# Feature adoption

Feature adoption is the extent to which users discover, try and keep using a specific feature of a product. It is tracked per feature, as the share of eligible users who use it repeatedly, and shows which parts of the product deliver value and which go unnoticed.

## How do you calculate feature adoption rate?

Feature adoption rate is the share of eligible active users who use a feature repeatedly in a given period.

`Feature adoption rate = users who used the feature repeatedly ÷ eligible active users × 100`

"Eligible" matters. Leave out users whose plan, role or permissions do not include the feature, or the number will look worse than it is.

Say a reporting tool has 2,000 monthly active users, 800 of them on plans that include scheduled reports. In October, 120 of those 800 set up a schedule that ran at least twice. Feature adoption rate is 120 ÷ 800 × 100 = 15% of eligible users.

Alongside the rate, track:

- **Time to first use:** how long after sign-up or launch people first try it.
- **Repeat use:** how many who try it come back to it.
- **Breadth:** how many core features the average account uses.

## Why does feature adoption stay low?

Low feature adoption usually has one of four causes:

1. **Users don't know it exists.** It shipped with a release note nobody read.
2. **They can't find it when they need it.** It sits three menus deep, under a name that isn't the word they would use.
3. **Setup costs too much.** The feature needs configuration, data or a decision before it does anything.
4. **The value isn't clear.** They saw it and didn't see why it matters to them.

Split the data to tell these apart. If people who try the feature keep using it, the problem is discovery or setup. If they try it once and drop it, the problem is the feature or how its value comes across.

## How do you improve feature adoption?

Fix discovery and setup first. Name features in the words users search for, show them in empty states and at the point of need, and ship good defaults so the first use works without configuration. Announcements and hints help, but they often arrive while the user is busy with something else.

The stronger fix is discovery through asking. A user who types "send this report to my manager every Monday" shouldn't need to know that a feature called scheduled exports exists. The product should map the request to the feature and run it.

That is how Foldspace approaches discovery: the user describes the job, the Product Agent runs the matching feature [against the product's existing APIs](https://docs.foldspace.ai/start/faq/), and Conversational Analytics shows what each request was after, including requests the product couldn't handle yet.

## Learn more

- [Product Analytics](/platform/product-analytics)