Media campaigns

A media campaign run properly has one brief, one record and one set of results, and the reason that is hard is that campaigns now span channels that each want to be measured on their own terms. Meta, TikTok, YouTube, retail media networks and the rest each report conversions their own way, and each claims more credit than it deserves because each sees only its own surface. We plan across channels to a single schedule and hold the results in one place, so that the campaign can be judged as a whole rather than as a pile of platform dashboards.

The measurement reform underway is the important story. The industry is moving away from last-click attribution and away from deterministic cross-platform identity, toward triangulation: media mix modeling validated by incrementality experiments, with platform attribution used only for tactical, in-flight signal. Google has invested heavily in its Meridian MMM through 2026, adding a Scenario Planner, geo-level incrementality testing and integration with Analytics, while Meta has quietly stepped back from its open-source Robyn framework. We treat MMM as the strategic backbone, incrementality tests as the validation, and platform numbers as directional. No single source is treated as truth.

• One brief, agreed before anything is booked.
• A schedule everyone can see and plan against.
• Spend and deliverables tracked in the same record.
• Honest reporting at the end, weak results included.

Where this heads is toward measurement that works without cookies and without individual tracking, built on modeled and experimental methods and on consented first-party data, often triangulated inside data clean rooms. We plan campaigns to be measurable that way from the start, because a campaign designed around identifiers that are disappearing is a campaign that will not be able to prove what it did.