<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=9151474&amp;fmt=gif">
Skip to content

Glossary · Data and operations

What is offline media spend in MMM?

Offline media spend is the money spent on channels with no API, such as TV, radio, out-of-home, print and influencer collaborations. For a marketing mix model it has to be collected from media plans and spot reports and structured at the same grain as digital, down to the individual spot.

Reviewed November 2026

By the Odins team · Updated November 2026 · About 3 minutes

Definition

Digital channels report themselves. TV, radio, out-of-home, print, cinema, sponsorships and influencer collaborations do not: the spend lives in media plans, agency invoices and broadcaster spot reports, in a different shape for each. Offline media spend is that money, and for a marketing mix model it is often the most important money, because it is the part attribution cannot see at all and the part where the budget questions are largest.

How it is collected

The rule at Odins is that offline channels get a pipeline, never a monthly upload. Media plans and spot reports come in from the customer or its media agency, through a standing feed, and are broken down to the individual spot or placement: this spot on this broadcaster at this time at this cost. From there each spot lands in the same hierarchy as the digital campaigns, at the same daily grain and in the same currency. With a well-organized agency the history is in place within a day.

By default, costs that vary with the media go in and fixed agency fees and mark-ups stay out, a choice made with you before the first model, so the model measures the effect of the media rather than the invoicing.

How the data comes together
SearchAPISocial and videoAPIDisplay and programmaticAPIAffiliate and price comparisonAPITV, radio, print, OOHpipelineSales and business driversfeedOne structured tablesame meaning in every rowcompanymktchannelspendNordicNObrand srch▮▮▮NordicNOTV spot▮▮▮▮NordicSEmeta pros.▮▮NordicSEaffiliate▮NordicDKradio▮▮The modela curve per channelThe same table feeds reporting and the model

Why the grain matters

A TV campaign entered as one monthly total tells the model nothing about which weeks the spots ran. Spot-level data lets the model see the flight pattern, estimate carryover from it, and credit TV's pull-through on search and social in the following weeks to TV. It is also what makes a question like "what did the Saturday prime-time spots do compared with daytime" answerable later, from the data already there.

Collecting and structuring offline spend is part of the managed service, and some customers use Odins for that alone, as the one place where their TV and radio spend sits next to their digital spend in a structure their warehouse can take.

  • Channel hierarchy: The one structure every source lands in: market, product, segment, then channel down to the spot. Digital and offline side by side.
  • Structured marketing data: Every source, one dataset, your hierarchy, your warehouse. The foundation the model sits on, and a product on its own.
  • Adstock: The carryover effect: the part of this week's advertising that lands in the weeks that follow.
  • Long-term brand effect: The slow lift in baseline demand from sustained brand investment, beyond the weeks adstock covers.

Go one level deeper: TV, radio and offline media connectors. Measuring TV, connected TV and YouTube with MMM. Or go back to all terms in the glossary.

FAQ

Frequently asked questions.

Do we have to upload TV spend every month?

No. Odins sets up a pipeline from you or your media agency, so plans and spot reports flow in as a standing feed rather than a monthly chore.

What if our agency only has monthly totals?

Spot reports exist for nearly every TV and radio buy; the agency or the broadcaster has them. Where only totals exist for the history, the model starts from them with wider ranges, and the pipeline captures spot-level data from then on.

Does the model treat TV like a digital channel?

On the same terms, yes: same hierarchy, same grain, same saturation curve and carryover. The difference is in the estimates, not the method; TV typically carries over longer and saturates later.

Talk to our team

See it on your own numbers.

A 30-minute walkthrough of how this plays out for channels like yours, on a model built on your data.

Book a demo