Custom Formula Metrics: Build Your Own Numbers

Vexo already tracks the numbers that matter for a React Native app — errors, sessions, active devices, new downloads, custom events, screen taps. But the number you actually care about is often a combination of those: not “how many errors” but errors per session; not “how many taps” but taps per download. Those ratios are the ones that tell you whether a release is healthier or whether users are engaging more, and until now you had to work them out in your head or export to a spreadsheet.

Custom formula metrics let you define that number once and keep it. You build a formula from the metrics you already have, give it a name, and Vexo tracks it like any other metric — a current total plus a trend over time — as a widget on your mobile dashboard.

What a formula metric is

A formula metric is a small, named piece of arithmetic over metrics you already track. It has two parts: a name and an expression. The expression combines existing metric series with an operator — for example, error-count / user-session-count-v5 is “errors per session.”

Because the pieces are your existing metrics, a formula metric always reflects your live data. When new sessions and errors come in, the number recomputes from the same series that power the rest of your dashboard — there's no separate pipeline to keep in sync and nothing extra to instrument in your app.

Building one

The builder is a single row: a metric on the left, an operator in the middle, a metric on the right, and an optional name. Pick the two metrics, choose how to combine them, hit Add, and the new metric appears in the list below.

The metrics you can choose from are the ones Vexo already tracks:

• Errors
• Active devices
• New downloads
• Sessions
• Custom events
• Screen taps

The operators are the four you'd expect — divide, multiply, add, and subtract:

/    divide    →  a ratio, like errors per session
*    multiply
+    add
-    subtract

Division is the one you'll reach for most, because ratios normalize away growth: “errors per session” stays comparable whether you have a thousand sessions a day or a million, where a raw error count just tracks how big you are.

An expression can also use a plain number in place of a metric, so you can scale a value — for example error-count / 1000 to read errors in thousands. The name is optional; leave it blank and Vexo labels the metric with the expression itself.

Reading it: a number and a trend

Every formula metric renders the same way. You get its name, the expression it's built from shown underneath so there's no mystery about what the number means, a running total, and a small line chart of how the value moved over time.

The trend is the point. A single “errors per session” number tells you where you are today; the line tells you whether the last release made it better or worse. The series is bucketed over your selected time range, so the same metric answers “how are we doing this week” and “how did we do last month” just by changing the range.

Formulas worth building

A few that fall straight out of the metrics above:

• Errors per session — Errors ÷ Sessions. A release-health number that stays honest as your traffic grows.
• Taps per session — Screen taps ÷ Sessions. A rough read on how much users are actually doing per visit.
• Sessions per active device — Sessions ÷ Active devices. How often the people who have your app come back to it.
• Custom events per session — Custom events ÷ Sessions. Whether the action you instrumented is catching on.

None of these are new data — they're the numbers you already collect, combined into the one that answers your question directly.

Getting started

Formula metrics build on the events Vexo already captures, so if your app is reporting to Vexo you have everything you need. Install the Vexo React Native SDK, open your mobile dashboard, find the Custom Metrics widget, and add your first formula. New to Vexo? Create a free account and connect your app in a few minutes. For how these sit alongside segments, funnels, and retention, see the product-analytics overview.

Frequently asked questions