INDIEZ: custom metrics need versioned definitions, not just editable formulas

INDIEZ's public case describes tailoring metrics to campaign objectives. That flexibility becomes useful only when the definition travels with the number used to make a buying decision.

What the public case establishes

Adjust describes INDIEZ using raw data and custom metrics to analyse sources, creatives and countries, alongside fraud prevention. The account says metric choices differed with campaign goals and network optimisation. It also reports identifying suspicious traffic from one partner; that observation is specific to the supplier's case. Adjust case study.

This debrief analyses a public customer case published by a commercial supplier. It is not an original UA Ledger interview. The chronology below follows the supplier’s account; we did not inspect the advertiser’s dashboards or interview its staff.

The decision worth examining

Two buyers can both report ROAS while using different revenue streams, time windows or exclusions. A configurable formula makes each calculation easier but does not reconcile them. The operational requirement is a versioned definition with a reason for every change, so historical decisions can be reconstructed without relying on the memory of the analyst who built the dashboard.

Separate formula changes from data corrections. Changing an attribution exclusion is a new interpretation of the cohort; receiving a delayed cost file is a new observation under the same interpretation. Both can move the output, but they should appear differently in the record. This distinction makes disagreements easier to diagnose and reduces accidental rewriting of earlier decisions.

A usable next check

Select one metric used by multiple teams and compute it from the same synthetic input rows using each team's current definition. Surface all differences before choosing a shared version. Preserve alternatives where they answer different questions. In production, attach the metric version and data cut-off to every export that can trigger a budget action.

Download the populated case worksheet. Its control, variant and decision rules are a proposed experiment, not an account of results already measured. Keep the source URL with the worksheet when circulating it so reported findings cannot become unattributed team benchmarks.

Evidence boundary

No INDIEZ dataset or vendor fraud classification was independently checked. The case does not supply a universal formula, and its suspicious-partner example cannot be generalised to an entire network or market.

For background, see reading an mmp dashboard without fooling yourself and cohort quality checklist ua spend. These are existing archive discussions, not independent certification of this case. No first-hand campaign result is asserted.