Increasing Return on Marketing Dollars

A Newsletter Published by
Lee Marc Stein, LTD.
January 2006 Issue

Contents

• The Why and How of Web-friendly Direct Mail
• Top 12 Rules for Testing Direct Mail Creative
• Support Vector Machines for Direct Marketing Applications

The Why and How of Web-friendly Direct Mail

The premise was that the Internet was going to wipe out direct mail. Over the past five years, we've seen that to prove as false as "2% is a good response to a mailing." What we've experienced instead is (no surprise) media convergence and synergism: direct mail helping online marketing and online marketing helping direct mail.

Direct mail helps online marketers by getting serious prospects to visit sites and register. Paid Search Engine Marketing doesn't capture all prospects. Because of all the online clutter, direct mail has been re-born as almost an under-the-radar way to motivate recipients to log on. By using personalized URLs and digital printing, today's mailings get significant results for marketers, and in particular business-to-business marketers. And, of course, online marketers, after burning through billions of dollars in the dotcom collapse, have learned that they need to take a direct response approach and not worry about eyeballs and branding.

How the Internet Is Shaping Direct Mail

For those of us who spend our time creating and managing direct mail, there are lessons to learn from our online brethren and sisteren. Follow these guidelines and watch your response rates soar –

1. Put more emphasis on the OFFER.Online marketing is all about strong offers and trials. Recently I worked on a direct mail package for an online marketer of business printing services. Their strongest offer is giving away the first 25 of the item they are selling. This "giving away what you sell" offer had been anathema to mailers, but now they should consider it. In your direct mail, you must at least test those offers working online, and particularly offers that generate customers with a high lifetime value.


2. Think differently about the components in your package. The classic direct mail package – outer envelope, letter, brochure, response form and BRE – may be dead for all but a few categories of mailer. First, if your objective is to drive consumers to your web site – pure traffic building – self-mailers and postcards may be your package. In business-to-business applications, what is working well is a personalized letter in a #10 envelope and no other components. The letter contains a personalized url (often first initial/first name and last name), and this device has been known to lift response by 50% or more. Many mailers (both consumer and b2b) are dispensing with both the response form and the brochure. They obviously want the transaction to take place online, but most offer an 800 number as well. What replaces the brochure frequently is a buck slip. In many cases, it merchandises the offer, but some mailers use it effectively as a preview of the web site.


3. Simulate landing pages. A landing page is a surface that offers traction, and we've always had those in direct mail. The fronts and backs of envelopes, the top of the letter, the front cover and perhaps the overleaf on the brochure are all places where the reader can gain traction and get into the rest of the package. Online, the landing page is a means of bridging the gap between the original motivator (paid SEM copy, email, banner ad) and the marketer's site. It presents the reason why the clicker should keep going. What's closest to that in direct mail? It's the Johnson box. Think of it as your landing page for the package and your letter will work better.


4. Shorten your copy. Yes, long copy still works… but only in certain applications – investment advisory newsletters, health newsletters and vitamin supplements, to name a few. Most direct mail copy has been on a continuing diet over the past 10 years. Consumers, shaped by their time online, just don't have the patience to read longer direct mail copy. Interestingly, if you send out a package on a high-ticket item, consumers often will visit your site AFTER they finish looking at the direct mail. They want to be able to delve deeper on particular points.


5. Pump up immediacy. Online marketing is all about the NOW. On certain sites, sale prices expire in minutes or hours. Email subject lines talk about offers expiring in just days. If you want Product A, but take Product B at the same time, you save more. Pop ups make the most of the moment. Think about shortening the deadlines in your mail pieces – not four or six weeks from mail date, but 10 days on first class mail, under three weeks on standard mail. Place "pop ups" on response forms, buck slips, PostIt notes.


6. Improve your "Shopping Cart." It was perhaps five years ago that online marketers found the abandonment rate once visitors reached their shopping carts was horrific. They were spending a fortune to get people to their site and navigating through it only to flop at the moment of truth. So the smart ones fixed the cart and survived. Offline cataloguers have had great shopping carts for years, but the rest of us need to improve.


7. Example: I recently completed a membership package for a professional association. The key membership benefits are a magazine and very strong web site. But the association didn't want joining online to be a response option. Huh? Nor would they allow prospects to respond by fax or toll-free number. "You vill shop my way or not at all" just doesn't cut it offline any more.


8. Consider multiple efforts. Email marketers will think nothing of releasing blasts to high-potential prospects and customers three times a week or more. Obviously there's a big cost differential between their use of multiple efforts and ours, but we are not mailing to our top segments frequently enough. Fundraisers in particular think they are hitting donors too often, but surveys of their donors do not indicate this.

Early in my career, I would go to the Post Office and watch how people opened their mail – which envelopes they discard, what they look at first and longest in the envelopes they opened. As Yogi Berra said "You can observe a lot just by watching." Now I observe what people do online.

Top 12 Rules for Testing Direct Mail Creative

Recently, I gave a 3 hour workshop on Testing Direct Response Creative for the Louisville Direct Marketing Association. Most of the workshop focused on direct mail. Here are the key points –

1. In a new product launch, always test two very different concepts (and not tactics). If you test only one and it doesn't work, you have no idea what to do next.


2. Similarly, when you are far from meeting budget, you must test radically against your control.


3. Never mix apples and oranges: the wording of the offer in your test must be exactly the same as in your control.


4. Similarly, don't mix concept/creative strategy testing with format testing. If you need a quantum leap in response, go with concept and creative strategy testing first.


5. Test the big ideas. These would include vertical and/or horizontal positioning of your product/service; what customers call it, how they use it, and what they tell their friends about it; and how the package leverages societal trends.


6. Consider creative segmentation. It's not a big idea, but it can provide the same kind of lift in response if you harness database technology and digital printing properly.


7. Next to testing big ideas, envelope testing is most important. You have to decide if you're going with a one-to-one correspondence strategy vs. envelope as store window. In b2b mail to executives, one-to-one almost always wins. Promotional envelopes haven't worked for most mailers for some time.


8. Letters are still the prime selling vehicle in most direct mail. (The latest trend, in fact, is to insert letters into self-mailers.) Key parts of letters to test are the headline or Johnson box, opening two paragraphs, P.S. and cross-heads or sub-heads.


9. If your control package is working fairly well, you may want to engage in component testing – new letter, new brochure, lift note, or even response form.


10. Go with your best effort first, then see what you can do to cut costs. Try eliminating the brochure first. You can also look at eliminating the response form and BRE, particularly if you are driving people online to order.


11. Learn from your competitors. Don't bother testing what they haven't been able to make work.


12. Test on a plan, not on a whim. You need to be strategic about improving your performance.

Support Vector Machines for Direct Marketing Applications

By Joe Weissmann

Do you really want to know a model modeler? Let me introduce Joe Weissmann. I've known Joe since my BusinessWeek days in the early '80s. Every once in a while he disappears for a few years because he's bored, then comes back into direct marketing with a vengeance and with a great new breakthrough.

A few years ago, he imported the concept of Support Vector Machines from the non-direct marketing world. SVMs are the basis of a new classification modeling technology that has been shown to provide substantial improvements in predictive power over competing techniques such as logistic regression models, decision trees, and neural networks. I'll let Joe take it from here. -- LMS

The way SVM works can be best explained to non-statisticians by using an analogy. Perhaps the best of these was expressed by Ernie Schell of MSA in Ventnor, NJ, who said: "Someone takes a bag of children's marbles and scatters along them along the boundary between two properties in a square dirt field, 100 yards on a side. Your task is to make a map of the field and draw a reasonably accurate boundary line, based on the trail of marbles.

"You don't know exactly how many marbles there are, but you can guesstimate by assuming that the bag is neither so small as to contain fewer than a handful nor so large as to contain more than a few handfuls. You assume that the marbles are all perfectly round and contain the typical combination of colors and patterns you would expect to find in a bag of children's marbles.

"How would you go about accomplishing the task? You might walk around the edge of the field, hoping to sight the trail of marbles there somewhere, and then follow that trail, making marks on the map where you find each marble. If you reach the other side of the field, you would have nearly completed the task. You'd need only to draw the most reasonable line that fits the pattern of marks on your map.

"You would probably look for the trend in the line as well, creating as straight a boundary as you could (not zigging or zagging in connect-the-dots fashion as a literal track-the-marbles line would do). If you found a few stray marbles outside of your trend line, you would probably just ignore them.

"But what if the trail — which had marbles every two or three feet — sort of curved off 20 yards from the far edge of the field and then you couldn't find any more marbles nearby, but you found one or two at the edge of the field? If you had collected about a "bagful" of marbles, you might assume that the gap was there because there weren't enough marbles to complete the line. If you didn't have a bagful, you might go back to see if there were a major deviation in the straight line at the point where the trail stopped.

"The point is, because you know from experience what a marble looks like, you don't have to explain to someone how to differentiate a marble from stones of all shapes, colors, and sizes; coins; bottlecaps; glass fragments, gumballs, cigarette wrappers; or any other debris that might be in the field. You also have a sense that the boundary line is probably relatively straight, or that at least it is not randomly meandering, and it doesn't go in a loop-de-loop pattern.

"In other words, using common sense and intuition, you go out and pick up the marbles, make your marks, and finish the job in good order.

Focusing on the margins

"Another reason you'd get the job done effectively is that you focused your attention on where the marbles were, and once you found the beginning of the trail, you went forward where you expected them to be. Going marble by marble, you might have discovered that the trail actually curved and nearly looped back around to its beginning, rather than cutting straight across the field, but you'd have had little trouble following the marble markers, so long as they were distributed more or less evenly, without major gaps. If you saw several bunched together, you'd probably have made a single mark, picked them up, and moved on.

"As noted, this is all common sense. And it happens to be how support vector machines work. But it is not at all how your typical statistical analysis works. To complete the analogy, with most other statistical methods you would have to perch a robot high above the field on a platform, teach the robot to distinguish between marbles and other objects such as stones, then have the robot look through binoculars to find each of the marbles. If there were a gravel patch in the field, you'd have no trouble finding the marbles, but your robot probably would!"

This analogy illustrates how SVM focuses on "the margin" — the relationship that each point (or marble) has to the boundary line between two spaces — and how it focuses on the points closest to the boundary. These closest points are called support vectors, hence the name support vector machine analysis.

Standard statistical methods are not designed to focus in the way that SVM methods are. The more the marbles in the analogy zig and zag, the more difficult it is to find a trend line. SVM is based on a much more practical relationship between the vectors and the trend line than are methods that examine the relationships among the points before attempting to establish a trend line. SVM behaves as you would — by focusing on what you intuitively expect a boundary line to be.

SVM is a form of data mining; it's also a form of artificial intelligence applied to data mining. By combining the two, bringing "pattern recognition" and "statistical probability" into the picture. SVM can handle a much larger number of inputs than traditional mining as well. And it knows when "close enough" is enough.

Direct marketing applications

In direct marketing, you are trying to draw a line (or make a prediction to distinguish) between those who will respond to an offer and those who won't, based on the characteristics of the offer and the characteristics of the target universe. If you want to give a weight to the prediction, you can determine how far from the boundary line a specific point, vector, or prospect is.

Modeling is the process of using historical or "training" data — in which the response results are known — to determine where the points are that best fit a boundary line. With methods other than SVM, there is a tendency to "overfit" the training data, so that while the model is extremely accurate in explaining what happened before, it is misleading in predicting what will happen in the future under similar but different circumstances.

SVM allows you to make accurate predictions of future behavior based on a practical analysis of past behavior. To use still another analogy, instead of categorizing all the brush strokes in a fine-art portrait, it gives you a paint-by-numbers template to reproduce it. The result will be recognizable and serve a useful purpose.

To find out more about SVMs and modeling for direct marketing applications, contact Joe Weissmann, President, JW Direct, jw@jwdirect.net, 310-841-6500 or visit http://www.jwdirect.net.

To unsubscribe, type "No More" in the subject line and e-mail to lmstein@leemarcstein.com.