Sep 10, 2026
By Victor Teran
What Is a Good Activation Rate? The Median Is 38%
Median activation for B2B SaaS is 38%, and the top quartile is 61%. We moved a fintech product from 38% to 67% in one engagement. It took three weeks, no new features, and one uncomfortable conversation about what activation even meant.
Activation
Benchmarks
Onboarding
Most founders have never seen their activation rate next to anyone else's, so they cannot tell whether theirs is a crisis or a Tuesday. That uncertainty is expensive, because it decides whether you spend the next quarter on acquisition or on the product you already have.
Here is the yardstick. Perspective AI's 2026 onboarding benchmark, drawn from roughly 1,400 product organisations, puts median B2B SaaS activation at 38% and the top quartile at 61%. Fintech runs higher, 44% median and 68% top quartile. Activation there means the share of new signups reaching the product's defined value moment inside seven or fourteen days (Perspective AI, May 2026).
Now the number that matters to us. Q.AI, a fintech product, came to us at 38%. Against its own category median of 44%, it was below average. One engagement later it was at 67%, effectively top quartile. Nothing was added to the product.
Key takeaways
Median B2B SaaS activation is 38% and top quartile is 61%; for fintech it is 44% and 68%. If you do not know which side of those numbers you are on, that is the first problem.
Q.AI moved from 38% to 67%, from below its category median to one point off top quartile, in a single engagement with no new features shipped.
The move was not a redesign. It was defining the value moment, instrumenting it, and then changing only the screens the data implicated.
A benchmark is only useful once your activation event is defined. Most products we audit are measuring signup completion and calling it activation.
What is a good activation rate?
Anything at or above 38% is median for B2B SaaS, and 61% puts you in the top quartile. For fintech the bar is higher: 44% and 68%.
Segment | Median | Top quartile |
|---|---|---|
B2B SaaS | 38% | 61% |
Fintech | 44% | 68% |
Source: Perspective AI's 2026 onboarding benchmark across roughly 1,400 product organisations, measuring the share of new signups who reach the product's defined value moment inside seven or fourteen days.
Two conditions before you use those numbers. Your activation event has to be a real value moment rather than a setup step, and your window has to be fixed in advance. Change either one and the comparison is decoration.
How do you calculate activation rate?
Activated users divided by new signups in the same cohort, inside a fixed window.
Take everyone who signed up in a given week, count how many reached your value moment within seven or fourteen days, and divide. Cohort in, cohort out. A blended all-time number moves too slowly to tell you whether last month's change worked, and it hides the only comparison that matters, which is this month's new users against last month's.
What does a 38% activation rate actually mean?
It means 62 of every 100 people who signed up never reached the thing your product is for.
Read that as a cost rather than a percentage. If you are acquiring 500 signups a month at a blended $40, that is $20,000 spent, and $12,400 of it bought people who never got to the value moment. They are not undecided. Most of them are gone, and the benchmark data puts the window at seven to fourteen days.
This is why the acquisition-first instinct is usually wrong at this stage. Doubling signups at 38% activation doubles the wasted half too. The same spend against 67% buys nearly twice the activated users without a single new visitor.
That is the whole argument for auditing before scaling, and it is arithmetic rather than opinion.
Why was Q.AI below its own category median?
Because activation had never been defined, so nothing in the product was built toward it.
Q.AI had a growing user base and a retention curve nobody could explain. The team had theories. Theories are what fill the space where a definition should be.
The first weeks of the engagement contained no design work at all. We built a funnel across the new-user path in Amplitude: signup, the onboarding steps, first core action, return session. The funnel showed the cliff. Then session replays, filtered to new users who reached that step and left, showed why.
The stumbles were specific. Effort demanded before value was shown. Screens where the next step was not obvious. A sequence that made sense to the team and not to a stranger.
None of that is visible in an activation percentage. The percentage tells you that you have a problem. It never tells you where.
FREE GUIDE
The 4 product leaks costing you growth
A short audit guide for founders. Find the four places your product leaks revenue, and what to fix first.
What changed to get to 67%?
Only the screens the evidence implicated, redesigned to remove the observed stumbles.
Value earlier. Fewer decisions. The next action as the loudest thing on the screen. Unglamorous, and narrow on purpose.
Because the changes were narrow, attribution stayed clean. The rest of the product was held constant, so when the number moved we knew what moved it. Sweeping redesigns destroy that. You ship forty changes, the number moves, and you have learned nothing you can reuse.
The results: activation 38% to 67%, a 29 point rise. Onboarding completion up 68%. Support tickets down 73%, which surprised the team and should not have, because confused users file tickets and the confusion had been removed at its source. Q.AI kept growing and was later acquired by Forbes.
We have written the full sequence up as the anatomy of that engagement, because the method generalises even though the percentages do not.
How do you compare yourself to a benchmark honestly?
Match the definition before you compare the number, or you will benchmark two different things.
The benchmark defines activation as reaching a defined value moment within seven or fourteen days. Most products we audit are not measuring that. They are measuring signup completion, or email verification, or profile setup, and calling the result activation.
Those numbers look far healthier and mean nothing. A product with 80% "activation" measured at account creation is not beating the benchmark. It is not measuring activation at all. The working definition worth holding to is the point a new user first experiences the product's core value, not the point they finished configuring it (Amplitude).
What you might be measuring | What the benchmark measures |
|---|---|
Signup or account created | The defined value moment reached |
Email verified | Within a fixed 7 or 14 day window |
Profile or onboarding steps completed | Behaviour that separates retained users from churned ones |
Any first session activity | One event, chosen deliberately, instrumented on purpose |
If you cannot say in one sentence what your activation event is, the comparison is not available to you yet. That is fixable in an afternoon, and we have written how to define yours.
Where should you look first?
At the earliest step where a cohort of new users stops, not at the step you like least.
Founders usually know which screen embarrasses them. It is rarely the one costing the money. The screen costing the money is boring, early, and asks for something the user is not ready to give.
The order that has worked in every engagement since:
Define the value moment in one sentence, and get the team to agree on it.
Instrument a funnel across the new-user path to that moment.
Watch replays of the users who dropped at the worst step, until the reason is obvious rather than theorised.
Change only what the evidence implicates.
Measure on the same funnel, new cohorts against old, with the window agreed in advance.
You will notice there is no step where you rebuild the product. In the engagements we run, the biggest single lever has sat in the first minutes of the experience nearly every time, and the fix has almost never required new functionality.
a16z's growth metrics guide segments its own benchmarks by revenue scale, software type and go-to-market motion (a16z), for the same reason: a benchmark you cannot match to your own situation is decoration.
It also decides what your advertising is worth, since paid acquisition against a leaky product buys the same leak at a larger scale.
The benchmark is useful for one thing: telling you whether the gap between you and top quartile is worth three weeks of attention. At 38%, with an acquisition budget already running, it is not close.
WHAT NEXT
Want this fixed in your product, not just explained?
Sep 10, 2026
By Victor Teran
What Is a Good Activation Rate? The Median Is 38%
Median activation for B2B SaaS is 38%, and the top quartile is 61%. We moved a fintech product from 38% to 67% in one engagement. It took three weeks, no new features, and one uncomfortable conversation about what activation even meant.
Activation
Benchmarks
Onboarding
Most founders have never seen their activation rate next to anyone else's, so they cannot tell whether theirs is a crisis or a Tuesday. That uncertainty is expensive, because it decides whether you spend the next quarter on acquisition or on the product you already have.
Here is the yardstick. Perspective AI's 2026 onboarding benchmark, drawn from roughly 1,400 product organisations, puts median B2B SaaS activation at 38% and the top quartile at 61%. Fintech runs higher, 44% median and 68% top quartile. Activation there means the share of new signups reaching the product's defined value moment inside seven or fourteen days (Perspective AI, May 2026).
Now the number that matters to us. Q.AI, a fintech product, came to us at 38%. Against its own category median of 44%, it was below average. One engagement later it was at 67%, effectively top quartile. Nothing was added to the product.
Key takeaways
Median B2B SaaS activation is 38% and top quartile is 61%; for fintech it is 44% and 68%. If you do not know which side of those numbers you are on, that is the first problem.
Q.AI moved from 38% to 67%, from below its category median to one point off top quartile, in a single engagement with no new features shipped.
The move was not a redesign. It was defining the value moment, instrumenting it, and then changing only the screens the data implicated.
A benchmark is only useful once your activation event is defined. Most products we audit are measuring signup completion and calling it activation.
What is a good activation rate?
Anything at or above 38% is median for B2B SaaS, and 61% puts you in the top quartile. For fintech the bar is higher: 44% and 68%.
Segment | Median | Top quartile |
|---|---|---|
B2B SaaS | 38% | 61% |
Fintech | 44% | 68% |
Source: Perspective AI's 2026 onboarding benchmark across roughly 1,400 product organisations, measuring the share of new signups who reach the product's defined value moment inside seven or fourteen days.
Two conditions before you use those numbers. Your activation event has to be a real value moment rather than a setup step, and your window has to be fixed in advance. Change either one and the comparison is decoration.
How do you calculate activation rate?
Activated users divided by new signups in the same cohort, inside a fixed window.
Take everyone who signed up in a given week, count how many reached your value moment within seven or fourteen days, and divide. Cohort in, cohort out. A blended all-time number moves too slowly to tell you whether last month's change worked, and it hides the only comparison that matters, which is this month's new users against last month's.
What does a 38% activation rate actually mean?
It means 62 of every 100 people who signed up never reached the thing your product is for.
Read that as a cost rather than a percentage. If you are acquiring 500 signups a month at a blended $40, that is $20,000 spent, and $12,400 of it bought people who never got to the value moment. They are not undecided. Most of them are gone, and the benchmark data puts the window at seven to fourteen days.
This is why the acquisition-first instinct is usually wrong at this stage. Doubling signups at 38% activation doubles the wasted half too. The same spend against 67% buys nearly twice the activated users without a single new visitor.
That is the whole argument for auditing before scaling, and it is arithmetic rather than opinion.
Why was Q.AI below its own category median?
Because activation had never been defined, so nothing in the product was built toward it.
Q.AI had a growing user base and a retention curve nobody could explain. The team had theories. Theories are what fill the space where a definition should be.
The first weeks of the engagement contained no design work at all. We built a funnel across the new-user path in Amplitude: signup, the onboarding steps, first core action, return session. The funnel showed the cliff. Then session replays, filtered to new users who reached that step and left, showed why.
The stumbles were specific. Effort demanded before value was shown. Screens where the next step was not obvious. A sequence that made sense to the team and not to a stranger.
None of that is visible in an activation percentage. The percentage tells you that you have a problem. It never tells you where.
FREE GUIDE
The 4 product leaks costing you growth
A short audit guide for founders. Find the four places your product leaks revenue, and what to fix first.
What changed to get to 67%?
Only the screens the evidence implicated, redesigned to remove the observed stumbles.
Value earlier. Fewer decisions. The next action as the loudest thing on the screen. Unglamorous, and narrow on purpose.
Because the changes were narrow, attribution stayed clean. The rest of the product was held constant, so when the number moved we knew what moved it. Sweeping redesigns destroy that. You ship forty changes, the number moves, and you have learned nothing you can reuse.
The results: activation 38% to 67%, a 29 point rise. Onboarding completion up 68%. Support tickets down 73%, which surprised the team and should not have, because confused users file tickets and the confusion had been removed at its source. Q.AI kept growing and was later acquired by Forbes.
We have written the full sequence up as the anatomy of that engagement, because the method generalises even though the percentages do not.
How do you compare yourself to a benchmark honestly?
Match the definition before you compare the number, or you will benchmark two different things.
The benchmark defines activation as reaching a defined value moment within seven or fourteen days. Most products we audit are not measuring that. They are measuring signup completion, or email verification, or profile setup, and calling the result activation.
Those numbers look far healthier and mean nothing. A product with 80% "activation" measured at account creation is not beating the benchmark. It is not measuring activation at all. The working definition worth holding to is the point a new user first experiences the product's core value, not the point they finished configuring it (Amplitude).
What you might be measuring | What the benchmark measures |
|---|---|
Signup or account created | The defined value moment reached |
Email verified | Within a fixed 7 or 14 day window |
Profile or onboarding steps completed | Behaviour that separates retained users from churned ones |
Any first session activity | One event, chosen deliberately, instrumented on purpose |
If you cannot say in one sentence what your activation event is, the comparison is not available to you yet. That is fixable in an afternoon, and we have written how to define yours.
Where should you look first?
At the earliest step where a cohort of new users stops, not at the step you like least.
Founders usually know which screen embarrasses them. It is rarely the one costing the money. The screen costing the money is boring, early, and asks for something the user is not ready to give.
The order that has worked in every engagement since:
Define the value moment in one sentence, and get the team to agree on it.
Instrument a funnel across the new-user path to that moment.
Watch replays of the users who dropped at the worst step, until the reason is obvious rather than theorised.
Change only what the evidence implicates.
Measure on the same funnel, new cohorts against old, with the window agreed in advance.
You will notice there is no step where you rebuild the product. In the engagements we run, the biggest single lever has sat in the first minutes of the experience nearly every time, and the fix has almost never required new functionality.
a16z's growth metrics guide segments its own benchmarks by revenue scale, software type and go-to-market motion (a16z), for the same reason: a benchmark you cannot match to your own situation is decoration.
It also decides what your advertising is worth, since paid acquisition against a leaky product buys the same leak at a larger scale.
The benchmark is useful for one thing: telling you whether the gap between you and top quartile is worth three weeks of attention. At 38%, with an acquisition budget already running, it is not close.
WHAT NEXT
Want this fixed in your product, not just explained?
Sep 10, 2026
By Victor Teran
What Is a Good Activation Rate? The Median Is 38%
Median activation for B2B SaaS is 38%, and the top quartile is 61%. We moved a fintech product from 38% to 67% in one engagement. It took three weeks, no new features, and one uncomfortable conversation about what activation even meant.
Activation
Benchmarks
Onboarding
Most founders have never seen their activation rate next to anyone else's, so they cannot tell whether theirs is a crisis or a Tuesday. That uncertainty is expensive, because it decides whether you spend the next quarter on acquisition or on the product you already have.
Here is the yardstick. Perspective AI's 2026 onboarding benchmark, drawn from roughly 1,400 product organisations, puts median B2B SaaS activation at 38% and the top quartile at 61%. Fintech runs higher, 44% median and 68% top quartile. Activation there means the share of new signups reaching the product's defined value moment inside seven or fourteen days (Perspective AI, May 2026).
Now the number that matters to us. Q.AI, a fintech product, came to us at 38%. Against its own category median of 44%, it was below average. One engagement later it was at 67%, effectively top quartile. Nothing was added to the product.
Key takeaways
Median B2B SaaS activation is 38% and top quartile is 61%; for fintech it is 44% and 68%. If you do not know which side of those numbers you are on, that is the first problem.
Q.AI moved from 38% to 67%, from below its category median to one point off top quartile, in a single engagement with no new features shipped.
The move was not a redesign. It was defining the value moment, instrumenting it, and then changing only the screens the data implicated.
A benchmark is only useful once your activation event is defined. Most products we audit are measuring signup completion and calling it activation.
What is a good activation rate?
Anything at or above 38% is median for B2B SaaS, and 61% puts you in the top quartile. For fintech the bar is higher: 44% and 68%.
Segment | Median | Top quartile |
|---|---|---|
B2B SaaS | 38% | 61% |
Fintech | 44% | 68% |
Source: Perspective AI's 2026 onboarding benchmark across roughly 1,400 product organisations, measuring the share of new signups who reach the product's defined value moment inside seven or fourteen days.
Two conditions before you use those numbers. Your activation event has to be a real value moment rather than a setup step, and your window has to be fixed in advance. Change either one and the comparison is decoration.
How do you calculate activation rate?
Activated users divided by new signups in the same cohort, inside a fixed window.
Take everyone who signed up in a given week, count how many reached your value moment within seven or fourteen days, and divide. Cohort in, cohort out. A blended all-time number moves too slowly to tell you whether last month's change worked, and it hides the only comparison that matters, which is this month's new users against last month's.
What does a 38% activation rate actually mean?
It means 62 of every 100 people who signed up never reached the thing your product is for.
Read that as a cost rather than a percentage. If you are acquiring 500 signups a month at a blended $40, that is $20,000 spent, and $12,400 of it bought people who never got to the value moment. They are not undecided. Most of them are gone, and the benchmark data puts the window at seven to fourteen days.
This is why the acquisition-first instinct is usually wrong at this stage. Doubling signups at 38% activation doubles the wasted half too. The same spend against 67% buys nearly twice the activated users without a single new visitor.
That is the whole argument for auditing before scaling, and it is arithmetic rather than opinion.
Why was Q.AI below its own category median?
Because activation had never been defined, so nothing in the product was built toward it.
Q.AI had a growing user base and a retention curve nobody could explain. The team had theories. Theories are what fill the space where a definition should be.
The first weeks of the engagement contained no design work at all. We built a funnel across the new-user path in Amplitude: signup, the onboarding steps, first core action, return session. The funnel showed the cliff. Then session replays, filtered to new users who reached that step and left, showed why.
The stumbles were specific. Effort demanded before value was shown. Screens where the next step was not obvious. A sequence that made sense to the team and not to a stranger.
None of that is visible in an activation percentage. The percentage tells you that you have a problem. It never tells you where.
FREE GUIDE
The 4 product leaks costing you growth
A short audit guide for founders. Find the four places your product leaks revenue, and what to fix first.
What changed to get to 67%?
Only the screens the evidence implicated, redesigned to remove the observed stumbles.
Value earlier. Fewer decisions. The next action as the loudest thing on the screen. Unglamorous, and narrow on purpose.
Because the changes were narrow, attribution stayed clean. The rest of the product was held constant, so when the number moved we knew what moved it. Sweeping redesigns destroy that. You ship forty changes, the number moves, and you have learned nothing you can reuse.
The results: activation 38% to 67%, a 29 point rise. Onboarding completion up 68%. Support tickets down 73%, which surprised the team and should not have, because confused users file tickets and the confusion had been removed at its source. Q.AI kept growing and was later acquired by Forbes.
We have written the full sequence up as the anatomy of that engagement, because the method generalises even though the percentages do not.
How do you compare yourself to a benchmark honestly?
Match the definition before you compare the number, or you will benchmark two different things.
The benchmark defines activation as reaching a defined value moment within seven or fourteen days. Most products we audit are not measuring that. They are measuring signup completion, or email verification, or profile setup, and calling the result activation.
Those numbers look far healthier and mean nothing. A product with 80% "activation" measured at account creation is not beating the benchmark. It is not measuring activation at all. The working definition worth holding to is the point a new user first experiences the product's core value, not the point they finished configuring it (Amplitude).
What you might be measuring | What the benchmark measures |
|---|---|
Signup or account created | The defined value moment reached |
Email verified | Within a fixed 7 or 14 day window |
Profile or onboarding steps completed | Behaviour that separates retained users from churned ones |
Any first session activity | One event, chosen deliberately, instrumented on purpose |
If you cannot say in one sentence what your activation event is, the comparison is not available to you yet. That is fixable in an afternoon, and we have written how to define yours.
Where should you look first?
At the earliest step where a cohort of new users stops, not at the step you like least.
Founders usually know which screen embarrasses them. It is rarely the one costing the money. The screen costing the money is boring, early, and asks for something the user is not ready to give.
The order that has worked in every engagement since:
Define the value moment in one sentence, and get the team to agree on it.
Instrument a funnel across the new-user path to that moment.
Watch replays of the users who dropped at the worst step, until the reason is obvious rather than theorised.
Change only what the evidence implicates.
Measure on the same funnel, new cohorts against old, with the window agreed in advance.
You will notice there is no step where you rebuild the product. In the engagements we run, the biggest single lever has sat in the first minutes of the experience nearly every time, and the fix has almost never required new functionality.
a16z's growth metrics guide segments its own benchmarks by revenue scale, software type and go-to-market motion (a16z), for the same reason: a benchmark you cannot match to your own situation is decoration.
It also decides what your advertising is worth, since paid acquisition against a leaky product buys the same leak at a larger scale.
The benchmark is useful for one thing: telling you whether the gap between you and top quartile is worth three weeks of attention. At 38%, with an acquisition budget already running, it is not close.
WHAT NEXT