In this guide
- MMM models sales as a function of spend by channel, seasonality and other factors, using weekly totals.
- It needs two or more years of data and variation in spend. It works without user-level tracking.
- Use it for budget allocation across channels, together with lift tests.
As cookies fade and platforms report their own numbers, marketing mix modelling has come back. It is an old statistical method: it looks at how sales moved when spend moved, across all channels, with no tracking of individuals. This page explains what it does, what it needs and how a small team can start. It follows marketing attribution models and conversion lift studies.
What it is
An MMM regression relates a business outcome, such as weekly revenue, to inputs: spend per channel, promotions, pricing, seasonality, holidays and sometimes weather or the economy. Two effects are built in: adstock (an ad keeps working after the week it ran) and saturation (the next dollar buys less than the last). The model estimates each channel's contribution and its return at different spend levels.
Ask Ari instead of digging through reports
Ari reads your tracking and ad data and answers in plain words.
What you need
| Need | Why |
|---|---|
| Two or more years of weekly data | To see seasonal patterns and enough variation |
| Spend by channel and week | The main inputs |
| Sales or conversions by week | The outcome |
| A record of promotions and price changes | So they are not credited to ads |
| Variation in spend | If spend never changes, the model cannot learn its effect |
Strengths and limits
Strengths
No cookies or consent needed for the core data. Covers offline and organic effects. Gives saturation curves for budget planning.
Limits
Needs a long history. Works on weekly totals, so it is slow and coarse. Cannot tell you about individual creatives. Results depend on modelling choices.
Best used with
Lift tests to calibrate and attribution for daily optimisation.
Connected in minutes
Add your domain, install the pixel, connect your ad platforms and watch the first events land.
Tools
- Open-source packages such as Meta's Robyn and Google's Meridian (check their current status and documentation).
- Vendors that offer MMM as a service.
- Your own analysis in a spreadsheet or Python for simple versions.
Start small
Collect the data
Weekly spend by channel and weekly revenue for 104 weeks or more.
Clean and log events
Mark promotions, stock-outs and big changes.
Fit a simple model
With an open-source tool.
Check it
Does it predict a held-out period? Do the signs make sense?
Calibrate with a lift test
If you have one.
Use it for one decision
For example, how to split next quarter's budget between two channels.
Where tracking fits
MMM does not need server-side tracking, but it needs reliable totals: spend and orders. Your first-party analytics and your store are the source. Open Tracked's analytics shows spend and revenue by channel on first-party data, and the same data can feed an MMM. A tracking server also keeps day-to-day attribution trustworthy, so you have a second view to compare with the model.
Orders from your store, back to the ad that earned them
Your store sends the order once. Open Tracked passes it on to each ad platform, cleaned and deduplicated.
Frequently asked questions
What is marketing mix modelling?
A statistical method that estimates each channel's contribution to sales from weekly totals of spend and sales.
Does MMM need cookies?
No. It uses aggregated totals, not individual tracking.
How much data do I need?
Usually two or more years of weekly data with varying spend.
Which tools can I use?
Open-source packages such as Robyn and Meridian, or vendors that offer MMM.
Should I replace attribution with MMM?
No. Use MMM for budget allocation and attribution for daily decisions.
Do I need spreadsheets to read my results?
No. Once tracking is live, the Open Tracked analytics shows spend, revenue, ROAS and profit, and you can ask Ari, the built-in AI assistant, questions about your own numbers in plain words. Ari's allowance depends on your analytics plan.
Is Open Tracked different from Open Track or Open Space Track?
No. Open Tracked is the official name of our server-side tracking platform, and the website is opentracked.io. Many people search for it as Open Track, OpenTrack or Open Space Track. All of them lead to the same product: server-side Google Tag Manager hosting, first-party analytics and a CRM pipeline.
Fix your tracking at the source
Open Tracked hosts your server-side Tag Manager, keeps first-party cookies alive and shows what each ad is really worth.
The Open Tracked team
We build Open Tracked: server-side tracking hosting, analytics and the tools around them, for stores that spend on ads. Questions about this guide, or an idea for the next one?
Comments
Loading comments...
Guides · 5 min read
