding experiment. Most of the time the new cohort just has different intent, and the flow was built for the old one. Segment activation by channel before you change anything.

One more point on Brian's 80% failure rate. Run activation experiments to answer a question, rather than to hunt for wins. The Pinterest email and phone activation tests that failed taught us more about where the real bottleneck was than the 50% signup conversion win did.

reforge

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

Every lever in this thread works, and each one fixes a different moment. That's why the first job is finding where your users drop, before picking a lever. Teams that try the Pinterest cuts, the Airtable wizard, Albert's early wins, and Duolingo streaks all in the same quarter spread experiments thin, and with roughly 80% of growth experiments failing, none of them move the metric.

YouTube

Casey's channel point is step zero of that diagnosis. Cut the drop-off by acquisition channel before mapping it to a moment, because a new channel can make setup look broken when the cohort just arrived with different intent.

Then map the levers to the moments:

1️⃣ Users stall in setup and never produce a first output. Pull Lauryn's lever: pre-answer the hard decisions with a wizard, template, or setup questions.

2️⃣ Users produce output and still don't feel anything. Pull Albert's and Hila's levers: make sure the first real action ends in a win the user earned themselves. Albert's reverse trial fits here too, because it puts premium value inside that first action.

3️⃣ Users get the aha and don't return. Pull Casey's post-session emails that deliver the promised value, Albert's metagame, and a push toward a second use case on horizontal products.

Albert's ordering rule applies to all three. A metagame on top of a loop where users lose, or emails pointing at a product they couldn't set up, amplifies the failure. So I'd fix the earliest broken moment first and move to the next one only once it holds.

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

Brian's map covers one user's journey. B2B PLG needs a fourth row, because the first few days also decide whether a second person ever shows up, and that second person is usually the one who pays. Users are cheap to acquire with B2C tactics, but companies are what convert to paid, so free-to-paid conversion is where a user metric turns into a customer metric.

So I'd map an aha moment per persona: the individual user, the team, and the buyer.

LinkedIn Lauryn's Airtable breakdown shows the same thing from the metric side. The biggest gap sat in the multi-user steps: invites, then more than two people active, then more than two still active in week 4.

That changes what to build in days 1 to 7. The individual's first output should be easy to share and should give the person receiving it a reason to act, such as a report to comment on or a merge request to review. When the first output lives in a private sandbox, the team and buyer aha moments never get a chance to happen.

For complex products, I also treat onboarding as continuous after week one. Each new feature gets its own mini aha moment, delivered on an engagement ladder, because a day 1 tour of everything overwhelms new users, which is the same failure Casey described.

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.
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.
Which lifecycle emails should a team build first?
The ones aimed at the steepest drop on the retention curve, usually the activation email. Depth and renewal emails come next, and win-back to fully churned users comes last.
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

Which activation lever to pull in a new user's first days

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?

Who should own acquisition and retention, growth or marketing?

Should a growth team report to product or marketing?

Which lifecycle emails should a team build first?

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.

Which activation lever to pull in a new user's first days

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?

Who should own acquisition and retention, growth or marketing?

Should a growth team report to product or marketing?

Which lifecycle emails should a team build first?

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.