January 04, 2007
If you didn’t make our live conference call last month on Email Marketing Benchmark Data for 2007, just click below to get your recorded MP3 audio file to find out what 3,637 marketers revealed about the latest in email marketing.
You’ll also receive our 16-page PowerPoint presentation PDF, including eight new data charts and an eyetracking heatmap. Highlights include:
- What email tests work best.
- List growth rates by industry.
- How to improve delivery.
- Has mobile marketing’s time finally arrived?
This is your downloadable version of a presentation first conducted live Dec. 19, 2006, by:
Anne Holland, President
Stefan Tornquist, Research Director
#1. Click this link to download the PowerPoint presentation PDF including seven new data charts and heatmaps (yes, you may share with colleagues):
#2. Click this link to download the MP3 audio file:
http://www.MarketingSherpa.com/tele/EMBG-IN-121906.mp3 (8MB file)
(Audio download note: This is in MP3, which almost anyone with a PC or Mac can play. If you'd like to listen now, just do a regular mouse click (left side) and it will start playing fairly quickly. If you'd like to download and save to listen later, just do a right-click with your mouse.)
#3. Here is the transcript of the teleconference:
Anne Holland: Good afternoon and welcome to MarketingSherpa's new teleseminar, The New Studies and Results Email Marketing Lists, Design, Tests and Deliverability for 2007. I’m Anne Holland, President of MarketingSherpa. With me today is Stefan Tornquist, our Research Director. This is our fifth annual email study, which is remarkable. I never believed we would have this many years of data to go over and so many trends going on in email. The nice thing, of course, is that email's not dead. We've got plenty of data to show that. Stefan, where is the data in today's presentation from?
Stefan Tornquist: This fifth annual study features the largest survey of email marketers, we believe, to have been conducted, with over 3,600 participants. That was very exciting for us. About two-thirds of those, or more -- about three-quarters of those are email marketers and about one-quarter come from ESPs and other email management agencies.
Anne Holland: OK. So, two-thirds are client side marketers at the real brand.
Stefan Tornquist: Actually, about 75% are client side. On that, we, as with every guide, we do a number of special reports based on lab tests and partnered research. We'll be looking at some of the results of a corporate firewall study that we did with KnowledgeStorm, looking at companies of over 500 employees and how the anti-spam folks work to stop spam and how the business folks are viewing. That was very interesting. We also look at the best-of research out there, and this time benefited from approximately 40 or 50 outside sources. Some of that has been published, but a lot of it hasn't, and then we take advantage of the editorial work that MarketingSherpa does collecting the great case studies. That helps us give some context to the numbers.
Anne Holland: Let's move on to our first data slide, and, of course, this is the good news about email marketing. If you've seen the executive summary from this report, which is on our Web site under the Free Stuff link on our nav bar, you'll notice that this is a chart that we put in there. I'm personally frustrated because it seems to me like every time somebody's pushing a hot new marketing type Web 2.0-this and podcasting and RSS and everything you can imagine, the first thing they do is they're mean about email. They say, "Oh, since email is dying, the world needs this new kind of marketing." It's their raison d'etre to tout the new marketing, and it's just a load of hooey. Email has been doing very, very well for years. It continues to do well, and, as you see here from the numbers, most marketers are saying that they expect the impact of email to increase at least a bit, if not significantly. So, it's an incredibly important marketing channel. We're not saying don't test other stuff. We're just saying don't abandon email. In fact, you probably want somebody who's dedicated to email within your organization if you don't already.
Stefan, what was the difference here -- when you looked at these numbers, business-to-business marketers versus marketers who target the consumer marketplace? I know that you look at all the data in our studies, really separating out those two groups because they are such different types of people. What did you see in terms of the differences about hopefulness about email?
Stefan Tornquist: To be honest, I was a little bit shocked that both groups were so hopeful. I expected a much larger percentage to sit in the middle. We do enough research into email to know that it works, so I didn't expect a negative reaction, but I was expecting a larger hump in the middle of this study and not such a progression of hopefulness, as it were. The business-to-business folks see it somewhat more hopefully. They see the increase as more significant. I would mostly attribute that to the fact that consumer marketers tend to be a little bit ahead of the curve when it comes to implementation of technologies, new tests and so forth, so they're a little bit more on top of how things are developing. For business-to-business folks, they still see a sharper ramping up of response rates and … overall effectiveness of email as a result of their internal efforts. Whenever I see trust like this, data like this, it's a shock that email, for all if its effectiveness, still in about half of the companies doesn't have a separate line item in the budget. It's just bundled in with other items.
Anne Holland: I think also they're suffering under the impression in the executive suite that, "Oh, email's free, and the reason you do email is because it's got such a cheap cost per thousand," and that really isn't the reason. The reason why you do email is that it can be incredibly effective. In the rest of our presentation, what we're going to focus on is real specific on how the data shows you can make your email campaigns and programs even more effective.
Why don't we move on to item No. 1 -- of course, growing your list. Everyone wants to grow their list, at least if they can grow it with highly qualified names. You don't just want a big list. You want a good list. Stefan, where does this data come from? What's going on here?
Stefan Tornquist: Well, this comes from our survey respondents, and let me explain what we're looking at. Let's take a look at over on the left hand side, B-to-B averaging. That shows us in an average month that a list will grow by 3.9% new email recipients. At the same time, you might lose approximately 2% of existing recipients to opt-outs or bounces, etc. That gives you a collective increase over the year of 22%, and that's the average.
One other thing about how we dissected this data: We took out those companies who were experiencing growth that was outside of the norm. Because some companies are just bringing an email program online, they may be experiencing monthly growth of 100% or even more. We took those folks out of the equation because it's hard for a marketer to compare themselves to a new email programmer who’s experiencing that level of growth. Had we included those, you would see these numbers as even higher.
Anne Holland: The one thing that I think a lot of marketers don't think about, and I've noticed this with our own reports recently -- and it's a nasty shocker sometimes -- is that when you actually look at that negative number, the percent that are leaving your list every month, you've got to be aware. You've got to, once every quarter or so, sit down and really look at the reasons why those names are coming off your list, because although it may only be 2 or 3% a month, it's a large number added up over time. That's a lot of money, especially if you’re assigning a cost per acquisition for that new name or if you're assigning a value per name.
That's a lot of names and you have to say, "Why are they leaving my list?" The majority of the time, it is not opting out. Very, very few people opt out. I've never met a marketer who had a high opt-out rate. Almost all of these names are leaving your list because they're bouncing. You may have a situation where they're bouncing because the people have changed email addresses. Actually, I was talking to Michelle Rutkowski over at Taunton Press about this, and she said that they have hobbyists in particular, people who like to do carpentry and different projects at home, and she said they had discovered they had a really big problem with loss on their list. These are people who adored them, who wanted to get their emails, but what happened -- and we're talking thousands of consumers, was that everyone was switching over from an old email account to a broadband situation.
So, people might be leaving AOL and getting a Comcast account, and it didn't occur to alert Taunton Press that they had a new email address. They actually went through a whole append project just for those old names that had been previously opt-ins, to see whether they had just, perhaps, switched over to a broadband account and to see whether they wanted to joint the list on an opt-in basis. She said it's made a huge difference to their bottom line. She's actually going to be presenting that case study at our Email Summit in March.
The other thing that you might want to look into is when it comes to bounces, and, of course, this has to do with your email software or with your ESP, your email service provider … what are they saying is a bounce and what are they saying is enough of a bounce situation to get that name knocked off your list. Everybody cares about deliverability. Everyone's saying, "Oh, we're going to get you the world's greatest deliverability and we're going to get all these names through." What that means is they're going to take off names that are potential bouncers a lot faster than they did in the past because email systems don't like things that have bounced before.
So, a lot of ESPs and email software providers are being very ruthless about stripping off bounced names, and you may not realize there may be a name that maybe you didn't want to go off. Maybe it bounced for a reason that you don't consider a serious bad bounce. You need to look into these things and say, "How are we handling our bounce situations? How are we handling email address changers? How are we contacting people who were our best names in the past and somehow or another dropped off?” Do you really want to go out there and reacquire them? I think a lot of excitement to go on -- not just on growing your list -- but around kind of remedial actions.
Anyway, let's take a look at our next slide. Now, Stefan, this is actually from a special report that you conducted for this benchmark guide with a multivariate test vendor and you had them look into a number for things.
Stefan Tornquist: Yes, Optimost analysts looked at a number of the most basic questions in registration forms. Here's an example of one of those. They looked at which of these standard variations works better than another. Here's a classic example. This is a registration form that, when it comes to email, it's asking for quite a bit, and this compares the two column with one column design. The green circle shows us which one works better. If you stop and think about it and about how you move through a form, maybe hitting the tab key, there's a good reason for that, but very often what we find is that those simple changes to a form can have a real difference to registration, and conducting multivariate tests is one of the best ways to accomplish those changes.
Anne Holland: Now, we actually had the folks from one of the divisions at Siemens give a talk at our summit recently, and they talked about how they'd been using multivariate testing on different areas of their Web site and they tried everything -- you know, the home page, some other landing pages -- and then they actually had also testing on the registration forms. And the most fascinating result for them was that the registration form testing had the biggest impact for their organization's bottom line, just for the number of names that they were able to collect, the amount of permission they were able to collect, the number of qualified leads that raised their hands … that these things mattered actually far more than graphics and then sometimes copy on the home page -- things that the creative team love to test. Guess what, it's that registration form that may be your sexiest place to put a little bit of money into multivariate testing for the next year.
I think that's very exciting. We used to think that registration forms were something that the Web department threw together and they weren't really marketing. They weren't creative. Of course, now they're turning out to be highly creative, something that you really do need to focus on in a big way for a big change in your opt-in list side.
Stefan Tornquist: I think another thing that people need to consider when -- if they're either thinking about putting up a new form for email registration or changing the one they've got is, always ask yourself that question. "Do I want a lot of information about fewer people or do I want more people but know a little bit less about them?" The more questions you ask, the fewer people are ultimately going to sign up, and I'm still being asked for my fax number on some forms and that's a real waste and if something like that is required, as it still is on some forms, it's a genuine cause for abandonment.
Anne Holland: Let's talk about the tests that do work best now that we're talking about things that are worth investing tests in. Stefan, why don't you walk us through this chart of the different tests that work best. This chart was business-to-consumer marketers. These are for marketers who are marketing to consumers at home.
Stefan Tornquist: That's right.
Anne Holland: OK, and this is for email marketers only.
Stefan Tornquist: Correct, and what's fascinating here -- No. 1, all of these tests, except for text-only messages, get an overwhelmingly positive response. If you lump in medium and high ROI together, all of them are hitting above 80%. But what's really great is that landing page copy gets the highest number of people saying that it has a high ROI return, and that's good because one of the things that every time we give one of these talks and get to talk to marketers after them, they say, "Well, you know, these tests are well and good, but I don't have the time. I don't have the resources. What's the one thing I should do? How should I spend my money?" Landing page copy is the most accessible test that we can conduct and we can conduct it in-house and make those changes in-house at a relatively low cost, so I think that's very exciting. It's not necessarily all the bells and whistles that make the real difference.
Anne Holland: In fact, the top three tests are really more about copy than anything else. It's copy-driven. It's not anything that you'd need any kind of fancy creative department for. I mean, you need a darn good copywriter, which is hard to find sometimes, but landing page copy, A/B testing, which is just copy about your offer, and subject line tests, those are all words as opposed to something just creative.
Stefan Tornquist: Absolutely. Now, at the bottom of the chart, we see that text-only messages don't do very well in comparison with the others, and that's interesting given that there's a lot of concern about HTML being blocked. We're always getting questions -- "Should I switch to text? Should I include text in the mix?" and I think it's not as simple as saying, "No, text doesn't work, not by any means." But on the other hand, HTML is certainly considered far more effective, even by marketers who are doing campaigns that are having their images blocked, retailers who send very visually intensive emails that are largely graphical. Still, everyone pretty much rates HTML as greatly superior.
Anne Holland: The one important thing -- there are actually two things to think about here. One of them is, of course, landing pages generally weren't things that are in the email department. In a lot of companies, the email department is dealing with an ordering system Web site. Maybe you're allowed to pop up some articles, but in a way you were the tail and the Web site was the body, and then you were like, "Oh, I'll get an email out because there's a new thing on there and I'll link to it."
What we're seeing more and more, and I've actually seen this from a few different marketers, including the National Association of Realtors, who do the largest business-to-business email newsletter in the world -- they have almost a million reading that site -- it's incredible for a B-to-B newsletter. Anyway, they say, "You know what? Now the email team are driving Web development." So, it's the tail wagging the dog. The email team is the one saying, "This is what we should do for headlines." The email team is the one saying, "Let's rethink this nav bar." The email team, because they're the ones often driving the traffic and they're saying, "We know what the entry point's going to be. We're already focusing that funnel in. Now, we're going to focus the Web pages, you know, sort of the after email response."
The other most critical thing is that now the email team has to have metrics that go beyond open and click. It's all about what happens post click. The next thing is a lot of the vendors are now banding together. I'm seeing announcements in the press pretty much daily, it seems like, with Web analytic firms and other marketing measurement firms kind of joining together with the email vendors and the email software firms so that you can have one dashboard look at everything in one place. So, that's more and more important, too, to get all those things joined together.
I think it's neat. Who knew that email would be the thing that would, maybe in the end, rule the Web? The other thing I wanted to point out is that these are all creative tests. These are not the most critical tests of all, and if we had added one more section of tests that would have whooped all the rest of these in terms of tests that really works best. The test that really matters, if you can only do one test, is list segmentation.
If all you have is your house list, then it's segmenting by your very newest names or names of a certain type, people who are just clients versus prospects, or multibuyers versus single buyers, or people who like this versus people who like that; whatever you can segment by. That kind of test always, without a doubt produces the highest ROI possible. Once you've tested that, then you move into creative tests, and these are the creative tests that you need to move into next. So, you don't want to leave out test lists because they matter always.
Stefan Tornquist: Well, in defense of the guide, we do take a pretty hard look at segmentation and other types of list management tests, as well, just to clarify.
Anne Holland: Absolutely. Why don't we move onto the next page. This is actually a landing page test example. If you're wondering, "Gee, I've got an email newsletter. Why am I testing a landing page?" Well, maybe you're testing whether or not it's a long article. This is actually an organization that really thought through their end prospect and they realized their end prospect was an executive on the go who wasn't going to read more than three words, and they thought, "Wow, instead of writing a 1,000-word article, why don't we just do a big honking pie chart." It's a pretty simple pie chart, but it's visual, it's eye candy and a paragraph, a paragraph and a half of copy, and that ended up doing better for them than articles did for that particular target audience.
So, it's saying, "What's going to work better?" You've got folks like iVillage who have been testing for over a year now doing what's called a video email newsletter, where they're not sending videos in the email, but they're sending hot links to video clips on line. So, you can either click on the hot link to the video and see a picture of Kate Hudson doing whatever or you can click and read an article. They have other newsletters that are just text-based content-based. It's saying what are the different types of content maybe I'm going to link to in my newsletter.
Some of the other tests that you can do obviously involve navigation bars, involve offers, involve -- obviously -- headlines matching headlines. There's all sorts of different things you can do. I'd like to hear more about landing tests from email. I don't enough. If you are doing them, please do let us know so that we can write about the glory that you are.
Why don't we move onto No. 3: improving delivery. First of all, this chart blows me out of the water. I was listening to National Public Radio the other night on my drive home and they had a big email filter expert being interviewed and he said, "All email filters are based on content," and I actually pulled the car over, got out my mobile phone and called the station and I was like, "He's wrong. I have the chart." Now, Stefan, what is this chart. What is the difference between reputation and content.
Stefan Tornquist: When we think about email delivery, the first questions we always get, "Should I put ‘Free’ in the headline?” “How many images are too many images?" All of these questions about what's in the email itself and what this chart from Return Path shows us is that some of the major ISPs are using to determine whether an email gets through or not is reputation much more than the content of an individual email.
Now, a lot of different things go into reputation, and it's very important to remember that content is one of the major factors. So, it's not that content doesn't matter. Far from it. Content is among the absolute top contributors to whether your reputation score is a good one or a bad one. If you have bad content, you're going to have a bad reputation score and get filtered on both sides, but reputation is a lot more than content. It's a combination of over 100 variables, depending on who's doing it. It's a little bit like figuring out the algorithms that Google is using to determine the top site. The factors are shifting constantly to prevent spammers from gaming the system. But, things such as having an individual IP address for the email server, numbers of spam complaints, obviously following canned spam regulations, presence on black lists. These and many other things contribute to a reputation score.
Anne Holland: I actually had one marketer -- he works on the agency side and they have several clients, or many clients actually, knowing this marketer, who were using their system to send out email campaigns as an agency and he said, "Gee whiz. We've noticed we've been blacklisted and we're blocked at a lot of corporations or certain destinations now. We're not doing anything evil. We're not spammers, but just things happen." We do understand that things happen. Even Microsoft has been blacklisted. MarketingSherpa has been blacklisted for no fault of their own. There are erroneous blacklists all the time.
Anyway, he said, "Should we routinely change our IP address?" -- and your IP address is the server that your email is sent from. Should we change that address once a year to get a fresh, clean, brand-new IP address so that we start out not on any of this filter and black lists any more and the mail will get through. Stefan, what would your take on that question be?
Stefan Tornquist: Well, my first response would be probably not. Aside from the logistical difficulties, when you change that IP address, you may even appear to be a spammer attempting to avoid reputation. You also void the good reputation score that you had already.
Anne Holland: Right. So, if you've been whitelisted anywhere, you lost that because they no longer know it's you.
Stefan Tornquist: If you're doing things right, chances are you've got more of a net positive from that reputation than otherwise.
Anne Holland: One way that I can tell the people who are using dedicated IP addresses, and this is something we've been writing about for a while, that if you're using, in particular, an email service provider, you should make sure that your contract includes the fact that you get a dedicated IP address just for you, just for your company, your brand, that you're not sharing it with one of the other clients.
The reason for that, of course, is a lot of your reputation is based on that particular IP address and if somebody else is sharing it and if they, by mistake, perhaps, get blacklisted somewhere, you share that blacklist. You're blacklisted too, through no fault of your own. People think you're the same person. They don't know it's the same IP address.
So, we've been writing about this for quite some time, that you need to get your own IP address. I can tell -- there's a quick and easy test to see if somebody actually has a dedicated IP address and that is if you happen to use Outlook for your mail and you have on the regular Outlook junk mail filters, in that, as with many email systems, you can white list somebody. You can say, "Please mark this name as safe. Email from this sender should get through at all times." Have you ever done that? I know I have. I've done that with certain email senders and then gosh darn it, they showed up in my junk filter the very next day or the next time they sent to me, and I couldn't figure it out. Well, it's because their email vendor was either switching IP addresses on them, maybe they were switching across systems or they didn't have one that was dedicated.
Actually, one of my favorite bloggers of all time is always in my junk mail filter no matter what I do to white list him, and I actually went up to him at a summit and I said, "Would you please get a dedicated IP address so I can be sure not to miss your stuff? I love reading it." He said, "OK", and he actually went out and did it, hopefully not just because of me, but because of all of his readers. So, this stuff really, really matters. There are a lot of mysteries explained by the IP stuff, as annoyingly technical as it is.
Anyway, why don't we move on to this chart: The B-to-B Delivery Challenge. Now, Stefan, this is the special study that you conducted with KnowledgeStorm for the Benchmark Guide this year. Can you describe what's going on here?
Stefan Tornquist: What was interesting about this study was that it gave us an opportunity not to just ask business users how email is changing in their practices, which we were certainly able to do. We were also able to survey the anti-spam people in the organizations themselves, the people who are writing the code and implementing the systems to interdict spam, and it was really interesting to see how very differently these two groups view the question of spam, view false positives, etc. It makes sense, because the anti-spam folks, it's their mission to get rid of spam and phishing emails. They're much more concerned with that than the possible negative impact on marketing and permission-based email that happens on the other side.
So, what we did here was -- this was one of the first questions to that group -- we wanted to know what they were doing to get in between spam and their end-users. Remember, this is at companies of 500 employees or more. We wanted to sort of separate them out of companies most likely to have a real firewall, which is where -- there's so much mystery with what's happening in B-to-B email once it gets thrown beyond the firewall. We don't know -- is it going to bulk folders? Is it not going through at all?
Here, we did an update of some Return Path's work from 2004. We thought it was good to get an updating of that two years later. As we can see, the most telling thing, perhaps, back a couple of years ago, almost 10% of companies weren't doing anything to stop spam. Now, none are. Not one company is doing at least one, probably more, of the different tactics that we're looking at here, and you'll notice that about one-third do look at public blacklists to determine whether or not emails get through.
It's very important to maintain some kind of review of where you in regards to the different blacklists. Now, probably the good news is that a lot use standard commercial filtering applications or appliances, and for those folks who are either very sophisticated in-house or who are using used deliverability consultants or companies, those commercial filtering applications are at least somewhat predictable in what they're going to block or what they're not going to block.
Anne Holland: Now, of course, this points out why, frankly, it's always been twice as hard to get email delivered to the work address than the home address. It actually is fairly easy to get through to the bulk of most consumers if you've got a decent reputation, you're a good mailer, you're a permission mailer, you are really only sending to people who want it and sending content that they really want and you're keeping your list fairly clean. If you are sending to folks at work, it's a heck of a lot harder because every company has their own filter and every company is applying the rules a little bit differently and they are, let's face it, not as worried about blocking permission email as they are about stopping as much spam as possible.
The CEO of your company is probably a lot more worried about labor lost due to people having spam in their inboxes than he or she is about the odd missing newsletter, which is too bad. This is why I would really like all of us to become ambassadors. We in the marketing department have to become an ambassador on behalf of permission email to our own organizations. You have to go and have a meeting with the IT department, whoever is in charge of the spam filters and email that's incoming for your organization. You have to say, "Hey, tell me about what filters we're using for our own email. Tell me how you guys decide what gets through and what doesn't. Tell me how we, the recipients, can white list folks when they send email. If I really want something, how do I white list it, aside from perhaps my own Outlook white listing? How do I make sure it just comes in through the company filter?
Most executives are shockingly unaware of how to white list an incoming sender. They're a little embarrassed about it, like, "I really don't know." It's our duty to fix this in our own companies so the rest of us then can just send email newsletter and make sure that the news gets out. A lot of marketers will admit to me that, "Oh, well, we don't actually -- can't get underneath that guy." Well, guess what. It's up to the marketing department to fix it, because nobody else is going to have really gut-level interest in doing so, so we can become the whitelisting ambassadors for our companies.
Obviously, we don't want to let spam through, but at least you can find out how do we white list. How do I put that, maybe, on the company intranet, so that other folks know how to white list email they're really interested in? It's sort of an internal politics that's totally worth doing.
Stefan Tornquist: And I think that the political campaign is exactly the right word, because one of the disturbing findings was the difference in how the anti-spam folks view spam as a trend compared to the end recipients.
Anne Holland: So, in other words, the IT department that may be going, "I'm stopping spam from coming into our company.”
Stefan Tornquist: Right.
Anne Holland: Versus the average guy sitting at his desk.
Stefan Tornquist: Well, versus the business email recipient. We asked what is the spam trend: Less? The same? Is more spam making it to your inbox. Only about 29% of the business email recipient said that spam was getting worse. A fair percentage said it was getting better or at least staying the same. Meanwhile, 60% of the people who are working to stop spam said that more spam is making it into the inbox. We know from other studies that that's probably not the case. It looks to be roughly stable. Not that more spam isn't being sent, but what's ending up in the in box.
Anne Holland: Getting through. Right.
Stefan Tornquist: It's a pretty shocking difference, and there's a lot of psychology involved in there. So, it's going to take a campaign, a public relations campaign, to see any real change, I suspect.
Anne Holland: I also think -- one last thing before we move onto the next slide -- and this affects business-to-consumer mailers. It's not just B-to-B, because a lot of consumers and a lot of executives like to get their personal mail at their work account. They may not use their home account, maybe because the kids are using it or it gets cluttered, or they're ordering -- I know I order all my Christmas presents and I have accounts at all the major ecommerce sites using my work account because I don't want my kids and my husband seeing what I'm getting for them next Christmas. A lot of people do get personal email at work. So, this is something that does affect you, and you can go to your email lists and sort by ISP. You can sort by EarthLink and Comcast and AOL, and all that, and say, "OK. What are the percent of potential at work accounts that I'm actually sending to?" and see, "Gee, this is the percent. This is big or small as a problem for me." It may be bigger than you think.
Why don't we move onto the next slide. This is email reputation management and affiliate marketing. Now, affiliate marketing, for those of you who are not aware, is a $6 billion industry. It affects mainly business to consumer. Ecommerce sites are the most famous for having affiliates. In fact, Amazon has claimed that up to 40% of their sales are through their affiliate network or their Amazon Associates, but actually, even more affiliate marketing is done on the consumer lead generation side of the business.
These could be anything from financial services, automotive, real estate. All sorts of different types of marketing are done via affiliate networks these days, and affiliates are people who are marketing on the brand's behalf. So, I might be, say, Forddirect.com and I'm trying to market Ford automobiles and Ford trucks, but I have a big affiliate network of all different marketers who are sending emails saying, "There's this wonderful special on Ford and you should click here and find out more about Ford.” All of these folks are sending to their own lists or renting lists to send out these offers via email on my brand's behalf. Now, Stefan tell me a little bit about this study that you conducted.
Stefan Tornquist: This company Lashback looked at 1,000 emails from the top 10 affiliate networks by volume.
Anne Holland: Now, this would be people like Commission Junction.
Stefan Tornquist: Exactly.
Anne Holland: Please note, this is not Commission Junction that sent the email. It is one of the largest networks that affiliates use to connect with their merchants.
Stefan Tornquist: Right. So, looking at these high-volume affiliates -- these are affiliates that are household names, virtually -- they looked at 1,000 emails and deconstructed them to see how well they were doing in terms of their CAN-SPAM compliance rate. As you can see, 70% had some kind of failure. The majority were from line accuracy, and CAN-SPAM requires certain prescribed ways of showing the “From” line. That’s by no means the only mistake that's made. Postal address -- here's almost 23% of those don't show the physical postal address and that is a very basic CAN-SPAM requirement. It's also one that the reputation services look for. So, you are, in part, responsible for the content going out in your name, and so these different mistakes, or omissions can have a real effect on your own reputation over time.
Anne Holland: I have talked to a lot of affiliate managers, and we also conducted an annual affiliate study and we ask them a lot of questions about their email and their church marketing practices and what they think about their affiliates. One of the interesting things is often affiliate managers say, "Well, you know, of course, I'm very careful about the email my affiliates send out. Whenever anyone complains, I immediately look into it and fix the problem." Of course, that's a very passive stance. That's saying, “Well, if an end recipient complains to me and I happen to get that complaint -- of course, it has to go through the customer service department to get to the right person, and if it's a large organization, that can be tough -- gee whiz, maybe I'll look into stopping that affiliate's marketing.” It's fairly rare that you're going to get that complaint. In the past, that may not have mattered that much, but now that reputation is so much of getting the rest of your email through, as Stefan said, you have to be more proactive.
You have to be seeding your name or a seed name onto every affiliate's list that you allow to send email, and you need to have someone going in there and just doing some quick checking. This is something you can have an intern check. It isn't like you need a highly paid marketing executive to go out and check these seeds. It's dive in once a week, take a look, make sure they got a street address on there. This is not hard stuff, but it is stuff that you're going to have to begin doing. Why don't we move onto the next chart. Oh, this is the fun one. This is the eye-tracking one. I love this stuff. Stefan, why don't you talk us through this eye-tracking chart.
Stefan Tornquist: I'm sure most of our listeners are probably aware by now of how eye-tracking works, but just very quickly, the image on the right-hand side shows where people looked, how intensely, where they left the page and where they clicked. The darker the red, the more people looked there. So, this is just one example. This is one of the original mails that were used for comparison. What we did in this study was we took several different types of email -- this sort of retail blast, a business-to-business newsletter and a webinar invitation. We took those three and had our designers change different elements within them to see how that would affect readership. How would it affect the way people moved through the page, how far they scrolled, etc. Where they clicked and so forth.
Anne Holland: What I think is most interesting is the fact that now when we set up the eye-tracking thing, none of these links were actually clickable. So, it wasn't like the person's little mouse icon was turning into a hand indicating where they could click. People were just clicking.
Stefan Tornquist: That's right. People still keep clicking away.
Anne Holland: What it looks like is that they weren't necessarily clicking always on things that would have even been clickable had this been a live email. For example, it looks like -- if you look at the picture of the woman and the man kind of hiking up the side of the mountain there, people were clicking on them and people were looking at them a lot. We always see people looking at people. So, you see a lot of eye attention on the models and a lot of clicks on the models, and, of course, I don't think Dunham was actually selling the models. I mean, sure, yeah, the glory of hiking is what you're kind of buying when you buy the shoes, but they were pushing a 10-30% off sale, and I would have probably, if I were them, preferred to get more people to click on the shoes than clicking on my little illustrative photo of some models.
Nevertheless, we also see a lot of clicks on the right hand, on sort of the edges of the email, and if you look at this, they had a nice little hot link just under and to the right of the picture where it says “View All.” So, it doesn't just say, “Gee, you can only click on these images here.” A lot of people clicked on that little View All link, and I don't always see that on ecommerce emails. So, that's interesting. Folks also clicked on a lot of the navigation aids. In fact, they were more likely to click on those places than actually in the Shop Now button, which was integrated into the image. So, there's a lot of sort of thinking that's going on there. At the very least, make sure that your images are clickable. If the image of the man and the woman hiking had been clickable and just perhaps taken folks to a general landing page with sale items with the image there again, it probably would have done fairly well as a campaign, and I'll bet you anything that image wasn't originally clickable.
Why don't we move on to are consumers even looking at images? Stefan, what's the data here?
Stefan Tornquist: Well, they are, to some extent. This study ask folks, "When do you active images?" because, of course, most HTML, most email clients, both consumer and in business, are now set as a default to block images. So, if you're sending out postcard style emails that are just one big graphic, for a lot of folks, all they're seeing is a little red x. So, this study asked folks, "How often do you activate those images and in what circumstances?"
Not surprisingly, the No. 1 answer was in statements and order updates, what we call service emails or transactional emails, that's when they say they're most likely to do so, at least occasionally. Now, what we don't know is how many people change the default setting on their email client. Our best guess based on previous research is it's no more than 30% are switching the entire default setting to show images all the time.
Anne Holland: Wow. No more than 30% of people just automatically see images no matter what?
Stefan Tornquist: That is --
Anne Holland: So, I'm one of those freaks.
Stefan Tornquist: I would have to qualify that as an educated guess based on sort of ancillary research, but I'm still fairly --
Anne Holland: It's sure not anywhere close to at least 50%. I thought it would be bigger than that.
Stefan Tornquist: Now, of course, some email clients still do accept images by default, but of those that have it set, that's our best guess currently.
Anne Holland: So, the majority of people receiving email are having to hand click on each email and say, "Yeah, show me those images.”
Stefan Tornquist: That's right, and, of course, you've got a difference for business users who are looking inside of their preview panes versus emails they open. You're going to see fewer people activating images inside the preview pane and then more, probably half or more, opening images, or viewing images, rather, in open emails.
Anne Holland: So, all this means, basically, is that you always, always, always have to look at the glory of your email creative from what's going to show up if images are not active, what's going to show up if that shows up with that little red x on that image -- well, can I still see something of interest, or am I just seeing a swathe of white with a little red x in it. We did run several case studies in this past year about marketers who tested adding a few lines of text at the top of their email and then running their images underneath and the lines of test were a navigation; just sort of, "Here's a list of the articles that are in this issue," or "Here's a little bit of information.”
It's just content links so that you would at least see something, and then if your image is run and you scroll a little bit you would see the beauty of all the images. In fact, I believe it was the Sierra Club who had tested both ways and saw a significant upswing in clickthroughs over a period of time after they switched to having those plain-Jane text content at the very top of their newsletter and then having the glory of their beautiful images below. I think there are a lot of interesting things going on there, and we are going to continue studying whether or not you should have text at the top, what size the images could be, all that sort of stuff. We'll keep up with case studies on all that.
But now, chart No. 5: mobile marketing. Now, by mobile marketing, what we mean -- and let's be clear, because some people when they say mobile marketing actually are talking about things like little ads that are on top of taxicabs driving around New York City. That is not what we are talking about. We are talking about mobile phones and all sorts of mobile devices -- BlackBerrys, etc., text messaging, stuff like that. Stefan, what's the data here?
Stefan Tornquist: Well, today, most mobile messaging, at least in the United States, is still about texting. SMS ads, like ‘American Idol’ where you could vote for which horrible singer you thought deserved to go on. I've actually never seen it. Maybe they're great.
Anne Holland: Let me add, we do have a case study about ‘American Idol’ in our database, along with their results there. Anyway, please go on.
Stefan Tornquist: Excellent. So, remembering that the US is way behind in mobile marketing -- only about 13% of US mobile subscribers have received an SMS ad, as compared to two-thirds of Spanish mobile users, half of the French. In other countries --
Anne Holland: Pretty much everyone in Asia.
Stefan Tornquist: And pretty much everyone in Asia, and I just heard a story of how in India it's not opt in. In India, you can just start receiving -- they're selling lists of mobile phone company consumers -- and so it could be very, very different here. In any case, only about 13% have received even a text ad. Only about 12% of mobile subscribers use their mobile devices to browse what we call the mobile Web, and that's using the built-in browsers in Web-enabled phones. So, it's still really on the verge. For most marketers, it's not a big part of the plan, but this was the year that some members hit 10% or above. So, for all the talk of mobile marketing, I think it really, in many respects, currently arriving.
We asked marketers what role mobile was playing in their budgets. About 52% say, "We don't see mobile applying to us in the near future," and then another solid 30% say that they're following it but not spending any money. So, that leads you with about 10% or a little more, doing something actively in mobile this year, whether that's testing it on the largest group, around 8%, and then about 4% saying, "You know, we've already spent money and we're going to be doing so again." There's certainly a lot of interest and a certain amount of money. There are some systemic problems in the US market that European providers don't necessarily face. The number of carriers, and US carriers really haven't decided quite how they want to proceed. They want to make the money, but they're very concerned with their customers. Mobile has a very high cost of customer acquisition. They are very concerned about anything they do that might alienate consumer. So, it's still much more of a Wild West show than in other countries.
Anne Holland: It's interesting, because on the press release front and on the news front, the hype surrounding mobile is almost becoming deafening. Almost every single day I'm hearing about a major media company talking about that. They're adding a new mobile channel, or MTV Mobile has launched. So, brace yourself. There's going to be this avalanche of stories around mobile, and, of course, all of the media companies that are launching these mobile channels are all hoping to lure advertisers and convince you it's the greatest, best thing. So, I would say, have a healthy, grain salt. At the same time, if you're considering going mobile for your marketplace or doing something in mobile, I would recommend that you look at your demographics first. Don't just look at how cool mobile is. Say, “Does this really match our demographics and are we going to be able to track it?”
For demographics, the folks that you're super obviously going to be looking at in particular, young men ages 18 to 24, and also, in addition, don't be surprised, but people who are addicted to BlackBerrys, and that's frankly quite of bit American business people, especially frequent travelers. That whole BlackBerry addiction is something that a lot of these folks are actually getting their regular email on their BlackBerry, reading it and reacting to it from there. So, a lot of B-to-B folks aren't really aware. You know what, mobile is already affecting you. You are mailing to the mobile device. You just didn't know it. At the very least, you might consider doing a reader survey. Why not? Do a quick poll. Find out what percent of the folks in your opt-in list are routinely reading your email on a mobile device, a PDA of some type.
The other folks who this really affect are marketers overseas. If you are marketing in China, the average Chinese person on the Internet receives and sends more text messages per week than they send email each week. That's been a fact for the past two years. We did a couple of case studies about marketers in South Africa recently who have integrated mobile into their email autoresponder campaigns, so that it was one of several steps of touches, routine touches, they would send in the course of making an incoming prospect, kind of nuturing them into a client. So, mobile becomes one of the nuturing steps.
Out of all of the folks that we've covered, and on the next chart you'll see -- here we are -- deep fried mobile goodness. It's an actual mobile campaign. It's a text campaign from Popeye's they just conducted this August and September. We had all the exclusive data on that, and what they told us was the biggest surprise was how quickly it took off. They had those signs erected outside, I believe it was like 79 Popeye's in the Houston area, and they had them erected on a Sunday night with the mobile offer. By 7 a.m. on Monday morning, folks were texting in, but the interesting thing, aside from that they had an autoresponder that would zap out the offer back to them saying, "Come in after lunch and get this offer.”
What they said that a lot of people, when they got the autresponder with the offer from Popeye's, would respond to the autoresponder with a little thank you. Mobile actually is what email used to be. Email used to be that super, super, super personal thing. It was sort of your mom would email you and your best friend would email you and that was about it. Mobile, in a way, is at that point now where it's still something that is private and you're not going to send as much a blasty thing as a private thing, and people feel it in a much more personal way than email.
They actually got a lot of thank-you notes, little personal notes from people who had loved Popeye's and wanted to let Popeye's know. So, you actually have a customer service thing there were someone has to be primed and ready to respond to the responders. There's a lot of positive stuff going on. By the way, the one last thing I want to point out -- we've got that Popeye's case study on the site, but the neat thing about Popeye's is this whole entire mobile campaign -- I mean it went on for a month, they got thousands and thousands of consumers involved -- cost less than $5,000, in fact, considerably less. It’s cheaper than you think if you want to start testing stuff. So something to look into for the future.