Shopify brand: The MMM said we could triple Meta spend. We did, and revenue bare
By gavintrippe · September 18, 2026 · Curated by George's Blog
Shopify brand: The MMM said we could triple Meta spend. We did, and revenue barely moved. What happened?
Media buyer: The model missed.
Measurement partner: No, the execution missed. The forecast assumed the extra Meta spend would reach more people. You tripled spend into the same audiences, so frequency went up and reach barely moved.
Media buyer: What are you talking about? We spent exactly what your model told us to spend!
Measurement partner: Yes, but the recommendation assumed that extra spend would buy more reach, not show the same people three times as many ads!
Shopify brand: So did the model fail or did the media execution fail?
Media buyer: The model!
Measurement partner: The execution!
Shopify brand: Great. And how am I supposed to know which one of you is right?
We used to sell models on their own at P1, but we pretty much stopped because a forecast is only as good as the execution behind it. If the people buying the media don’t understand what the model is actually assuming, you’re basically handing them a number and hoping they execute it correctly. That’s where these setups fall apart.
Talked through this with Will Haire on the Selling on Giants An eCommerce Show for Brand Operators podcast, full episode linked in the comments below.