Case study

26% more conversions, from data alone.

A performance agency · Last updated: July 2026

A performance agency was making budget decisions on numbers that looked precise and were quietly wrong. We rebuilt the tracking underneath their client accounts, and the campaigns didn’t change, but the results did.

The situation

Strong campaigns, shaky numbers.

The agency ran strong campaigns for a portfolio of ecommerce and lead-gen clients. Reporting came from GA4 and platform dashboards that rarely agreed, and every monthly review started with a debate about which number was real.

What was broken

Signal stripped before it counted.

Browser restrictions and ad blockers were stripping a large share of conversions before they were ever counted. Enhanced conversions were half-configured. The platforms were optimizing budget on a fraction of the true signal.

One client, one month, three reports. The ad platform, GA4 and the client's own back office each report a different conversion total, and the note under each explains why: one counts the clicks it can claim including modelled ones, one counts the sessions the browser still allowed, one counts the orders that were actually paid. An arrow leads to the budget decision that was made on one of them, marked with a question mark.

What we fixed

Configuration, per client, verified.

A first-party server-side setup, configured for each client rather than copied between them

Enhanced conversions and CAPI verified live against the existing numbers before switchover

A single source of truth in BigQuery, so every client report finally reconciled

The same client account before and after, four rows on each side. Collection in the browser becomes collection server-side, consent defaults set late become defaults set before any tag fires, half-configured enhanced conversions become fully mapped, and no verification step becomes verification on every release. The four rebuilt rows lead into three outcomes: customer match, lead scoring, and profit-based optimization.

What recovering the data was worth.

38%

of client traffic affected by tracking prevention, now recovered

+26%

average uplift in measured conversions

14 hrs

saved per project on manual reporting

What became possible

Clean data opens doors guesswork can’t.

Customer match

Recovered first-party data made audience match usable again across platforms.

Lead scoring

Clean events fed a lead-quality model instead of guesswork on form fills.

POAS, not just ROAS

Real margin data flowing through meant optimizing on profit, not revenue.

The next step

See what you’re losing.

We map where conversion data goes missing across your client accounts, and show what recovering it would be worth. Not a demo, not a sales call.

Request a free audit →