Domain Reputation and Recipient Engagement

Today's guest post comes from Chris Wheeler, the Director of Deliverability at email service provider Bronto Software. No stranger to the email experience, Chris's past experiences have included building a deliverability program from the ground up at a major online retailer and manning the d-team at a enterprise level ESP. Chris contributes to several blogs online and is part of a handful of key industry committees and think-tanks for taking email to the next level, both for marketing and recipient effectiveness. When not neck deep in email related things, he's busy with his two dogs and enjoying his home town of Austin, TX. Take it away, Chris!

I've been getting questioned more frequently from clients and folks in the industry about exactly what domain reputation and increased recipient engagement means for email deliverability. A few months ago, the email industry started posting about reputation shifts from IP to domain. Ken Magill's bellwether of an article showed, for the first time, how domain reputation has moved into the general forum as opposed to back-office leaks from the large ISPs. There have even been some vendor announcements about tools helping to measure this.

There are two major changes that have already begun in beta behind the scenes at some ISPs that you, as a savvy marketer, should be aware of. Akin to your postal mail today being tracked by the return address (your actual street address), email has historically been tied back to the sending IP it originated from when handing the message off to the receiver. However, imagine your postal mail being tracked by your last name instead; regardless of where you actually reside. And then, by how many people who receive your postal mail actually open the envelope and do anything with it. That's what's happening with email -- your sending domain, IP agnostic, will be used in tracking the type of mail you send augmented by recipient engagement.

Let's break it down:

1) One major ISP is already using this behind the scenes while another is in the works of implementing sender domain reputation. The first step to ensuring compliancy is getting your authentication set and verified. DKIM and SPF should be positively showing that mail sent from your domain (or your email transport vendor if using an ESP) is conclusively returning a pass for both the envelope headers (SPF) and the content of the message as it leaves the sender's server (DKIM). ISPs will use this to determine whether domain reputation is even possible when analyzing incoming mail. But, if you're not signing, then your mail will probably get moved to the back of the line and face sub prime filtering as IP reputation will be considered lower on the reputation ladder. Secondly, audit all mail coming from your domain. If you've got a rogue marketer or department sending email for your domain, clean it up since all mail (regardless of the actual IP sent from) will be aggregated to determine your domain reputation and thus trickle down to affect all mail coming from it. A good way to do this is by setting triggers to log all outbound email across your outgoing email servers. Again, strategy is key. If Bob over in coupon sales is killing it with blasting recipients twice daily and aggravating them, Sally over in order processing trying to get shipment confirmation emails out will suffer if both mail streams are coming from the same domain.

2) ISPs have had access to recipient data since they developed technology a while back (read: they've known what's being sent to recipients for a long time now). Recipient engagement metrics, as legally recorded by ISPs, is yet another substantial data point being used by some ISPs currently. What does this mean? The same metrics you as a marketer watch are being monitored by the Yahoo!'s, AOLs and Gmails as well -- open, click and mail movement (i.e., spamming it, forwarding to a friend, etc.) events. In order for ISPs to drive down the number of irrelevant and unwanted email their recipients' inboxes receive, they will match recipients to user profiles based on the type of mail received and what's done with it. So, if I have a high propensity of receiving the same mail as my friend Al Iverson, AOL will assume that we both like the same types of mail and if he marks a message from a sender as spam(authenticated by #1 above), I'll most likely not see that mail. Or, if he never opens. Or if he never clicks. This is grossly oversimplified but you get my point. Recipient model peers engaging in email now are going to be critical to having mail delivered to target audiences. This will drive marketers to either conform and crank up their relevancy and targeting approach or be shunted to the bulk folder. Take special note if you engage in cross-selling, up-selling or reactivating lapsed customers (also known as "Win-Back" programs) since now, by widening your targeting criteria around business rules, you risk offending recipients and those deemed their peers in the ISP network.

If you already have great recipient engagement (as a result of sending email that recipients want) and a safe domain, these steps make logical sense as the next evolutionary gain for ISPs. You shouldn't see much of a change since their requirements are already covered by your current practices. However, if your email strategy is a bit off, or shady, at this point, you'd better clean house before you see your deliverability tank in lockstep with ISPs rolling these initiatives out.

Chris Wheeler
Director of Deliverability, Bronto Software
Post a Comment