# KPIs that adapt to what each product is trying to achieve

Each project’s board leads with its objective and picks the figure that fits each growth step from the sources you connected, from active users to trial-to-paid, without a spreadsheet of metric definitions.

[Open Smart KPIs in the demo workspace](/tractionscout/demo/workspace?view=portfolio) (sample data, no account)

## What you can do

- **Led by your objective:** The card’s big number is the objective you set, or one Scout proposes from connected evidence and the server checks before adopting it.
- **Figures placed by meaning:** Every metric lands on a step of the growth chain by what it means, not which tool sent it. Crawler counts go to data quality, and an ad network’s own conversions are not counted as yours.
- **Active users from any source:** DAU, WAU, MAU and stickiness from the one source that identifies the most people, whether the tracker, the SDK, PostHog or Supabase sign-ins. Populations are never added together.
- **Trials and trial-to-paid:** Trials, trial conversions and trial-to-paid rate as cohorts by trial start day, from your billing lifecycle or the source that started the most trials.
- **Weakest link and guardrails:** Each card names the step with the clearest change for the worse, and stays quiet about speed, errors or refunds until they clearly get worse.
- **Withheld, not invented:** When a figure cannot be measured honestly, such as unlinked trial conversions, it is withheld instead of shown as zero.

## Why fixed dashboards fail small teams

A dashboard designed for one product rarely fits the next. A subscription app cares about trial-to-paid, a shop about orders, a content site about returning readers. Setting all of that up by hand, for every product, is the work most founders skip. TractionScout's board adapts instead.

## How a figure earns its place

Every figure from every source is placed on a step of the growth chain, **Reach**, **Visitors**, **Engaged**, **Converted** or **Revenue**, by what the metric means rather than which tool sent it. Figures that are not on the chain go to guardrails, data quality or activity beside it. For example, crawler counts are data quality, AI mentions count as reach, and an ad platform's claimed conversions are not counted as your conversions.

For each step, one figure leads. The order of preference is:

1. the objective you set for the project,
2. Scout's current focus,
3. a sensible default for the step, such as sign-ups, then trials and activation, then purchases for Converted,
4. the figure the source marks as its headline,
5. totals over the period before point-in-time readings,
6. the freshest figure.

## Objectives, automatic or yours

The card's big number is the project's objective. In automatic mode, Scout proposes the outcome from the evidence you connected, and the server checks that evidence before adopting it. Choose the objective yourself and your choice is kept.

## Active users and trials, from any source

Active users (DAU, WAU, MAU and stickiness) come from the one source that identifies the most people with product activity: the TractionScout tracker or SDK, PostHog with a user ID, or an account source such as Supabase sign-ins. Scout classifies each event name as product use, conversion or navigation, using only the name and aggregate counts.

Trials, trial conversions and trial-to-paid are cohorts by trial start day, from your billing lifecycle in Stripe or from the source that started the most trials, such as RevenueCat. When conversions cannot be linked to trials, the figure is withheld rather than shown as zero.

## Weakest link and guardrails

Each card names its weakest link: the step with the clearest change for the worse. Guardrails for speed, errors and refunds stay silent until they clearly worsen, so the board only asks for attention when something has moved.

Read the full method in [How TractionScout measures growth](/docs/measurement-methodology.html).

## Questions

### Do I have to define my KPIs by hand?

No. The board picks the figure for each step from what you connected. You can set the objective yourself, and a manual choice is always kept.

### How are active users counted?

Over the last 1, 7 and 28 finished days, from the single source that identifies the most people with product activity. Scout classifies event names as product use, conversion or navigation using only names and aggregate counts.

### What if a metric comes from two tools?

Each step shows one lead figure, chosen by the objective, Scout’s current focus and a preference for that step. Figures from different populations are never summed.


Maintained by TractionScout · Updated 2026-11-05


## Setup guides

- [How TractionScout measures growth](https://tractionscout.com/docs/measurement-methodology.html)
- [Connecting PostHog](https://tractionscout.com/docs/connect-posthog.html)
- [Connecting Stripe](https://tractionscout.com/docs/connect-stripe.html)

## Start with this question

What is the one number I should watch for this product, and which step is weakest?

[undefined](https://tractionscout.com/signup/tractionscout?prompt=What+is+the+one+number+I+should+watch+for+this+product%2C+and+which+step+is+weakest%3F)

Your question is carried into setup. Review it, connect the data you need, and send it to Scout when you are ready.

## Related examples and guides

- [Every product you run, on one screen](https://tractionscout.com/features/portfolio-dashboard.md): The portfolio puts each project on the same growth chain, from reach and visitors to sign-ups and revenue, so you see at a glance which product needs you today.
- [Cookieless web analytics that ends at sign-ups and revenue](https://tractionscout.com/features/web-analytics.md): A first-party tracker without cookies, with goals, live sessions, conversion journeys and site health, tied to the sign-ups and payments each visit led to.
- [App Store and Google Play numbers next to your web, search and revenue](https://tractionscout.com/features/app-store-analytics.md): Connect App Store Connect and Google Play Console to see the path from store impressions to installs, ratings, reviews, crashes and subscriptions, beside the rest of your product’s growth.
- [Scout, the AI that reads all of it and asks before it acts](https://tractionscout.com/features/ai-growth-assistant.md): Scout compares your numbers across every source, explains what may be in the way, prepares the change and checks whether it helped. It remembers what worked and works with your own agents through MCP.
