Notes
Field notes from marketing data.
Practical writing on tracking, CDPs, measurement and making marketing data ready for AI, from thirteen years of doing the work.
Is your marketing data ready for AI? A practical checklist
Most AI projects in marketing stall on the data, not the model. Here is what "AI-ready" actually means.
Read →What a tracking audit usually finds, and how to fix it fast
Broken conversions, duplicate pixels and a data layer nobody owns. The same problems show up almost everywhere.
Read →CDP or warehouse first? How to choose
Packaged CDPs and warehouse-native setups both work. The right answer depends on your team, not the vendor pitch.
Read →Chrome kept third-party cookies. Plan as if it hadn't.
Cookies stayed in Chrome, but privacy rules and other browsers still decide what you can measure.
Read →Holdouts: the test that shows what your marketing caused
A campaign can look great and still change nothing. A holdout tells you which.
Read →How we cut 2,000 reports to 200 after a merger
Executives had stopped trusting the numbers. The fix started with interviews, not code.
Read →