I still hear the pitch: take your sales history, add your media spend, fit a trend line through the rest, and you have a Marketing Mix Model. It’s an appealing simplification, and it is also close to useless, because it ignores almost everything else that was actually moving the consumer.
Imagine a mobile retailer runs a campaign built around a single headline offer: iPhone from £50 a month. Sales jump. Drop that period into a model with no price or promotion variable, and every extra unit sold gets credited to whichever channels carried the message. The media plan looks spectacular. None of it is true. The offer sold the phones. The advertising told people about the offer. A model that can’t tell the two apart will hand the credit to the wrong one every time, and nobody reading the deck has any way of knowing.
The same failure runs the other way too. A brand can grow while doing nothing especially well, simply because the category it sits in is growing, or a competitor has stumbled, or a macro trend is doing the work for everyone in the market at once. A rising tide carries every ship in the harbour. Without a term in the model that separates the tide from the swimming, growth gets read as marketing effectiveness by default, and a brand that is coasting looks, on paper, exactly like a brand that is winning.
Both failures share the same root cause: a model built around the data that happened to be easy to collect, rather than the full set of things actually influencing the decision. Price, promotions, distribution and availability, competitor activity, seasonality, the wider category’s growth rate, consumer sentiment, even the weather – leave any of them out and the model doesn’t ignore their effect, it hands that effect to whatever variable moves with it. Often that is media, because campaigns are timed to coincide with offers and seasonal peaks, and because media is nearly always in the model while everything else is a fight to get there.
The single most important meeting in any Marketing Mix Modelling exercise happens before a spreadsheet is opened. Its job is not to inventory the data you already hold. Its job is to work out everything that plausibly influences the consumer’s decision, whether or not you currently have a way to measure it, because the honest starting question is never “what data do we have”, it’s “what would change this person’s mind”.
Put yourself in the room with the consumer at the actual moment of the decision. What are they weighing? The price, obviously, and whether it just moved. What’s in stock, and where. What a competitor is doing that week. Whether it’s the month everyone buys this category anyway, or whether something in the news has just made the whole market more or less inclined to spend. Build that list first, exhaustively, before asking which parts of it you can source data for. The factors you can’t measure well don’t disappear from the consumer’s decision just because they’re hard to find data for. They disappear from the model, quietly, and get misattributed to whatever’s left.
A model missing half the picture will still hand you a number with three decimal places. The precision is real. The story it’s telling you almost certainly isn’t.
If your model has never had that meeting, drop a line to hello@themmmdoctor.com.