Campaign measurement
Attribution, incrementality and MMM
Google, Meta, and your finance team can report different numbers for the same sale, because they measure it differently.
We establish what each number represents and what your campaigns are actually contributing. So you can put your marketing budget where the evidence says it belongs.
Know what you're actually paying for
Google and Meta don't give you an independent view of marketing performance.
They measure their own contribution, using their own data, attribution rules, and windows.
We bring platform, analytics and business data together, then use the method that fits the decision you need to make.
Attribution
Which touchpoint gets the credit, and where the platforms' claims overlap. Useful for day-to-day optimisation, not proof of cause.
Incrementality testing
How many additional outcomes the advertising actually created.
Marketing mix modeling
How investment across channels contributes to business outcomes over time.
Attribution tells you which activity gets credit for a conversion. Not whether the advertising caused it.
We compare a group exposed to advertising with a suitable control group that is not. The difference in outcomes estimates the effect the advertising created.
Say 1,000 customers buy after seeing an ad.
800 would have bought anyway
200
The platform reports 1,000. The advertising created 200.
Getting the test design right is critical. Sample size, seasonality, the control group and differences between test populations all affect what you can conclude. We design and validate the test, run the analysis, and compare the result with what the platforms reported.
If the platform reports a strong effect and the test doesn't support it, you want to know before you put more money into the campaign.
Did the advertising actually cause the sale?
When is marketing mix modeling worth it?
MMM is for the bigger investment questions: how channels contribute to business outcomes, and where changes in spend are having an effect.
It uses historical data such as media spend, sales, seasonality and promotions to estimate those relationships over time. It needs enough evidence to work with.
| History | roughly 3 years of consistent data |
|---|---|
| Below that | platform data and targeted incrementality tests usually give a better answer |
Before any budget advice, the model has to pass fixed checks: it explains at least 70% of the variation in your results, its error stays below 15%, and its estimates are stable. If it fails, you get the fit report and our view of what's missing, and no budget advice.
The team assessing campaign performance isn't the team spending your media budget.
Campaign strategy and execution stay with your media agency or internal marketing team. Where relevant, a specialist team within LevelUp Group can handle that work. Our responsibility is the measurement: we bring the data together, establish what the numbers represent, and test the evidence behind reported results.
We work directly with the teams running the campaigns, without taking over their role.
Your media team runs the campaigns. We measure what they deliver.
Frequently asked questions
-
Why can't we just use GA4?
GA4 gives you a cross-channel view of the activity it can observe and attributes conversions according to its methodology. It cannot tell you how many of those conversions would have happened without the advertising.
-
What do you need from us?
Start with what you already have: platform reporting, analytics data, campaign history and business results. We establish what each system measures, why the numbers differ, and what the evidence can reliably tell you.
-
Is MMM a type of attribution
No. Attribution assigns credit to touchpoints in individual customer journeys. MMM doesn't use journeys at all: it models total spend against total sales over time. They answer different questions, and they work best side by side.
Find out what your campaigns are really contributing
Tell us which decision you need to make. We look at what you already measure and which method fits: attribution, an incrementality test or MMM.
Contact details
Cloud Nine Digital
Johann Siegerstraat 24
1096 BH
Amsterdam, Netherlands
T: +31 (0)20 787 1323
E: info@cloudninedigital.nl
Need something else?
Please do reach out. We are happy to help!