Issue 338

March 31, 2021


Growth

Diagnosing Incrementality Issues on Facebook: Benchmarking View-Through Conversions as a % of Total Conversions

When I audit FB ad accounts, I always look for tell-tale signs that they might have an incrementality problem. Any time view-through conversions account for a large percentage of attributed conversions from prospecting campaigns, I rush to run an incrementality test. In this post I did a quick analysis of 10 US, D2C brands to try to establish a benchmark for the ratio of view-through / click-through conversions. So essentially, if view-through conversions are over indexing (say 40%+) it's likely that an incrementality test will tell a very, very different story than what the pixel reports.BLOG.THESISTESTING.COM

3/31 UPDATE – Not RIP for Holdout Tests on FB?

Despite what the FB UI + this help page would suggest, multiple FB reps have told me in the last week that holdout testing will not be going away entirely for web advertisers, though it will likely be severely limited. I'm still waiting for them to either update this Help page or provide a separate update, but I haven't seen anything yet.FACEBOOK.COM

FB Rep on How the 1-Day Click Opt-Out Data Will Work in Ads Manager

It's clear from prior announcements that opt-out users' data will track only the highest priority event. And it's clear that this data will use a 1-day click attribution window and will be delayed + modeled. The new info (or at least what I think is new info) we received this week from a FB rep was that "1-day click opt-out data will only be included in Ads Manager reporting totals if your campaign/ad set is optimized for 1-day click (with or without 1-day view). Opt-out data will not be included in 7-day click reporting." And so that implies a tradeoff between "Optimizing for 1-day click and receiving opt-out data but no data for opt-in/non-iOS customers beyond 1-day post-click or Optimizing for 7-day click and receiving all available data for opt-in/non-iOS customers but no data for opt-out customers." I'm going to keep looking to find this confirmed in the actual FB docs.FACEBOOK.COM

‘A V-Shaped Recovery’: Slow Recovery Ahead for Mobile Marketers Post-Apple Tracking Crackdown

"“We expect eCPMs to deflate by 35% on non-IDFA supply,” said Sergio Serra, head of product management at InMobi. “While we must acknowledge a monetization deceleration for iOS pubs is likely to extend over the next few months, we are also positive a higher number of delivered impressions will generate similar absolute revenue overtime. Such an assumption particularly holds true for performance buyers.”DIGIDAY.COM

2021 App Retention Benchmarks

These retention rates seem SUPER high to me. They have the *average* day 30 retention rate at 66%!!! Appsflyer's data from last year had day 30 retention at less than 5% in 2020 & 2019 (https://www.appsflyer.com/resources/app-retention-benchmarks/). Maybe they are using a different approach to the retention calculation?APPTENTIVE.COM

CRM

Join the Grow.co Weekly Newsletter Which Consistently Opens at +45%

Why is the open rate so high?

We think it's because we take the time to curate the best articles on acquisition and retention marketing we can find.

Plus, you'll get access to job listings, our own original content and exclusive invitations to our roster of events.

We respect your privacy.
We'll never share your address and you may unsubscribe at any time.