Nowadays, smartphone users, on average, use only 30 apps each month. Therefore a big challenge that mobile marketers face is retargeting their current and lapsed user base. As programmatic advertising becomes more popular, various strategies and methods have been devised to address these challenges - one of them being incremental lift analysis.
An incremental lift test is used by advertisers who want to measure the effectiveness of retargeting campaigns and help them decide on optimal strategies.In this test, the targeted audience segment is randomly split into two groups, the treatment group and the control group. Treatment groups are the users who are retargeted and shown a brand-related ad. The control group is shown a public service announcement (PSA) ad or no ad at all, depending on the advertiser’s strategy. In terms of audience size, most advertisers allocate 20% as a control group, although some use 50% to immediately gather data.
Methodology
There are two widely used approaches in running an incremental lift test. The first is Intent to Treat (ITT), and the second is serving a Public Service Announcement (PSA), or a placebo ad.
ITT is where the control group doesn’t see any of your ads. Most advertisers use this approach because it costs less and is easier to implement. Although, this strategy has a lot of noise, and a like-for-like comparison is not possible given that not all the users from the targeting group will be included in the analysis. A high match rate and good inventory supply are essential to this strategy since you’re only including data on your targeting group who have seen the ad. It’s not like a control group wherein all users are included in the analysis.
On the other hand, the PSA or placebo ads are where both the treatment and control groups see an ad. The treatment group is shown ads that are specific to the app, and the control group is served with a PSA ad. In terms of running the analysis, this approach is easier as all events and performance are being tracked in real-time by both the network and the advertiser.
Case Studies
To measure the effectiveness of the retargeting strategy we ran incrementality tests on two of our top-performing social casino clients - a bingo app and slots casino app.
We ran a PSA test for the bingo app by dividing the audience into two groups and showing respective creative ads to understand the lift and incrementality of the retargeting campaign.
For the slots app, we ran an ITT and used a holdout group that weren't shown any ads. Our key purpose was not only to see the lift and incrementality, but also to measure what group of users we would have better results with. This approach enables us to understand better what group of users to retarget.
In terms of control and target group size, we used 6% and 20% respectively. As early as 2 days have lapsed, and retargeting users improved app revenue.
Metrics:
- Lift - % likelihood that a user will purchase if the ad is shown
- Incrementality - actual % of purchases received because the ad was shown
We were able to see positive results as early as 2 days lapsed given that for other app categories mostly longer lapsed periods are generally used (D7, D14, D30, etc). The below table shows the revenue lift and revenue incrementality of the slots casino app per different lapsed day ranges:
Lift Per Lapsed Day Range (Slots)
Aarki also retargeted lapsed users who have been inactive for more than a year and still saw positive lift and incrementality. This is proof that retargeting campaigns help to improve user retention and engagement.
Take Action
If you’re running out of strategies to retarget your mobile app users, try elevating your campaigns through an incremental lift analysis. It’s a great tool to use when deciding the optimal strategy that will work for your brand. Want to learn more? Shoot us a message here.