I replace last-click guesswork with incrementality, experimentation and mix modelling — so spend follows what genuinely drives growth, not what gets the credit.
Client names withheld and figures sanitised. Each one followed the same spine: a question worth answering, a method that could survive scrutiny, and an outcome in dollars.
Platform-reported conversions implied a story finance didn't believe — and budgets were set on it.
Decisions ran on opinion and HiPPOs. Nothing was being proven before it shipped.
Measurement was about to break as third-party cookies went away across five brands.
Every new customer was treated as equal, so bidding chased volume over worth.
If a number can't survive a holdout or a control group, it isn't a result — it's a coincidence with good PR.
Every analysis ends in an action and a dollar figure. A chart that changes nothing is a cost, not an asset.
Marketing's measurement and the CFO's revenue should reconcile. I'd rather have one defensible truth than two convenient ones.
A public side project — forecasting weekend takings, decomposing word-of-mouth decay, and pressure-testing whether marketing spend actually moved the opening. Built in the open because the best way to prove a method is to publish it before you know the answer.
Multi-touch attribution feels rigorous and isn't. What it's actually measuring, and what to use instead.
A plain-language tour of holdouts, geo tests and the traps that make a "lift" disappear under scrutiny.
Prioritisation, power, and the cultural work of getting an org to ship the test before the opinion.
Why marketing and finance disagree on the same number — and the reconciliation that buys back trust.
I'm a digital analytics manager who's spent eleven years turning marketing and customer data into decisions people can defend in a boardroom.
My background runs across web, app and campaign measurement, with deep ownership of the Adobe and GA4 stack and a specialism in the harder end of the discipline — incrementality, experimentation and mix modelling. I've built and mentored analytics teams, and I care most about the bit where a number turns into a budget decision. Outside work I write about measurement, run a box-office data project, and occasionally argue with people about whether their dashboard is doing anything at all.
Whether it's a role, a hard attribution question, or an experimentation program that needs building from scratch — let's talk.