iOS 15 Changes How We Have to Track Newsletter Engagement

The way media companies validate the success of newsletters has to evolve. Historically, we have used open rates as our primary metric, but this is becoming increasingly complicated. Unfortunately, many media operators I know are not spending more time looking at this, and it will impact their businesses.

In September, Apple rolled out iOS 15, including its Mail Privacy Protection. This did two things for user privacy. First, it masked IP addresses, which was annoying but manageable. Second, it opened all images up in the background, which completely screwed up open rates.

Here’s how… We would drop a 1×1 pixel in the newsletter to track an email open. When the newsletter opened, the image would “load,” which would tell us that the user had opened the email. However, if Apple is opening images up in the background—irrespective of whether the user has actually opened the email themselves—we’re seeing opens that are not actively opened.

This has resulted in significantly inflated open rates. For example, if before iOS 15, a reader using Apple Mail opened 40% of the time, they’ll now look like they’re opening 100% of the time. Extrapolate that out, and it can start to have a material impact on your opens. As time has gone on, I’ve seen inflation larger than 10%, and for some publishers, it can be much higher.

The good news is that this only affects readers using Apple Mail products. If a reader uses Gmail, even on an Apple device, we’ll be able to track their opens correctly.

So, why does this matter to begin with? There are a few reasons worth exploring.

First, if publishers sell their ads on an open rate, if that inflates by 15%, the publisher may charge 15% more. However, the number of ad views isn’t changing, which means the advertiser is not getting what they paid for. This can impact performance, which means the advertisers will walk away. This isn’t the worst thing because you can look at your non-iOS 15 subscribers and use that as the baseline for open rates.

Second, it can interfere with your ability to keep your email lists clean, which is vital for deliverability. In July 2020, I wrote:

The first is a poor sender reputation. Internet service providers assign a score to each sender to determine how reputable we are. Unfortunately, that score takes time to build, and it can slowly deteriorate as time goes on.

A variety of factors play into your score:

– The number of emails you send
– The number of times your content is marked as spam
– The open rates and CTRs
– Responses
– Hard bounces
– Having domainkey identified mail (DKIM) authentication

The ISPs will know the actual open rate even if publishers see an inflated rate. That means that subscribers might not be opening the newsletter, the publisher thinks they are, but the ISP dings the publisher for a lower quality subscriber. This is problematic because deliverability will suffer if an email list is sent to many people that are not opening.

Without a viable open rate, we cannot comfortably churn. How do you know if someone is really opening versus appearing to open due to Apple’s update?

The third reason is that it interferes with your growth experiments. Whenever we identify a new possible source for newsletter sign-ups, it is essential to track how that source performs. Are readers more engaged or less engaged? That can inform whether we invest more time and money into that source. But, without the data, you’re operating a bit more blindly.

All three of these are obvious problems. And it means that publishers are going to have to evolve some of their strategies to make sure they are getting better data. This means we need to refocus on the click.

There is one universal truth when it comes to email. The only way for a user to click something in an email is to open it first. Whether Apple screws around with all sorts of open rate data, you can’t click something you don’t see. The instinct, then, would be to start loading the newsletter up with links so that people click more, right?

Not exactly. I’ve long been an advocate for newsletters to be treated as products in and of themselves. A user should read the newsletter end to end like a newspaper and walk away with the information they want. This is why newsletter-first media companies have become so successful over time. The email is an incredibly high-quality product.

So, we have to balance farming for clicks and delivering a high-quality product that users want. Here are a few ideas that could be helpful.

Ask the user questions. I’ve seen some newsletter operators put a question at the end that asks if the reader enjoyed the newsletter. Another type of question you can ask is relative to the content. For example, if it’s a b2b story, maybe ask: “Does this impact your business?” Finally, I’ve seen some publishers have a daily survey at the bottom. This can help get users to click, but if done right, you can also capture more data about your audience that could be helpful for other initiatives.

Unfortunately, this may not be enough. Most people don’t click. Even with an increase in options to click on things, users still won’t click.

In this case, we have to get more aggressive and start explicitly asking users to click. For example, for users who show a 100% open rate and a 0% click-through rate, we should message them asking them to click something. This can get a little murky, but I think it’s the most important thing we can do. Here’s simple copy you can try: “To keep our email lists clean, we remove people who haven’t clicked in 30 days. Do you still want to receive this newsletter?”

In this case, you want to give them as much opportunity to click on the link. Perhaps you run it for 1-2 weeks every single day. If they then don’t click, you should unsubscribe them. Here is how I would set up the progression:

• 0-30 days: Track people who haven’t clicked
• 31-45 days: Explicitly ask a user to click a link saying they still want to receive the newsletter
• 46th day: Unsubscribe

You can get more aggressive or less aggressive; the choice is ultimately yours. But you want the user to see a prompt in the newsletter asking them to click. If it’s in their face enough, hopefully, they’ll click.

We must keep our email lists clean. Before iOS 15’s introduction, we could easily track open rates and use that as the metric for churning subscribers. Now that the open rate is artificially inflated for iOS 15 users, we can’t depend on it as much. However, the open rate still works for those who don’t use Apple products. For how long, though, remains unknown.

As with all things in media, we have to evolve. For years, open rate worked. Now, it’s not the perfect metric. But to simply ignore these changes would be irresponsible. It’s more challenging, but we need to become more innovative and understand different engagement mechanisms to understand our audience’s use of our newsletters. If we do that, our newsletters will be healthy for years to come.

Thanks for reading today’s newsletter. Please hit reply or join the AMO Slack channel if you have thoughts. I hope you have a great weekend!