Email ‘Moneyball’: Why Your Goal Shouldn’t Be 100% Inbox Placement

Posted by George Bilbrey 

I recently checked in with a client who told me the company’s biggest goal right now is “getting to 100% inbox placement.”  My reply was, “That’s the wrong goal.”

It may seem counterintuitive, but more is not always better when it comes to inbox placement.

With the rise of subscriber level, engagement based filtering, the deliverability game has changed. In the past, filtering was more binary: either you had a problem (once you passed a certain reputation threshold), or you didn’t.

Now having a deliverability problem is more subtle. There is a complex interplay of inbox placement rate, volume, and cadence, long term open/click rates, and email program design.

Let’s take a delivery problem at Gmail as an example of how these tradeoffs work. Although the filtering process at Gmail is very complex, I find a simple mental model consisting of three layers explains most delivery performance issues:

Global IP and domain-level reputation based on overall sender reputation.Gmail looks at traditional reputation metrics (complaints, spam trap hits, etc.) as well as the quality of addresses being mailed to (for example, what percent of your recipient addresses are secondary addresses) at a global IP and domain level. If the reputation hits certain key thresholds, mail can be blocked at a global level. This is relatively uncommon.

Subscriber-relative engagement adjusted for global reputation at the domain and IP level. Gmail appears to look at how engaged each subscriber is with mail from a particular source (domain) compared to all other sources. If the relative engagement is high, Gmail will deliver that mail into the inbox. The level of relative engagement appears to be adjusted up or down depending on the global reputation of each sender.

Content “marker” reputation. The previous kinds of filtering are tied to more permanent sender identifiers (domain, IP). This kind of filtering is more short-term and tied to a large number of  “markers” (URLs,  text in email, HTML tags, kind of message) that are found or can be derived from the message. This “content” reputation appears to be measured at a global level (for all subscribers) and at the subscriber level (for a particular subscriber).

Let’s use an example to show why 100% inbox placement isn’t always optimal: a marketer with ~100% inbox placement at Gmail that sends campaign-based email very infrequently.

In this case, the marketer probably has the chance to send more email, even campaign-based mail. For this marketer, the global IP and domain reputation is strong enough that even least-engaged subscribers are being delivered at a high rate. Since least-engaged subscribers have an extremely low propensity to open, click, and buy compared to the most engaged, lower deliverability in the least engaged bucket doesn’t cost the marketer a lot.

In this case, if the marketer sends more mail, the likely results will be:

1. Highly engaged subscribers will continue to receive mail in the inbox despite lower global reputation and lower engagement.
2. Since highly engaged subscribers are receiving more email, this drives significantly more opens, clicks and conversions. It is also reducing average engagement, but (in this example) not so much that deliverability suffers.
3. At some point the deliverability for less-engaged segments will decrease because of lower relative engagement and decreased global reputation.
4. This lower deliverability for “low engagers” means that there are few clicks and opens from this group.
5. However, the loss in (4) is more than made up for by the gain in (2), which increases overall opens, clicks, and conversion.

Please note that you must also consider the above results over a longer period of time, and also consider the impact of increase unsubscribes and complaints on overall list size.

Inbox placement has always been a means to an end. The true goal is to increase client loyalty by consistently demonstrating the usefulness of the email program, which results in consistent opens, clicks, and conversions.

Inbox placement is not the goal in and of itself. Under the new deliverability rules, more inbox placement isn’t always better. The secret is having the right data to understand when you are starting to damage the inbox placement at the most valuable mailboxes.

This post originally appeared in Media Post.

Author Image

About George Bilbrey

George Bilbrey is the founder of the industry’s first deliverability service provider, Assurance Systems, which merged with Return Path in 2003. He is a recognized expert on the subjects of email reputation and deliverability and is active in many industry organizations, including the Messaging Anti-Abuse Working Group (MAAWG) and the Online Trust Alliance (OTA). In his role as president of Return Path George is the driving force behind the ongoing innovation of our products and services. Prior to Return Path, George served as Director of Product Management at and as a partner in the telecommunications group at Mercer Management Consulting. He holds a B.A. in economics from Duke University, and an MBA from the Kenan-Flagler School of Business, University of North Carolina.

Author Archive