Approach to Linear TV Attribution

Description: This article describes our approach to linear attribution and outlines how the probabilistic spike model works vs. the linear impression-based model.


Introduction

Innovid’s unified approach to linear TV attribution combines different models to deliver unparalleled accuracy and insight into overall TV performance for advertisers.

Our linear spike model measures the immediate response to traditional broadcast linear campaigns and works for any advertiser regardless of market size or campaign spend. It uses linear ad occurrence data in conjunction with online response data.

Using household-level data, our impression-based model measures the delivery and impact across linear and streaming campaigns. It uses a range of impression-based datasets, including ad serving data from certified OTT/CTV publishers and platforms and online response data.

Our platform is a useful tool for marketers looking to improve their understanding of the impact of their advertising campaigns and optimize their media spend accordingly.


Probabilistic spike model

Our spike model is used for broadcast linear campaigns. It works by assigning each TV spot a weighting based on the probability it drove a viewer to engage with the advertiser’s website or app. 

How spike model linear data is processed

We use a three-step process:

  1. Collect data and track response
    Website: Our first-party Response Tracker collects data from the advertiser’s website from visits through to onsite actions and sales.
    App: The advertiser activates InnovidXP as a conversion endpoint through an app partner, and this integration brings data through to the platform.

  2. Calculate and generate the baseline
    Continuously measure website traffic.
    Filter out traffic that is highly unlikely to have been driven by TV.
    Evaluate the baseline for every market, every minute of every day.
    No need for historical data.

  3. Attribute and assign response
    Isolate and attribute above the baseline activity.
    Assign a probability score to each user session.
    Deal with overlapping spots.

The diagram below outlines this process:

Spike model data is processed daily - modeling is run on all spots aired in the 24 hours since it last ran, and results are published in the platform. This processing deals with any overlapping spots, i.e., spots that are aired at the same time on different channels.

The example below outlines this process: 

Spike Processing Example
Example weekday Description Always on
Before spot airs Evaluate the baseline. Continuously measure and filter website traffic.
Wednesday, June 1st Spot airs. Run every day.
Period in between spots Calculate the lift of activity above the baseline after spot airs. Deal with overlapping spots.
Wednesday, June 8th Spot airs again.  
After each spot airs

Attribute above the baseline activity.

Assign a probability score to each user session.

 

Impression-based model 

Our impression-based approach uses granular viewership and ad occurrence data sourced from smart TV data providers (Inscape or Samba), ACR, post logs, our ad server, or integrations with third-party OTT/CTV publishers and platforms; alongside response data from the advertiser website and/or app. 

Once we have captured all the impression and response data, we cleanse, filter, and model it, mapping viewership to households and online responses. From there, data is put through numerous extrapolation processes to ensure it provides an accurate representation of the total TV viewing population. The final result is actionable reach, frequency, incremental reach, and outcome insights.

How impression-based linear data is processed

We use the following process:

  1. Collect smart TV, ad airing, and website/app response data.
  2. Cleanse and filter the data.
  3. Process 1:1 matching.
  4. Calculate TV Influence (i.e. we use a control group method to isolate TV uplift).
  5. Upscale the results (because the Inscape/Samba datasets only cover a small, but representative. fraction of the total TV viewing population).

The diagram below outlines this process:

Impression-based linear data is processed weekly. All linear impression-based spots undergo processing twice since this method only processes the previous two weeks’ worth of data. After roughly two weeks, you can view provisional impression-based results. These are usually finalized around three weeks after spots have aired.

The example below outlines this process:

Impression-based processing example
Example weekday Description
Wednesday, June 1st Spots are aired.
Thursday, June 2nd

Spots from the day before are uploaded. The processing for these spots happens on the following two Mondays.

Note that the traditional linear results will be available, as the spike model processes daily.
Monday, June 6th “Processing 1” begins. Impression-based spots from June 1st are attributed, but only up to Sunday, June 5th. This is a 5-day attribution window at this stage.
Monday, June 13th “Processing 2” begins. Impression-based spots from June 1st are attributed again, extending to a 7-day window up to June 7th.
Wednesday, June 1st Spots are aired.
IMPORTANT: Each processing runs for the previous two weeks and usually takes several days. This means results are finalized up to three weeks after the spots are aired, although interim results are displayed on the platform before this.

Key differences

  • Traditional linear attribution results are available daily via spike model processing
  • Linear impression-based attribution results are provisional after two weeks and finalized after three weeks (these are approximate timelines) via a data attribution panel
Approach Probabilistic Spike Impression-based
Description Measures the performance of traditional broadcast linear campaigns. Measures delivery and impact across linear and streaming campaigns using a consistent count.
Types of Insight Immediate impact of TV.

Reach & Frequency.

Incremental Reach.

Immediate and Longer-term Impact.

Attribution Window Minutes/Days. Days/Weeks.
Data

Ad occurrence data.

Online (web/app) visit data.

Ad occurrence data.

Impression data.

Online (web/app) visit data.

Geo Coverage Global - All markets. U.S., UK, DE.
Primary Use Case
Markets and campaigns where impression-based modeling is unviable. National/regional campaigns with advertisers who have sufficient levels of customer engagement to generate strong impression-based results.

 

Key benefits

Impression-based approach

  • Measurement metrics and insight such as reach and frequency and direct outcomes such as web visits, sales, registrations, etc.
  • Incremental reach analysis of OTT/CTV beyond linear TV and across individual streaming publishers
  • Longer-term impact analytics go beyond the immediate impact of a TV campaign to give a richer picture of TV performance

Probabilistic spike approach

  • Performance insight for any market covering all campaigns where impression-based modeling is not available or limited (e.g., countries outside of the US, UK, and DE)
  • Outcomes and performance metrics for traditional linear broadcast, including granular spot-level insight
  • Measures the immediate impact of TV ads on advertisers’ online web and app activity

Related content

FAQs: Using the Product

Was this article helpful?
1 out of 1 found this helpful