Budget Simulator
Move the money.
Read the revenue.
Drag each channel's share of the total and read modeled revenue off the fitted response curves. The same curves MMM produced, so the simulator and the model never disagree.
Share of budget
started → allocated
Total spend
unchanged
held fixed for a fair comparison
Modeled revenue
higher
per the fitted response curves
That gap is the case for reallocating. Read off the fitted response curves, not a rule of thumb.
How it works
01
Start from the fit
The simulator opens on the current allocation and the response curves from the latest MMM fit.
02
Reallocate
Move share between channels. The total stays fixed unless you change it, so the comparison is fair.
03
Read the gap
Modeled revenue against the current mix, with its interval. That gap is the case, or not, for the change.
What you get
Same curves as the model
No separate assumptions. If MMM says a channel saturates, the simulator shows it flattening.
Fixed-total comparisons
Reallocation is judged at constant spend so a bigger budget is not mistaken for a better mix.
Interval on the outcome
Modeled revenue carries the fit's uncertainty into the decision.
Shareable scenarios
Save an allocation and hand the link to whoever owns the plan.
What it will not do
Boundaries we state up front, so the numbers are trusted.
- It does not simulate without a fit. No response curves means nothing to simulate, and the page says so.
- It does not extrapolate far past observed spend without widening the interval.
- It does not plan a campaign funnel; that is the Campaign Simulator.
Also in the product