Five tactics to get faster learnings from your Apple PPO tests
- IRINA DAN
- April 26, 2024
IN THIS ARTICLE
SHARE THIS ARTICLE
Three years after its release, Apple’s Product Page Optimisation feature is still far from meeting developers’ expectations. Limited to testing only visual elements (icon, preview video, screenshots) and to only one test at a time, the biggest challenge of Apple PPO tests is that they take forever to show high confidence levels, making it difficult to reach any conclusion about which variant to implement.
Here are some tactics to get faster learnings from your Apple PPO tests:
1. If your app has low traffic, consider sequential testing
When you set up the product page optimisation test, you’ll get an estimate of how long it might take to reach your desired improvement in conversion rate based on recent traffic to your app’s product page. A test can be active for up to 90 days, but who can afford that long? If the estimated test duration exceeds the 90-day limit, consider sequential testing.
While sequential testing is not a real A/B test, it will allow you to gain insights in a shorter time.
2. Group localisations to increase audience sample
HERE’s a list of all languages supported by Apple in each country.
Basically, if you group localisations like EN-US, EN-CA, EN-UK, EN-AU, your test will be exposed to a larger audience and reach a higher confidence faster. The downside to this method is that you won’t be able to segment results for each localisation.
3. For clearer impact, test bigger changes
Test background changes, new design concepts, preview video vs. no video, different imagery.
Avoid testing minor font or copy changes, screenshot order, new screenshots late in order.
4. For more reliable results, run an A/B/B test
While the requirement is for each variant to be unique, you can make minor tweaks and set up an experiment with 2 almost identical B variants. The advantage to this method is that you can compare the results of the 2 B variants after a week and rule out false positives. If both B variants return similar results, then you can skip the B/A test which is normally used to make sure the results are correct.
5. Test the new variant as a custom product page in Apple Search Ads
If you have a stable campaign in ASA (such as Brand), test the new variant as a CPP, making sure to not change campaign settings or bids.
CTR and install rate can be used as indicators to the success of the new product page.
Unfortunately, you can’t use this method to test app icons. And while it may not be a proper A/B test, it can help shorten the decision time.
Conclusion
None of the solutions above is perfect, but most apps don’t meet all requirements to run perfect PPO tests.
Until Apple allows running multiple tests simultaneously, with the possibility to reach confidence levels in more decent times, we’re left with imperfect solutions to get faster learnings.
LOOKING TO IMPROVE YOUR PRODUCT PAGE CONVERSION RATE?
Every successful project begins with a conversation.