.png){kind=icon}
Apple’s privacy thresholds in SKAdNetwork have been obfuscating conversion values – and preventing advertisers from getting an accurate read on performance, especially on Facebook. Facebook recommended ensuring that you hit at least 128 installs a day per campaign so you can overcome Apple’s privacy thresholds.
Will hitting the 128 install daily threshold solve your problems? How does this privacy threshold apply to channels other than Facebook – ad networks and SANs?
We try to answer these questions in this episode, based on what we are observing.
ABOUT ROCKETSHIP HQ: Website | LinkedIn | Twitter | YouTube
KEY HIGHLIGHTS
💍 What happens if you do hit 128 installs a day per campaign?
🧐 The difference between Facebook and other networks.
🧶 Why do other networks behave differently from Facebook?
💪 How are we responding?
FULL TRANSCRIPT BELOWToday we follow a thread from one of our recent episodes: What’s working now post ATT/iOS 14.5 in mobile user acquisition: 6 opportunities to address performance deterioration. Facebook has recommended setting up campaigns so that you hit at least 128 installs a day, so that these campaigns can overcome Apple’s privacy thresholds and you can actually see conversion values.
For many smaller advertisers, 128 installs a day per campaign can be expensive and risky. However, we were able to test this with some of the larger advertisers that we work with. We asked – does hitting 128 installs a day solve all your problems? Do you then see all conversions, at least as reported by SKAdNetwork?
The answer, my friends, is disappointing.
Yes, getting to 128 installs a day *did* make a difference. A number of campaigns that had zero conversion values started to see non-zero conversions. *However* these conversions were still so few that they just weren’t meaningful to draw performance conclusions from.
For instance, on a couple of accounts we saw thousands of dollars in spend with 2 or 3 purchase conversions, which was very poor compared to their erstwhile day 0 CPuP of $150 or so. (And we did benchmark SKAdNetwork reported performance metrics against day 0 metrics from the past – to allow for the signal loss due to Apple’s system of timers.)
This makes it extremely difficult to evaluate performance. Especially, if you are required to commit to hitting 128 installs a day and spend thousands on a campaign, only to fly blind.
**
What makes this even more curious is that this privacy threshold is NOT a problem on other SANs or ad networks. On Snap, TikTok, and ad networks, we do see the cost per unique payer (or cost per conversion) to be comparable to the erstwhile day 0 cost per conversion. It appears that there is not a lot of performance degradation on these platforms as compared to Facebook.
What gives? Why is Facebook measurement seemingly broken?
Here is our strongest hypothesis. As we’ve spoken about in our workshop Mobile Growth Lab, because Facebook has limited advertisers to 9 campaigns per account, Facebook is likely setting up campaigns ‘under the hood’ that it is dynamically using for creative testing and iteration. By dynamically setting up these black-box campaigns and using them behind the scenes, Facebook can stay under Apple’s required 100 campaign limit, while exposing only 9 campaigns to advertisers.
If the 9 advertiser-facing campaigns are actually 100 Apple-facing campaigns, then each advertiser-facing campaign will need to hit 11 times as many installs and conversions as an Apple-facing campaign.
What we hypothesize is that ad networks and other SANs such as Snap and TikTok right now have their advertiser-facing campaigns to be similar to Apple-facing campaigns. As a result, conversion values on ad networks and other SANs are obfuscated far less than Facebook’s are.
Given that Facebook’s conversion tracking is not even directionally accurate, these are some of the things we recommend:
a. Treat your other SAN and ad network performance metrics as close to accurate(as comparable to d0 CPAs).
b. On Facebook, optimize for installs for now – and target getting to the lowest cost installs, so you have at least some confidence in the numbers that you are seeing. Don’t compare Facebook reported SKAdNetwork metrics vs. other channels – because these are not comparable.
c. As we’ve said before, lean on blended numbers as your source of truth.
These aren’t perfect solutions – because we aren’t living in a perfect world just now, especially with the broken measurement post iOS 14.5.
BUT I hope this episode helps you understand that scaling to 128 installs a day per campaign isn’t the only path forward – if anything, that is a very risky path. I’d recommend looking at your other options to get as much visibility and transparency as possible under the circumstances.
A REQUEST BEFORE YOU GO
I have a very important favor to ask, which as those of you who know me know I don’t do often. If you get any pleasure or inspiration from this episode, could you PLEASE leave a review on your favorite podcasting platform – be it iTunes, Overcast, Spotify or wherever you get your podcast fix. This podcast is very much a labor of love – and each episode takes many many hours to put together. When you write a review, it will not only be a great deal of encouragement to us, but it will also support getting the word out about the Mobile User Acquisition Show.
Constructive criticism and suggestions for improvement are welcome, whether on podcasting platforms – or by email to shamanth at rocketshiphq.com. We read all reviews & I want to make this podcast better.
Thank you – and I look forward to seeing you with the next episode!