Skip to content
home

Data and analytics engineering · 2 min read

The MMM data readiness checklist

Twenty checks to run before you commission a mix model.

By WebIQ Analytics · 20 November 2025

Most disappointing mix models fail on the data, not the maths. These checks take an afternoon and save weeks.

History

  1. At least 104 weeks of data.
  2. Weekly, not monthly.
  3. No unexplained gaps or step changes.

Spend

  1. Every paid channel, from one consistent source.
  2. Each channel's spend varies over time.
  3. Channels aren't always planned together.
  4. Delivered spend, not plan or budget.
  5. Regional spend is real, not national spend split by population.
Illustrative: weekly spend checked channel by channel for enough variation to measure.Media inputsChannelWeekly spendTotalVariation (CV)IdentifiabilityPaid search£1.9m0.40Limited variationTV£5.8m0.62Varies enoughPaid social£2.5m0.32Limited variationOnline video£590k0.37Limited variationRadio£456k0.06Prior-driven
Weekly spend checked channel by channel for enough variation to measure.

Outcomes

  1. One agreed outcome definition.
  2. Total outcomes across all channels, not just tracked digital sales.
  3. Returns and cancellations treated consistently.

Controls

  1. Price and price changes.
  2. Promotions, with dates and depth.
  3. Seasonality and holidays identifiable.
  4. Competitor activity, even as an index.
  5. Known events logged.

Structure and evidence

  1. Sources, definitions and owners documented.
  2. A refresh process exists.
  3. At least one experiment or holdout result.
  4. Someone can explain what changed in marketing each quarter.

How to read your answers

  • Mostly yes: you're ready to model.
  • Gaps in controls or structure: model with caveats, and fix them for the next refresh.
  • Gaps in history, spend variation or outcomes: fix the data first. A model now would mostly reflect its assumptions.

Two checks matter more than the rest: whether spend varies (5) and whether channels move independently (6). If either fails, no modelling technique will rescue the result. The fix is planned variation in spend, or a test.

If you'd like a second opinion on your answers, a short readiness assessment is the first step in every WebIQ mix modelling engagement.

The data foundations your measurement depends on

No model is better than the data under it. We build the weekly, validated dataset your mix models, tests and dashboards need, and keep it trustworthy.