Illuminating the carryover effect in TV advertising with Single-Source data

Author: andrew

Published: 15th June 2022


Every TV advertiser wants to measure the effectiveness of their TV campaign. But in order to do this – and efficiently decide on how to split budgets between different advertising channels and optimise their media mix – they must understand its true performance.

Many advertisers still rely on the 5-minute attribution model to measure the performance of their TV campaigns. In a previous article we proved that only 1% of traffic is generated within 5-10 minutes of a TV ad, meaning it’s highly inaccurate to just use this data alone for campaign optimisation as you don’t get the full picture. Furthermore, these models always undersell TV advertising as they only show a small part of the effect the TV campaign has on a brands’ KPIs.

We also demonstrated that the right window in which to measure TV effectiveness is one week which captures the majority of the traffic driven by TV advertising.

But even this is not enough to understand the full power of TV.

Gaining richer insights with Single-Source data

Using Single-Source data, which maps consumers’ online and offline behaviour, including viewing habits, location, browsing and purchase data, over a period of time for the same individual, it is now possible to unlock and see the full effect of the campaign. This includes the effect of carryover after the campaign has ended.

The carryover effect refers to a delay between consumers being exposed to advertising and responding to the advertising. For example, after a TV campaign airs, some customers might respond to the ad immediately, but for others it might take a week or a month.

ViewersLogic extracts this information from our consumer panel of over 8000 people. This opt-in database allows brands to determine the consumer’s journey to purchase to understand how media affects it.

We studied how ASDA’s TV campaign affected website visits during the final week of the campaign compared to 4 weeks after it aired to discover the carryover effect on consumers.

The results? 

The findings showed us what we expected; TV has a branding effect that does not end the day the campaign ends, but rather has a strong carryover for a few weeks post-campaign.

As you can see from the above chart, there was a 22% uplift in ASDA site visits during the last week of the campaign among the exposed group. In the 3 weeks after the campaign ended, they still had a 15% higher chance of visiting ASDA’s site and this dropped to 8% in the 4th week after the campaign ended.

The carry over effect was the missing component of understanding ASDA’s true campaign performance. With this data those responsible for the advertising budget can be more confident and precise in their funds’ allocation, deliver more bang for their TV buck and reallocate funds to complementary channels without reducing the impact of TV.

To find out how you can get this level of control over your media budgets, get in touch with us here:



We used a natural A/B testing methodology to compare a test group to a control group and successfully identify the effect of TV on different KPIs.

  • Control Group: Individuals who were not exposed to TV activity from the brand in question during the campaign period, in the 2 weeks prior to the campaign and in the 4 weeks after the campaign ended.
  • Test Group: Individuals who were exposed to TV activity from the brand in question during the campaign period but not in the two weeks prior and in the 4 weeks after the campaign ended.