Increasingly, marketers are aiming to move away from legacy, single-touch attribution and tracking mechanisms like first-click and last-click attribution and fully embrace multi-touch attribution models. But what does multi-channel attribution mean and what does a full path model really entail?
A multi-touch attribution model describes any methodology that accounts for the various key touchpoints an individual will see and experience before they take a desired action. In performance advertising, individuals will see a variety of ads from different networks before taking the desired action. Multi-touch attribution enables marketers to account for all of these ads when determining who gets credit for this action.
“Last click attribution as the sole measuring stick for digital performance advertising should go away within the next few years,” says Anne Frisbie, Senior Vice President of Global Programmatic and North America at InMobi. “Our belief is that this a very important trend that we hope plays out as quickly as possible. Last-click attribution was great when search marketing first launched at the turn of the century, i.e. compared to no measurement at all. But now, nearly 20 years later, it is clear that last-click attribution is not sophisticated enough and that it misattributes value back to media partners since many organic activities that marketers shouldn’t have to pay for are captured, i.e. they aren’t truly incremental.”
For example, let’s say someone sees a paid search ad first, followed by an ad in app and then finally a social media ad before taking a desired action. Legacy single touch models would assign 100% of the credit to either the first ad or the last ad. But, just because an ad was closer to the conversion doesn’t mean it was entirely or solely responsible for the final action.
Noted marketing expert Neil Patel has compared multi-channel attribution to soccer/football: “Multi-touch attribution gives the credit for a goal to not only the scorer but also (gives some credit to) the players who prepared the goal, too. Soccer player statistics often calculate scores based on the goals and the assists of the players. That means the statistics are based on what could be called a double-touch analysis that takes into account the last touch and the touch before the last one. Since the default model in marketing still seems to be ‘last touch’ only, it looks like soccer has overtaken marketing in terms of analytical sophistication.”
It provides a more holistic and realistic view of the customer experience, as it’s unlikely that just the first ad or just the last ad really and completely drove the final action. Likely, multiple actions were working in concert with one another, and all relevant and important drivers should receive equal credit via a weighted modeling attribution arrangement.
It also provides the ability to measure incrementality and understand customer journeys better (which in turn leads to better spend optimization). And, it helps to reduce fraud, making tactics like ad stacking, click cramming and click sniping far less lucrative for fraudsters.
“The ability to measure quality, combined with a deep understanding of which engagement metrics drive performance, will result in tangible outcomes for brands, transforming how advertisers define value in their media spend,” says Matt McLaughlin, Chief Operating Officer of DoubleVerify.
Less than one in 10 marketers in the U.S. say their company effectively uses and understands data-centric attribution, while close to 35% say current usage is very poor or below average. But, 58% of companies in the U.S. expect to implement multi-touch attribution in 2019.
In 2017, less than half of all U.S. businesses leveraged multi-touch attribution in digital advertising campaigns. Expect that to change dramatically in the future though. Companies worldwide spent $816 million on multi-touch attribution solutions in 2018, but that figure is expected to reach $1.6 billion by 2023.
“It is very clear that clients are beginning to put in place cross-device attribution solutions that enable them to have a much better view across different devices (not just browser based digital placements) on the impact of their media. And, many of these clients are also executing offline sales attribution studies through their partners. We fully believe in cross-device attribution and full-funnel attribution efforts. Overall, the value of high impact in-app mobile advertising opportunities (such as video and other interactive experiences) are more accurately reflected when measured across the full funnel and via cross-device attribution, in-store visits and in-store sales not just last-click attribution or online web sales - so we think this is a very healthy change for the ecosystem,” says Anne.
Work with the right partners to help with both implementation and ongoing tracking. A custom model may be necessary depending on what’s considered a conversion event, whether or not marketing touchpoints need to be tracked in real time, what the customer journey looks like for each persona/segment, etc. There is no one single type of multi-touch attribution that works well for every organization out there.
Also, advertisers should define an “Incentive Compatible” attribution model, where the incentives and the attribution mechanism helps to both identify the most effective channels and motivate those channels to do things right (e.g, holistically fight fraud).
Interested in Learning More?
Send us a note at contactus@inmobi.com to learn more and to determine if multi-touch attribution is right for your next mobile marketing campaign!
For a more detailed overview, be sure to check out this video:
And for even more insights on everything from the time decay model and linear models to weighted modeling, U shaped modeling and everything in between, peruse these blog posts at your leisure:
Name of Beneficiary: InMobi Hong Kong Limited
Company Address：18th Floor, Edinburgh Tower, The Landmark, 15 Queen's Road Central, Hong Kong
© 2024 InMobi 邑盟信息技术（上海）有限公司,
京ICP备17056665号-5
地址：上海市西藏南路228号永银大厦802 电话：021-63302138
