# MMM brief: ________________

_A brief for a marketing mix modeling project. Agree it with the people who will act on the result before any data is pulled. From the Vexpower MMM library: vexpower.com/mmm/mmm-measurement-plan._

Prepared by: ______________   Agreed with: ______________   Date: ______________

## The one-sentence brief

We will use the model to decide ___, over ___, measured by ___, while respecting ___.

_If you can't fill this sentence in, the project isn't ready for a model yet._

______________________________________________

## 1. Decision

What will someone actually change because of this model? Name the budget, the period and the levers that can move.

_e.g. How to allocate next quarter's £300k across search, social and brand._

______________________________________________

## 2. Decision owner and date

Who acts on the result, who signs it off, and when the decision is made.

_e.g. Marketing lead, with finance sign-off, at the December planning meeting._

______________________________________________

## 3. Outcome KPI and why

The quantity the model explains, defined in full: revenue, orders, units, new customers or contribution; gross or net of returns; which geography. Say why this outcome matches the decision.

_e.g. Incremental net revenue after returns, checked against contribution after advertising at a 40% margin, because finance approves budgets on contribution._

______________________________________________

## 4. Time horizon and granularity

History available (ideally two or more years), granularity (weekly is typical), geography, and the planning horizon including carryover.

_e.g. Weekly UK data for about three years; the decision covers one quarter plus six weeks of follow-up._

______________________________________________

## 5. Channels and levers

Every paid channel, with the metric you'll use (spend, impressions, GRPs, clicks), plus non-media levers such as price, promotions, distribution and email.

_e.g. Paid search, paid social, brand video, price and promotions, email sends._

______________________________________________

## 6. Known events and other drivers

Anything else that moved the outcome: seasonality, holidays, promotions, price changes, outages, stock-outs, launches, competitor activity, macro shocks. Include dates.

_e.g. Annual January sale; website outage in March 2025; 8% price rise in September 2025._

______________________________________________

## 7. Constraints

Management limits the recommendation must respect: total budget, channel minimums and maximums, contracts, delivery or fulfilment capacity, brand presence.

_e.g. Total fixed at £300k; brand stays at £30k; search between £90k and £180k._

______________________________________________

## 8. Existing evidence

Experiments, lift studies, holdouts, platform reporting, analytics and previous models, with dates and scope. Note which have already been used to calibrate anything.

_e.g. 2025 conversion lift study on prospecting social; 2024 regional search holdout; platform ROAS._

______________________________________________

## 9. How results will be used

Where the output goes and what it won't be used for: quarterly allocation, annual planning, scenario comparison; not weekly bidding or creative choices.

_e.g. Set the quarterly split; rerun each quarter._

______________________________________________

## 10. What would change the decision

The threshold or result that would make you choose differently. Agree it before the model runs.

_e.g. If extra social spend returns less than £2.50 per £1 (breakeven at 40% margin), keep the current mix._

______________________________________________

## Sign-off

| Role | Name | Date |
| --- | --- | --- |
| Decision owner | | |
| Finance | | |
| Analyst / modeller | | |