jdte.digitalBook a call
Measurement

Marketing Mix Modelling

Marketing mix modelling answers the question the board asks: what would we lose if we stopped this? Joseph has built and run MMM at £50m+ annual media investment scale, where it proved a +9.0 ROI from YouTube brand investment over an 18-month cycle.

What you get

Data assemblyTwo years of weekly sales, media spend by channel and market, price, promotions, seasonality and external factors, cleaned into one dataset.
Model buildA model with adstock and saturation curves, validated against the incrementality tests already running.
Quarterly refreshThe model is re-run each quarter so every budget round is planned from the latest read.
Board readoutOne page: what each channel returned, where the next pound should go, and the confidence range around each number.

Who this is for

  • Businesses spending across several channels, including YouTube, TV, radio or brand, who cannot see what each one adds.
  • Marketing directors who need to defend the budget with evidence finance will accept.
  • Teams whose platform attribution claims more sales than the business made.

What MMM does that attribution cannot

Platform attribution counts the touchpoints it can see and gives credit to whoever was last in the room. It cannot see TV, radio, out of home or most of YouTube, and it cannot say what would have happened without the ad. MMM provides a cross-channel view of contribution, including channels that are difficult to assess through attribution alone. It needs no cookies and no platform data.

How Joseph runs it

  1. Assemble the data. Weekly sales and media by channel and market, with price, promotions, the seasonal calendar and the events that drive demand.
  2. Build and validate. Open-source tools such as Meridian and Robyn make MMM affordable for businesses spending a few million a year. The model is checked against geo tests and holdouts so the numbers are trusted.
  3. Run it as a cycle. Quarterly refreshes, so the model informs each budget round rather than sitting in a drawer.
  4. Report it plainly. One page for the board, with the return by channel and the recommended reallocation.

What it proved before

For a multi-brand betting and gaming group, MMM over an 18-month cycle showed a +9.0 ROI from YouTube brand investment, a channel the business had previously cut at every budget round. The wider performance programme delivered a 3.1 ROI in betting and gaming.

See the measurement and MMM case study.

Talk to Joseph about your account

A thirty-minute call. You describe the account, Joseph tells you what he would look at first. No deck, no pitch.

Book a calljoetobin17@gmail.com07568 172261