Is Apple Search Ads Conversion Data Reliable?

Incipia is back to the marketing geekiness, this time with an analysis of Apple Search Ads campaign conversion trends. Specifically, we investigated the average conversion lag of Apple Search Ads campaigns over a 1-week time period, as well as the average discrepancies in conversions reported by Apple Search Ads, vs installs reported by an MMP.

In the conversion lag investigation, we looked into trends on how campaign CPA changed after the day that a user's tap was paid for, across 10 accounts. We discovered the following interesting tidbits:

• CPA decreased in the first and second day but did not seem to continue decreasing after that point. That is, the CPA from 5/5/17 would decrease on 5/6/17 and 5/7/17, but would not continue decreasing on 5/8/17.
• CPA always decreased, but sometimes for only one day, not two.
• The average total decrease in CPA was ~41%. That is, a CPA of $10 recorded on day 0 could be expected to decrease to $5.90 on day 2.
  • The average decrease in CPA on day 1 vs day 0 was ~36%.
  • The average decrease in CPA on day 2 vs day 0 was ~18%.

Apple does record conversions for up to 30 days from the original tap, meaning that CPA very well could continue to decrease for up to a month. But it seems that the majority of conversions (per our 1-week study) will arrive in the first two days after the tap.

Moving on to the comparison of conversions as reported by Apple Search Ads vs an MMP, we studied trends from five accounts (for a week's cohort that was more than 3 days past day 0), and the results were pretty astonishing. Surprisingly, not even branded keywords were safe bets!

• Overall average Apple Search Ads conversions recorded = ~56% more than MMP
• Branded average Apple Search Ads conversions recorded = ~65% more than MMP
• Exact match average Apple Search Ads conversions recorded = ~48% more than MMP
• Broad match average Apple Search Ads conversions recorded = ~72.19% more than MMP
• Search match average Apple Search Ads conversions recorded = ~60.63% more than MMP

Some reasons for why this discrepancy exists (per an AppsFlyer support thread) include:

• Apple reports on downloads (possibly even just download button clicks), while MMPs typically report on actual installs or even first-opens.
• Apple's reportedly delivers data with a delay to MMPs, which can cause issues in accuracy.
• Users with LAT enabled; reportedly, Apple counts these installs but does not pass install data to MMPs.

Our takeaway from this study is that it really underscores the importance using an MMP to measure Search Ads performance, rather than solely relying on Apple's conversion data.

If you have comments to contribute, tweet @incipiaco  and we'll include your comments here!

That's all for now, folks! Be sure to bookmark our blog, sign up to our email newsletter for new post updates and reach out if you're interested in working with us.

Incipia is a mobile app development and marketing agency that builds and markets apps for companies, with a specialty in high-quality, stable app development and keyword-based marketing strategy, such as App Store Optimization and Apple Search Ads. For post topics, feedback or business inquiries please contact us, or send an inquiry to hello@incipia.co.