Attribution System
Triangulation
One Question, Several Answers
Ask three measurement sources how a channel performed and you'll get three numbers: the platform's own attribution, Google Analytics, and MMM incrementality each measure differently, and they rarely match. Most teams either pick one and ignore the rest, or spend meetings arguing about which is right instead of deciding what to do.
Triangulation is the discipline of reading them together — and Maxma does the alignment automatically. Every source is mapped to the same channel and tactic cuts, the same markets, and the same time axis, so a side-by-side comparison is actually apples to apples:
Same channel, same tactics, same weeks, same scale — the platform claims nearly 2× what the MMM measures, and the gap concentrates in one tactic.
Agreement Builds Confidence; Discrepancy Builds Insight
Where the sources track together, you can act with confidence — the result doesn't depend on which measurement you trust.
Where they diverge, the shape of the gap is itself the insight. A discrepancy spread evenly across a channel usually points to a definitional difference, like attribution windows. A gap concentrated in one tactic — as in the example above, where platform-reported conversions run far ahead of measured incrementality for a single automated format — usually means that tactic is claiming credit for conversions that would have happened anyway. That's not a data quality argument; it's a budget decision waiting to be made.
From there, the next steps are concrete:
- Recalibrate what "good" means — set the tactic's targets on incrementality-calibrated numbers rather than platform-reported ones (see Calibrated Attribution in the measurement framework).
- Settle it causally — when the stakes justify it, run an incrementality test as the tie-breaker.
- Act on the answer — cap, restructure, or reallocate the over-credited spend, and watch the next measurement cycle confirm the move.
Views Built Around How You Read Data
Triangulation lives in Maxma's business intelligence dashboards, and no two teams read performance the same way. Views can be built and customized to each customer's needs — compare sources by channel, tactic, market, or funnel stage; pick the metrics and breakdowns that match how your team plans; and save the views your weekly rhythm actually uses. The comparison above is one layout, not the only one: