Activation rate: how to define it, calculate it and judge benchmarks
How to define activation rate, pick the activation event, calculate it per cohort, read the 2022 and 2025 benchmarks and raise it without gaming it.
Activation rate is the share of new users who reach your activation event, the first action that proves value, within a fixed window. Picture a PM whose investor just asked what a good activation rate is, while the dashboard counts anyone who clicked anything. The honest answer starts with the definition, not a benchmark. Fix the event, the base and the window, and the number shows whether new users stall before they see value.
What activation rate measures
Activation rate measures how many new users reach the first moment your product proves its value. Lenny Rachitsky and Yuriy Timen write the formula like this:
activation rate = [users who hit your activation milestone] / [users who completed your signup flow]
Their milestone is the earliest point in onboarding that shows the product's value and predicts long-term retention. Both halves of that definition matter. An event that shows value but does not predict retention is a feature tour. An event that predicts retention but comes late reports problems weeks after they start.
Placement is where most definitions go wrong. "Too early would be simply finishing the sign-up process for a SaaS product," the authors write. Too late would be three purchases on a marketplace or e-commerce site. Their rule of thumb: the milestone "should be a leading indicator of having built a habit, not having proven the habit."
Open your dashboard and write down which event your current activation rate counts. If the answer is sign-up completion, you are measuring a different step of the funnel.
How to calculate activation rate
Calculate activation rate from three choices you fix in writing: the event, the window and the base. Change any one of them, and last quarter's number no longer compares with this quarter's.
| Choice | What to fix | Example |
|---|---|---|
| Event | One action, logged the same way on every platform | Created a first report |
| Window | Days after sign-up in which the event still counts | 7 days |
| Base | Who enters the denominator | Users who completed sign-up |
Time windows are rarer than they should be. In the 2022 survey, about 6% of respondents had time-bound activation, with a median of 10 days and a mode of 7 days. The authors think the real share is higher, because the survey did not ask about windows directly.
Without a window, older cohorts keep activating for months and always look better than last week's. A 7-day window suits a product people open daily. A team tool with a weekly rhythm needs a longer one.
Count per weekly sign-up cohort, not per calendar week of events. Take everyone who completed sign-up in one week, then count who reached the event within the window. Counting events by calendar week mixes old and new users. A re-engagement campaign then shows up as an onboarding win.
The base decides what the rate means. Counting from users who started sign-up mixes sign-up friction into activation, and the same cohort reads lower. Kreto's Masa work deferred registration until users had tried the core experience, and sign-up conversion rose from 32.7% to 63%. A change like that redefines who counts as a new user, so restart the activation baseline from the first cohort after it.
Put the event, window and base into your metrics doc, then rebuild the last eight weekly cohorts on that definition.
How to choose your activation event
Choose the earliest event that predicts retention and that enough new users reach. Rachitsky's process has three steps:
- List candidate moments by brainstorming with the team and exploring product usage data.
- Run a regression analysis to see whether retention inflects when users hit each moment.
- Run experiments that push more users to a moment, and check whether their retention rises.
The third step separates a real activation metric from a lucky correlation. In Rachitsky's words, "A good activation metric is causal for your retention, not just correlative."
For a quick version of the second step, use the threshold from the 2022 survey: "Users who hit your activation milestone should retain at a rate at least 2x better than those who do not complete the activation step."
Run the comparison on a cohort table. For each candidate, split new users into those who did it within the window and those who did not. Divide the first group's week 4 retention by the second group's.
In the illustrative example, sharing a report has the strongest ratio, but only a small minority of users ever reach it. Creating a first report clears the line and covers far more users, so it makes the better event. A rare event describes a small group and gives you nothing to work on for everyone else.
Multi-user B2B products need an account-level event. One person finishing setup does not mean the team adopted the tool. The authors' example of an actionable account milestone is a "workspace with 10 items created and 2+ active editors by W4".
Pull four candidate events from your analytics and compute the ratio for each. Then check the winner against the retention curve of each cohort, not against one blended average.
What is a good activation rate?
A good activation rate is one that rises across your own cohorts while retention holds. Public benchmarks only tell you whether your number sits in a plausible range. The two public sets below measure different things.
| Source | Year | Sample | Average | Median |
|---|---|---|---|---|
| Lenny's Newsletter reader survey, all products | 2022 | 500+ responses | 34% | 25% |
| Same survey, SaaS only | 2022 | SaaS respondents | 36% | 30% |
| Userpilot benchmark report | 2025 | 62 B2B SaaS companies | 37.5% | 37.04% |
The Lenny's Newsletter survey drew over 500 responses across B2B SaaS, B2C, marketplaces, e-commerce and DTC. Userpilot's 2025 report uses first-party data from 62 B2B companies on its Activation Dashboard.
The two sets look close, but they are not the same measurement. They differ in year, in sample and in definition, and in both each team picks its own activation event. The SaaS median of 30% and Userpilot's median of 37.04% sit in a similar range. That is as far as the comparison goes, so neither one is the benchmark, and averaging them means nothing.
Most of the spread comes from what teams count. Asked why rates vary, the survey authors answer: "Mostly due to the milestone definition, and how much effort it takes to get there." Marketplaces and e-commerce report the lowest rates, since their milestone is a first transaction. B2C freemium products report the highest, with low-friction milestones.
Userpilot's industry split shows the same effect inside one dataset, with rates from 5% to 54.8%. The page carries its own warning: "Chasing the numbers of other companies without considering your specific context might not lead you in the right direction for your own success."
A FinTech product and an AI tool with the same rate stand in very different places against their peers. Compare yourself with products that share your category and your kind of milestone, and only loosely.
When an investor asks, give your definition first, then your number, then the trend across your last eight cohorts.
How to raise activation rate
Raise activation by fixing the one step where most new users stop between sign-up and the event. The survey respondents named eight common changes, and they fall into three places in the funnel.
| Where the change happens | Changes respondents named |
|---|---|
| Before sign-up | Smarter top-of-funnel targeting |
| Inside the product | Simpler onboarding UI and UX, less onboarding friction, better copy, showing value earlier |
| Outside the product | Emails and follow-up messages, sales outreach, incentives |
Treat the list as a menu, not a plan. Work through it in this order:
- Map every step between completed sign-up and the activation event, and count users at each one for a single cohort.
- Pick the step that loses the most users. Watch session recordings of people who stopped there, and talk to five of them.
- Change that one step only. The guide to SaaS onboarding best practices covers the usual fixes for a single drop-off point.
- Compare the next cohorts with the baseline on the same event, window and base.
When users leave the product before the event, an onboarding email sequence with one task per email gives them a reason to return to it. Kreto's product marketing work covers this stage when a funnel's bottleneck sits between sign-up and first value.
Watch for changes that move the number without moving the product. Picking an earlier event raises the rate overnight. Narrowing targeting raises it by sending fewer, better-fit users. Neither proves onboarding improved.
Check that retention moved for the same cohorts. An activation gain that leaves retention flat means the event was correlated, not causal. Start with step one on last week's cohort.
Common activation metric mistakes
Broken activation rates fail in a handful of predictable ways. The survey names the first five. The sixth is a reporting fault that is easy to miss.
- Too early. Sign-up completion or a filled profile counts as activation. Nearly everyone reaches it, so it predicts nothing.
- Too late. The event proves a habit instead of predicting one, such as a third purchase. Problems surface weeks after they start.
- Not predictive. Users who reach the event retain about as well as users who skip it.
- Not actionable. No onboarding change can move the event, so the team watches it without acting.
- Too complicated. The rule stacks so many conditions that nobody on the team can explain or debug it.
- Redefined without a reset. The event, window or base changed, and the chart still joins old and new numbers in one line.
Check your current definition against all six before the next board or investor update.
Frequently asked questions
What is the difference between activation rate and sign-up conversion?
Sign-up conversion counts visitors or installs who finish creating an account. Activation rate starts where sign-up ends. It counts users who completed sign-up and then reached the activation event within the window. Track both, because a change to the sign-up flow shifts who enters the activation base.
How is activation rate different from retention?
Activation is a leading indicator, measured in the first days or weeks. Retention measures whether users are still active weeks or months later. A good activation event predicts retention, so validate it against your cohort retention. Users who activate should retain clearly better than users who do not.
How long should the activation window be?
Match the window to how often the product gets used. Check the window against your data: plot when activators reach the event, and set the window where most of them already have. A team tool where adoption depends on colleagues needs longer, closer to the four-week account milestone described above. Pick one window, write it down and keep it fixed across cohorts.
What if we have too little data to test events?
With a few hundred sign-ups a week, the retention ratio for rare events is noisy. Pool several weekly cohorts before comparing candidates, and favor events that many users reach. Pair the numbers with interviews of users who stayed and users who left. Revisit the choice when volume grows.
Find where new users stall
You now have a candidate event, a fixed window and a weekly cohort table. If most users drop between completed sign-up and that event, onboarding is the stage to look at first. Talk to Kreto if you want a second view on that stage.