DM News hosted senior direct marketers last month to parse key issues affecting e-mail marketers and vendors. Here’s part 2 of a four-part, monthlong series on spam filters, reputation systems, multiple e-mail addresses and transactional e-mail.
David Daniels, Research Director, JupiterResearch
Dave Lewis, VP of Alliances & Market Development, StrongMail Systems
Greg Martz, Director of New Member Experience, The Motley Fool
Matthew Seeley, President, Experian’s CheetahMail
Charles Stiles, Postmaster, Time Warner’s AOL
Moderator: Dianna Dilworth, Associate Editor, DM News
Contributor: Mickey Alam Khan, Editor in Chief, DM News
Dianna Dilworth: I’d like to move on to discuss reputation. This question is addressed to you, Dave [Lewis]. You were the author of the Email Sender and Provider Coalition statement on reputation systems. What should be their key drivers, complaints, list management practices, etc. What should be the priority and why? And how do these reputation systems differ from anti-spam systems?
Dave Lewis: Let me try and tackle it in stages. But first, it was a committee effort, it wasn’t my authorship of the position paper. I was one of the co-chairs, I had a lot to do with it. I think it reflects the consensus at least among the members of that organization as to how reputation systems should be driven. I think we’re at a point where we have recognized that there are limitations – despite the good efforts of the ISPs – to spam filtering.
The endgame is to make this a reliable, trustworthy medium so that we can conduct communication and commerce. That’s what we’re all here to achieve, and I don’t think we’re going to get there through prevention. Prevention itself is not a cure. The spam filters are good, but it’s like horseshoes and hand grenades. Close enough isn’t good enough.
When your transactional e-mail from your corporate domain gets caught in these spam filters, you begin to reflect. Can I really rely on this medium to convey my business-critical communication? And if you can’t answer that question definitively “yes,” then you’ve got an issue.
What we look forward to in what we’re talking about here, relative to accountability, is that a) we can identify who the senders of mail are; and b) through reputation systems, associate a reputation with a known identity. That’s what we’re trying to achieve. The drivers of reputation are something that I’d like everyone in the industry engaged in, because it will change the way we use the medium.
The drivers suggested in the ESPC position statement are as follows – but first, you know, it’s important that the factors that drive reputation be objectively measured and transparent. It’s kind of like your credit rating. You can go in, look at what your credit rating is and can challenge those things that are showing up on your credit rating. We don’t have that ability, and probably shouldn’t have that ability, with spam filters that are on the perimeter. But when we’re talking about reputation services that drive whether our mail is accepted and where it is placed, we as legitimate senders should have that ability.
So we see four factors that would drive reputation systems. The first is feedback from the recipients, and that’s absolutely No. 1. Not some panel, not some model that determines whether the content is acceptable, but what the recipients of that mail want to receive and how they view it. As a marketer, I’d much rather be judged on that measure than any other. Hold me accountable for what my own customers think of me.
The second is acceptance of authentication. What we were proposing is that all of the reputation service providers require that companies wishing to access their services authenticate their e-mail. That seems a straightforward requirement, and the way we’re defining authentication is that we allow the senders to segregate their mail by brand by class of mail and that the measurement of reputation be done at the domain level.
The third point is looking at some mailing practices around things like deliverable addresses. Senders have a responsibility to be diligent in how they compile and manage their lists. We think that should factor in to reputation. But we were encouraging the ISPs to differentiate between, say, addresses that never existed when we established honey pots used in judging whether a sender has a good reputation, and addresses that have simply been abandoned by the user.
The fourth one was around bounce management. And that is kind of a joint responsibility between senders and receivers. It’s important that the receivers, the ISPs, tell us when they’re bouncing e-mail and that those codes be consistent so that we can take the kind of action on our side to understand what’s causing the bounce, whether it’s because of a block or because of an address that no longer exists. There has to be consistency, and of course senders have the responsibility to take that information and clean up their list.
So those are the four factors. But the one that is most important, in my mind, is basing a reputation on what recipients of your e-mail think of it. Do they want it?
Charles Stiles: I would have to echo a lot of what Dave said. The tools and processes that have gotten us thus far in the spam-fighting effort are not the tools that are going to take us to the next level. We have to have something else.
Typically when I talk to people about accountability, they look at that as a four-letter word. There is a good aspect, a positive aspect, to accountability. If you’ve been working hard at maintaining your lists and keeping track of your customers and sending them relevant content, you ought to be able to benefit from that. It’s not just about preventing your mail from coming through because of bad accountability. It’s also about making sure your mail is delivered because of good accountability.
Matthew Seeley: When we look at our customers, when you look at those lists and you’ve done the right things. You’re careful with the cleansing of your lists; you’re not over-mailing. You’re making sure you take all of the processes into account. Again, respecting the customer. We have seen that the unsubscribe rates are extremely low. And they get rewarded in this process, because we’ve adopted these authentication procedures and, for us, we agree with you. I think it’s a gold star for those that get it right.
David Daniels: Again on the feedback loop, I think AOL in particular has done a great job of being transparent with that for quite a while, but I think that is that next level of getting all of the large receivers on some same mechanism to make that easier or better for all the large senders.
But I agree – that last point you made, Dave – about the suppression and the bounces. I think we’ve seen in this past year e-mail address turn increasing [and will] probably see another blip, once it’s so easy for everybody now to get an AOL address. I suspect that we will see more e-mail address turn in the short term. But marketers need to be diligent in not only suppressing that but just looking at the behavior in their overall list.
I’ll throw out one more number, and that’s 27 percent of marketers, when we ask them, “How do you segment your list, how do you target?” only 27 percent say, “We look at what people click on in the e-mails. We look at click-through behavior.” [I’m] not suggesting that only 27 percent know their click-through rate, that’s much higher. But there is only about a quarter of people who are going down to that next level to say, ‘These people are clicking. These people are not clicking.’
So we have a good sense that people have large lists, but there is a good portion of that list that’s dormant, that probably shouldn’t be mailed to anymore. That’s just creating issues for the ISPs, and it’s a bit of a snowball effect. While the industry and the technology serve a role, the responsibility of the marketer certainly comes first to do the right thing and follow those best practices.
Mr. Lewis: You’ve raised a good point. I think this move toward accountability does require a change in the mindset of direct marketers. I tend to think that we’ve been seduced by how inexpensive and easy it is to pump out e-mail. And what you often see is customers doing things through the e-mail channel [they] would never dream of doing offline.
I think the metrics we use and hold ourselves accountable for need to be looked at very hard, and one of them is list size. That’s not the Holy Grail. The Holy Grail, if there is one, is about the size of the list of your engaged customers, not just how many names you have on your list.
The correlation I’ve always found is that if people aren’t engaged, there is a higher propensity to complain. Customer satisfaction is the driver of blocking today. What’s going to drive reputation systems in the future? You better take a hard look at customers, measured by complaints, unsubscribes, opens and clicks. You’d better look at making that a key metric.
Mr. Stiles: I don’t think anybody would argue that has definitely evolved. But consumers have evolved as well. What they used to view in their inboxes was a large amount of spam. We’ve done a good job at fighting that. So now they see a lot less spam.
I would guess or suspect that their definition of spam has changed or what they consider to be spam has changed. Marketers need to be careful in what they are marketing to consumers and make sure there isn’t that likelihood that they’re going to report it. We need to be more careful now than ever.
Greg Martz: From a sending perspective on reputation systems, I think it is going to be a vital part of determining whether you get into the inbox. What we’re seeing in the marketplace, it’s very fluid. I’ve gone through this recently over the past few months, trying to figure out what reputation provider to go with, or multiple ones.
You almost get a sense of paralysis, because there seems to be a lack of transparency, a lack of control and a lack of being able to determine how much benefit it does provide you. We’re willing to spend the money to get valuable service through reputation providers or even guarantee delivery, but we need to know what that ROI is going to be.
It’s in its early stages. It would be interesting to see, if we could reconvene in a few years, what we’re talking about in terms of reputation. But from the sending perspective, we’re still on the fence in terms of what do we do with reputation?
Sept. 18: Multiple e-mail addresses.
