# Read GA4 performance by device

> Tier: recipe. Operating recipe: real tool sequence, core rules, known traps.

Device cuts on engagement and conversion, the site-side check on device debates.

**Intent.** Device-class truth from the site's own measurements.

## When to use

- Before device-based bidding.
- Mobile-performance arguments.

## When not to use

- No specific contraindication beyond the preconditions below.

## Sources

Required: Google Analytics 4

## Preconditions

- The account is ticked in the dashboard, confirmed via list_accounts.

## Inputs

- `window`: Analysis window. Default: last 30 days.

## Procedure

1. Run the device report. — tools: `ga4_run_report`
2. Cut the worst device by browser before concluding. — tools: `ga4_run_pivot_report`

## Decision rules

- **When** One browser drags the device class → **conclude** A compatibility bug wearing a device costume: name the browser for the product team.

## Evidence requirements

- Every figure carries its account, metric and window.
- Anything below a readable sample size is reported as unjudged, not as zero.

## Human approval boundary

- The readout changes nothing. Any change it motivates goes through Safe Writes: preview, human confirmation, then apply.

## Failure modes

- Tablet is a rounding error most places: do not let its noise reshape mobile conclusions.

## Output

- **Readout**: The figures, with windows and accounts named.
- **Flags**: What deserves a deeper skill or a human decision.