For the complete documentation index, see llms.txt. This page is also available as Markdown.

UMM

UMM combines your MMM results with Roivenue multi-touch attribution in one model. Overview of the UMM pages and the order to set them up.

UMM combines your marketing mix modelling (MMM) data with Roivenue multi-touch attribution (MTA) data into a single view. MTA sees the digital touchpoints in customer journeys. MMM also covers activities MTA cannot track, such as TV or out-of-home. UMM mixes both, so you can see the contribution of all your marketing in one model.

UMM is a section in the left navigation of Roivenue. It brings together the pages for uploading MMM data, connecting it to your MTA data, building UMM models and analyzing the results.

The section consists of these pages:

Page
What it is for

Upload the results of your MMM model to Roivenue from an Excel template.

Assign a category to every MMM driver and map online media drivers to your MTA data.

Create UMM models and set how MTA and MMM are mixed for each site.

Add results of real-world tests and use them to calibrate MTA or UMM models.

Compare UMM, MTA and MMM results side by side.


How to set up UMM

Set up the pages in this order. Each step needs the previous one.

  1. Upload MMM data in MMM Data Upload.

  2. Categorize the drivers in Driver Categorization. Drivers without a category are excluded from UMM.

  3. Create a UMM model in UMM Models and add a configuration for each site.

  4. Optionally, add experiments in Experiments and connect them to the model.

  5. Analyze the results in UMM Analysis.

Tip: After you upload data, change a driver, save a model or save an experiment, the model recalculates. This usually takes a few hours. The results appear in UMM Analysis once it is done.


FAQ

1. What is the difference between MTA, MMM and UMM?

MTA attributes conversions to the digital touchpoints in customer journeys. MMM estimates the effect of marketing activities from aggregated data, including channels MTA cannot track. UMM mixes both into one model.

2. Does Roivenue build the MMM model?

No. You upload the results of your MMM model in MMM Data Upload. Roivenue then mixes them with your MTA data.

3. Why is UMM Analysis empty?

At least one step of the setup is missing, or the model is still recalculating. Check that the MMM data is uploaded, all drivers are categorized and the UMM model has a configuration for the site.

4. How long does the recalculation take?

Usually a few hours. The results appear in UMM Analysis once it is done.

5. I cannot see the UMM section. Why?

UMM may not be available for your account or your user role. Contact the Roivenue team through the Roivenue Help Center.

6. Can I use experiments without UMM?

Yes. You can connect an experiment to an MTA model. This creates a version of the MTA model that includes the experiment. See Experiments.

Last updated

Was this helpful?