Both ROAS and incrementality measure retargeting campaign results, but their methodology and goals are different.

With attributed ROAS, we’re looking to receive as many actions as possible within a click or view-through window. In this case, it’s best to target users who are already likely to convert and push them to convert even more.

Incrementality is a long game. In this case, it’s best to ignore users who are already converting in the app or likely to convert. Instead, we should target lapsed users who have previously converted, but maybe haven’t made a conversion in awhile. The goal is to turn these users into repeat converters and increase their LTV.

It’s clear already that the strategy for these two goals is quite different. When running a campaign, it’s important to choose one KPI and stick to it. Are you looking to increase app opens from active users? Attributed ROAS may be the best strategy. Do you want to boost in-app purchases from lapsed payers? Incrementality is the best way to truly know if your campaign has worked or not.

Some of our clients also run retargeting campaigns with Facebook, but we don’t recommend running retargeting with only Facebook. Facebook makes a good complement to other personalized retargeting strategies, but there are a number of issues with using a self-serve platform as your sole retargeting platform. For one, performance data on Facebook is often overestimated. In the documentation of Facebook’s platform, it’s written that “multiple actions count as a click. Any engagement action counts as a click. As an example, if someone likes, comments on, or shares your ad, this action counts as a click and the conversion will be attributed to that date.” What this means is that any engagement at all with your ad is counted as a click and there is no difference accounted for between the different types of engagement (click, view, like, reaction).

Also, Facebook relies on user IDs rather than device IDs. Users often have more than one device linked to their Facebook account, which may mean that users that appear lapsed on their account may actually be active on a different device, and vise-versa. This can further skew retargeting performance results. A dedicated retargeting partner has the ability to reach users in a different environment who are not reachable via Facebook, and users who aren’t responsive to Facebook ads.

Finally, a dedicated retargeting partner is capable of offering more flexibility with creatives and strategic recommendations. At Adikteev for example, we offer custom creatives in all formats (including playables) free of charge, and advise on the best formats for each app to boost engagement. We also offer a custom pre-launch analysis to give app marketers a 360° view of their app’s needs.

Sharing organic data with your retargeting partner ensures that your campaign will be optimized for user behavior instantly.

We recommend evergreen retargeting because for high paying users, it’s essential to continue reminding your users why they should return. Users don’t churn only during holidays and special occasions. Because user churn is something that happens constantly, it’s essential to maintain a robust re-engagement strategy that will keep users in the app all year round.

For gaming apps and some ecommerce apps, bursts and promos can work well for high-value purchasers. However, if you’re just highlighting new features in the app that have been recently released in an update, performance can be just as good evergreen as it is for only bursts and promos.

There’s no one-size-fits-all strategy for retargeting. Each app has its own particularities and audience behavior. It’s essential for app marketers starting retargeting to get a clear sense of what their app’s needs are before deciding on a strategy. At Adikteev, we offer a free prelaunch analysis for every app we begin working with.

That said, there are some commonalities for different verticals. One thing that retargeting cannot do is create new user behaviors. For example, if you have a user who downloads your app and then never interacts with it again, it’s going to be impossible to encourage this user to become a purchaser. However, if you have an app user who has made at least one purchase at some point in their user journey, and now have not made another purchase, it’s going to be much more likely that this user will be influenced to make another purchase with retargeting. Lapsed purchasers such as these are often good candidates for retargeting as they are also the most incremental: it’s the easiest to measure the impact of a retargeting campaign on these users over time with an uplift test.

Another common strategy that’s often a bit counterintuitive is the idea of going after active users. As mentioned, user churn is constant, and maintaining the right level of engagement to keep active users interested is a delicate balancing act. By treating active users as a sure bet and not retargeting them at all, app marketers risk losing them to churn. A smart retargeting strategy will keep these users active and engaged in the app, without causing ad fatigue or annoying them with too many reminders.