Product Adoption Metrics: The 7 That Matter and How to Calculate Them

The core product adoption metrics are adoption rate (share of eligible users who used a feature), activation rate (share who reached first value), time to value, breadth of adoption (how many features an account uses), depth (how intensively), stickiness (DAU divided by MAU), and retention lift among adopters versus non-adopters. Adoption rate alone is the weakest of the seven, because it counts one-time use. Stickiness and retention lift are what tell you a feature actually earned its place.

Most adoption reporting stops at a single percentage: how many accounts touched the feature. That number is easy to produce, easy to present, and nearly impossible to act on. Below are the seven metrics that let you tell the difference between a feature people tried, a feature people use, and a feature that changes whether they stay.

1. Feature adoption rate

Formula: (users who used the feature in the period / users eligible to use it) x 100

The headline metric, and the one most often calculated wrong. The error is in the denominator: measuring adoption of an admin-only feature against your entire user base guarantees a small number that means nothing. Eligibility should reflect who could plausibly use it, filtered by plan, role and whether they have even reached the part of the product where it lives.

Pick a window that suits the natural frequency of the job. A feature used at month-end close should be measured monthly, not weekly, or you will report a failure every three weeks out of four. And be explicit about whether you are counting users or accounts, because in B2B those two numbers can differ by an order of magnitude and teams routinely compare one against the other without noticing.

2. Activation rate

Formula: (new users who reached the activation event / new users in the cohort) x 100

Activation measures whether a new user reached the first moment of real value, not whether they clicked around. Defining the activation event is the hard part and worth doing properly: find the early action that most strongly separates users who are still around at day 30 from those who are not. It is usually something concrete, like connecting a data source or completing a first real piece of work, rather than finishing a tour.

Track activation by cohort rather than in aggregate. A blended activation rate that looks steady often hides recent cohorts activating worse, which is the earliest warning you get that acquisition or onboarding has drifted.

3. Time to value

Formula: median time from signup to the activation event

Use the median, not the mean, because a handful of accounts that activate six months later will drag an average into fiction. Time to value tells you how much patience your product currently demands, and it is one of the few adoption metrics with a fairly universal direction: shorter is better, almost always.

Segment it by how the user arrived. Self-serve signups from a landing page and sales-assisted accounts with an onboarding call usually have completely different curves, and averaging them produces a number that describes nobody.

4. Breadth of adoption

Formula: average number of core features used per active account in the period

Breadth captures how much of the product an account has genuinely taken up. It matters because accounts using one feature are far easier to replace than accounts that have woven three or four into their weekly routine. In most B2B products, breadth correlates strongly with retention and with expansion revenue.

Define your "core features" list deliberately and keep it stable, ideally five to ten capabilities that represent distinct jobs. If the list changes every quarter, the trend line is meaningless. Watch the distribution as well as the average, because a mean of 2.4 could be everyone using two or half your base using one and half using four.

5. Depth of use

Formula: average uses of a feature per adopting user per period

Depth separates the tried-it users from the rely-on-it users. A feature with 45% adoption where the average adopter uses it 1.2 times a month is a feature people tested and abandoned. The same 45% at 14 times a month is a feature in someone's daily workflow. Adoption rate reports these identically, which is precisely why it should never travel alone.

Depth is also the fastest way to spot a feature that is nearly working. High adoption and low depth usually means the value is real but the experience is costly enough that people avoid returning, which is a very different fix from low awareness.

6. Stickiness (DAU/MAU)

Formula: (daily active users / monthly active users) x 100

Stickiness estimates how many days a month a typical user shows up. A ratio of 20% means the average monthly user is active roughly six days a month. Whether that is good depends entirely on the job your product does: a team messaging tool should be far higher, a quarterly planning tool far lower, and comparing across those categories is meaningless.

Metric	What it answers	Fails when used alone
Adoption rate	Did anyone use it?	Counts a single try as success
Activation rate	Did new users reach value?	Says nothing about ongoing use
Time to value	How long did it take?	Skewed badly by outliers if you use the mean
Breadth	How much of the product is in play?	Hides the distribution behind an average
Depth	Is it a habit or a trial?	Only meaningful within an adopting group
Stickiness	How often do they return?	Not comparable across product categories
Retention lift	Does the feature matter?	Correlation, not proof of causation

7. Retention lift among adopters

Formula: retention rate of accounts that adopted the feature minus retention rate of comparable accounts that did not

This is the metric that decides whether a feature earns its maintenance cost. If accounts that adopt a feature retain at 91% and similar accounts that never touch it retain at 78%, that 13 point gap is the strongest argument you will ever make for investing in its adoption.

Two cautions. Compare like with like, or you will simply rediscover that engaged accounts do more of everything. And remember this is correlation: the feature may attract the accounts that were always going to stay. The way to firm it up is to look at whether retention improves after adoption within the same accounts, rather than only comparing groups.

What is a good product adoption rate?

There is no single figure, and any article quoting one is guessing. It depends on how broadly the feature applies. A core workflow capability every account needs should reach a large majority of active accounts within a few months, and under half signals a discovery or value problem worth investigating now. A specialist feature built for one persona can be a clear success at 10% adoption if those are exactly the accounts it was built for and they use it deeply.

Two better questions than "is our number good": is adoption growing within the segment the feature targets, and do adopters retain better than non-adopters? A feature with modest adoption that meaningfully lifts retention in its target segment is doing its job. Wide adoption with no retention effect is decoration you now have to maintain.

Why adoption metrics stall without the reason attached

Every metric above describes the gap. None of them explains it. When a feature sits at 9% adoption, the number is identical whether users never found it, found it and did not see the point, tried it and hit an error on their plan, or already solve the job with a spreadsheet they like better. Those need four different responses, and the standard reflex, adding a tooltip, only addresses the first.

The reason is usually sitting in data you already have. Non-adopters file support tickets, answer surveys, leave reviews and write in-app comments, and they tend to name the barrier plainly. Theming that feedback and tying each theme to the accounts behind it turns a flat adoption percentage into a ranked list of causes with revenue attached. That is what product adoption software should give you, and it is the difference between reporting adoption and improving it.

One practical warning before you act on any of these numbers: adoption metrics are only as trustworthy as the event data underneath them. A renamed event or a tracking change on one platform can produce a convincing adoption cliff that never happened, so it is worth being able to monitor your data for freshness, volume and schema changes before a phantom drop sends a team chasing a problem that does not exist.

Where to start

If you are building this from nothing, start with three: activation rate by cohort, depth for your two most strategic features, and retention lift for those same two. Those answer whether new users reach value, whether the features they reach become habits, and whether any of it affects whether people stay. Add breadth and time to value once the first three are stable and trusted.

Then go read what the non-adopters said. For the measurement side of the picture, our guide to product analytics metrics covers the broader set, funnel analysis covers the step-by-step path users take to adoption, and user behavior analytics shows where they get stuck along the way. And if a weak number has you weighing a guidance tool to lift it, our explainer on what a digital adoption platform is covers when that actually helps and when the better move is fixing the product.