The Revenue Data Aggregator

Every platform tells a different story. GA4 says one revenue number, your store platform says another, Google Ads claims credit for a third — and Clarity knows why pages fail but nobody looks. This agent reconciles all four into one weekly truth per store, so my team at MageCloud optimizes against reality instead of a platform’s flattering version of it.

Get this skill — free in the pack MageCloud CRO services

At a Google workshop on using AI in marketing

What it takes off your team’s plate

The stitching work that eats an optimizer’s Monday, done before the coffee is ready:

Cross-platform reconciliationPulling GA4, Google Ads, Clarity, and store orders into one sheet, per client, every week — and doing the trust math between them.

Attribution argumentsDeciding whose number to believe when the ad platform claims 60% of revenue and the store says otherwise.

Movement diagnosisWorking out whether revenue moved because of traffic, conversion rate, or order value — by channel and by device.

The why behind the whatDigging through session recordings to explain a conversion dip instead of guessing.

How a run works

Five passes, same order every week, producing one page per store:

Reconcile revenue on the trust ladder. Platform orders are ground truth. GA4 is the measurement layer. Ad platforms are claims. The agent builds the ladder — for example: store £48,200, GA4 £45,900 (95% capture), Google Ads claiming £19,300 — and flags capture below 90% as a tracking emergency.
Decompose the movement. Revenue = sessions × conversion rate × AOV. The agent attributes this week’s change across the three factors, by channel and device: ‘sessions flat, mobile CVR down 11%, AOV up 3% — the problem is mobile conversion, not traffic.’
Attach the why. For whichever factor moved, it pulls the Clarity evidence — rage clicks, dead clicks, quick-backs, checkout abandonment — and links two or three actual session recordings.
Cross-channel sanity. Blended MER against per-channel claims, new-vs-returning revenue split, and the paid-dependence trend.
Publish the digest. One page a practitioner reads in ninety seconds, plus a machine-readable metrics file every other agent in the pack consumes — one version of the truth per store, per week.

✋ The approval gate

This agent is the instrument panel, not the pilot:

It makes no optimization recommendations — the optimizer and the other skills decide
When sources disagree, it reports the disagreement; it never averages it away
Every number carries its source and date range; unverifiable numbers do not ship
One exception: broken tracking gets flagged immediately as an emergency

What you get

REVENUE-DIGEST-[store]-[week].md — the ninety-second weekly read

metrics.json — reconciled figures every other agent reads from

Capture-rate alerts when GA4 drops below 90% of real orders

A revenue time series that compounds into the agency’s asset

Year-ahead planning with Roger from A&E Leisure

I have spent twenty years watching stores make decisions on the wrong number. The fix is never a prettier dashboard — it is agreeing what the truth is before arguing about what to do. That discipline is what this agent automates.

I have written about this from the trenches:

Works together with

Run it yourself, or bring my team in

The skill is free in the pack — install it and start this week. If your store does seven figures and you want this running inside a full revenue-optimization engagement, my team at MageCloud handles it end to end.

Download the pack Book a call