Research: How Should You Follow Up a Mailbox Full Bounce?

A few weeks ago, we shared our research on different types of bounces and their implications for senders. One of our most notable findings pertained to Mailbox Full bounces — specifically, that almost ⅓ of recipients with a Mailbox Full bounce opened another message within the next 12 months.

The surprising results sparked conversation in the industry, with senders and providers debating how to handle Mailbox Full bounces in light of the data. Rumors have also been swirling of Gmail making changes to the way they handle Mailbox Full bounces, so we thought a deeper analysis of the results by ISP was in order.

Gmail

Since Gmail makes up the majority of the average marketer’s list, we placed a special focus on their data. Our data set included millions of Mailbox Full bounces from Gmail recipients — but how many of those addresses were likely to re-engage later?

After seven days, just shy of 8% had opened a subsequent email — not far off from what we saw in the original results. By one year, though, the percentage of recipients who opened after Mailbox Full had risen to over 38.5%.

Given Gmail’s bountiful free email storage, industry advice often asserts a full mailbox there is likely to have been abandoned. Even so, almost 40% of Gmail recipients with a Mailbox Full bounce opened another email within the next 12 months.

Microsoft

Like Gmail, Microsoft provides 15GB of free email storage for all users, so should we expect to see similar results? Let’s look at the data.

Somewhat surprisingly, Microsoft recipients were even more likely to respond after a Mailbox Full bounce. A week after the bounce, over 15% of the recipients had opened another message; a year later, that number was over 50%. My theory is that many of these are “O.G.” Hotmail users who have years’ worth of saved-up email, but still use their accounts regularly. For what it’s worth, our outbound data indicates Hotmail.com still gets the highest volume of mail of any Microsoft domain (by far).

Verizon Media Group (Yahoo/AOL)

Although Gmail was the first to increase email storage space, VMG currently leads the free storage pack with a whopping 1TB of email storage provided free to all recipients. With that much storage, how often could their recipients possibly have a full inbox?

According to the data, not often. We saw only a few thousand Mailbox Full bounces during our research period, a fraction of what we found for Gmail and Microsoft domains. It may take an astronomical amount of mail to fill a VMG inbox, but VMG recipients  who bounced with Mailbox Full were much more likely than others to correct the situation quickly. An impressive 30% of the bounced recipients opened a message within a week, and over 75% engaged within a year of the bounce.

Everyone Else

The ‘Big 3’ providers may make up the bulk of your list, but there are still lots of other domains out there that you need to successfully reach. Since bounce-handling methodology can vary pretty widely across the remainder of the mailbox providers, we lumped them together for our final analysis.

Week One saw around 8.7% openers — better than Gmail and the overall average — but at one year, just over 29% had opened.

Putting it all together

In the last article, we advised senders to avoid the urge to suppress Mailbox Full bounces after a single occurrence. Further analysis of the data supports that assertion, especially for the Big 3 providers. Suppressing Mailbox Full bounces too soon is all but guaranteed to suppress active users who are likely to open in the near future, but continuing to send to them could also damage your sender reputation. Our recommendation remains one of moderation: for typical senders who mail recipients 1-3 times each week, consider suppression if you see Mailbox Full bounces and no successful deliveries for 2-4 weeks.

If you use an ESP, we’d also recommend confirming whether or not they automatically suppress “soft” bounces — these often include Mailbox Full and can occur after as few as three bounces. These rules are typically designed with good intentions, but could be using outdated assumptions about Mailbox Full bounces. Many ESPs with these rules allow flexibility to modify or remove the suppressions based on user input.