Mobile Attribution: How to Avoid Mismeasurement

Mobile Ad Attribution

Data measurement and analytics reporting are at the core of every successful marketing initiative. While desktop remains the incumbent device for digital advertising, there has been a prominent growth shift towards mobile websites, ads and apps. Just five years ago, mobile browsing and app usage were budding concepts, but according to Yahoo’s Flurry Analytics, 90% of consumer’s mobile time is now spent in-app.

The mobile environment is founded upon last-click attribution, but advertisers use several ad networks and platforms to reach users, complicating the attribution process with so many touchpoints. Unlike on desktop, mobile attribution doesn’t have the luxury of using cookies or pixel tags to identify when a user executes an action. As a result, advertisers need to identify trusted attribution platforms that have a bird’s eye view of the click-to-install path across all integrated networks. Correct attribution will ultimately save advertisers money by preventing multiple charges by ad networks. Below are some recommendations for advertisers to get the most accurate attribution from the partners and networks they work with.

Use location-based data.

In order to measure the impact of your brand’s online ad, use location-based data to analyze and determine offline and in-store purchases. Companies, such as Ninth Decimal, leverage location data for accurate cross-channel attribution. This technology is possible because attribution vendors can aggregate data from all of the devices that were reached during a campaign, and then later cross-reference it against devices that visited the location again.

Be wary of double attribution and multiple charges per user action.

Double attribution occurs when two or more sources are attributed with credit for the same app installation, meaning that a marketer can be charged upwards of two times for the same action. This is common with marketers who work directly with an ad network because measurement begins once the SDKs (attribution analytics tools) is associated with the respective media partner.

However, media companies don’t have a holistic view of the user’s path to conversion or have an understanding of other networks’ activity. In other words, when a mobile user clicks on network X’s ad and downloads an app, then network X charges the advertiser based on the cost per install (CPI) model. Since the mobile ecosystem is dominated by last-click attribution, an advertiser should only have to pay for ONE last click. Not only does double attribution waste budget with multiple charges, but it can actually negatively impact an advertiser’s ability to optimize performance. Instead of relying on an ad network’s attribution model, work with an independent unbiased mobile measurement platform because they have integrations with hundreds of media partners, have a ‘bird’s eye view’ of the ecosystem and rule on last-click attribution.

Know the difference between biased and unbiased attribution partners.

There are two governing mobile attribution bodies: end-to-end solution providers and third party unbiased attribution-only providers (such as Kochava). End-to-end solution providers are companies that buy media, manage campaigns, measure and optimize them. Since this type of company is intrinsically motivated by driving revenue, attribution is a secondary concern.

Contrary, an attribution-only provider is in business to solely regulate attribution, appropriating their efforts to fighting for transparency and reliability between advertisers and networks. As such, it’s in an advertiser’s best interest to work with attribution partners to prevent inherent conflicts of interest.

Don’t rely on deep links for Facebook attribution.

Facebook is undoubtedly a dominating force in the mobile ad space, accounting for more than 50% of the marketing share. The platform allows advertisers to use deep links for mobile apps, essentially removing extra steps between a user clicking on an ad and getting to the ad’s content that triggered their interest. Without a deep link, users would have to manually search for content they’re interested in, and the advertiser risks user drop-off. In the mobile ad attribution space, deep linking only offers insight into where the user came from but does not explicitly show time-based attribution (such as last click or first click).

Instead, advertisers can measure app promotion by embedding its SDK or working with one of Facebook’s mobile measurement partners (MMPs). Often, advertisers opt to work with a trusted mobile ad attribution company instead of using a network’s SDK for measurement. With this approach, the network (in this case, Facebook) is dependent on the third party measurement provider to send the data. If a third party attribution company is not involved, then the data is instead sent from the network to the attribution provider.

The mobile advertising space is largely dominated by measuring engagement rather than acquisition. While there is no one-size-fits-all approach to achieving mobile measurement at scale, it’s important that advertisers embrace testing new methods and technologies to ensure post-install and post-click marketing analytics are accurate.