The old joke goes that if you took all the economists in the world and laid them out end to end, they'd all be pointing in different directions. The same can be said about marketing people when discussing the relative importance of the various components of a direct mail campaign.

In October 2001, David Cabrera of Aura Corporation (www.auracorp.co.uk) posted a reasonable model to the UK-Netmarketing list (www.chinwag.com/uk-netmarketing):

• Targeting 600% difference between the best and worst lists

• Incentives 400% difference between the best and worst incentives

• Creative 35% difference between the best and worst creative

• Response Device 15% difference between best and worst

• Timing 10% difference between best and worst

In a classic TIMITI approach, you want to check out all of these elements. And when it comes to measuring email marketing, you want to check them at various stages in the email marketing life-cycle:

• Emails Sent

• Emails Opened

• Emails with Clickthrough Response

• Emails Bounced

• Emails Unsubscribed

Emails Sent

This one's easy. Either you have a database and you determine how many of your potential targets are going to be on the receiving end of your onslaught or you're renting lists from respectable email list managers who believe in opt-in-only email. So you know very well how many emails are supposed to go out.

When dealing with your own data managers and with third parties, make sure they provide you with reports showing how many actually went out. You may want to have the ISP that manages the mail servers provide a backup report. The email promotion process starts getting more interesting, however, when you look at how many of your messages have been opened.

Emails Opened

You know how many direct mail pieces you mailed and you have a receipt for postage to prove it. You know how many went to bad addresses because you have the unopened envelopes unceremoniously returned to your desk. The same is true with bad email addressesthey bounce. And, of course, you know how many people responded to your offer because, well, they responded. But wouldn't it be nice if you knew how many people opened your envelope before tossing it in the wastebasket? With email, you can. ReadNotify (www.readnotify.com) provides a service that makes use of the "return-receiptrequested" feature that requires the recipient to acknowledge the email.

Much more elegant and much less intrusive is the single white pixel maneuver. It's a simple little trick, really, but it only works with email recipients who accept HTML messages. Most email clients come preconfigured to receive HTML, but due to personal preference (like when I'm in a foreign country and paying $3 per minute for a very slow hotel connection) and corporate firewalls, there is a significant percentage who want/need text-only messages. For the rest, a link to a single white pixel is the trigger.

Send messages out with HTML code that fetches the graphics from your server. A link to a single white pixel will tell you exactly when your email message was opened. If you code those links right, they'll tell you which recipient opened it as well: www.company.com/email/LandingPage.html?mssg982359 So now you know how many messages went out and how many were opened. You'll also find out how many were misaddressed.

Emails Bounced

One of the nice things about email is that it automatically tells you when you've got the wrong address. Emails bounce. If you're sending out your own, those bounces will come right to you. If you're using a third party, make sure they provide the proper reports indicating how many bounces there were, to ensure you don't pay for faulting labeling. There is one other negative bit of feedback you can expect and here, too, the fault might be shared: people who unsubscribe.

Emails Unsubscribed

If I directly sign up to get your email announcements or newsletter and I don't like what you send me, I'll sign off. That's an absolutely wonderful way to measure whether you're hitting the mark from a content perspective. When people like what they get, they tell their friends and your subscription rate goes up with every send. If they don't, they let you know in no uncertain terms. If you rent (opt-in-only, please) email lists, then there's another layer of meaning when subscribers choose to no longer be subscribers:

• The email address collector may have collected the addresses under false pretenses.

• Your message may have been inartfully sent to the wrong list.

• Your content sucks.

Having content that sucks is more often than not a surprise. You think it's smashing content. Your subordinates all tell you you're right. Just count yourself lucky that you've learned something valuable at a small comparative expense. The other two problems, however, tell you a lot about the vendor you're dealing with.

Some email list database merchants collect names however and wherever they can. The worst collect them willy-nilly (which stands for will he or nil he-whether the address owner wants it or not). Any of your messages sent to this hodgepodge list will be met with a higher degree of disdain due to the lack of serious targeting.

But even the best email address accumulators may run afoul of the salesperson who is so intent on renting you tens of thousands of names he or she doesn't focus on which lists should and which should not be on the recipient list for your offer. Fortunately, this works against them.

At YesMail (www.yesmail.com), they are very careful to track which types of messages are of interest to which categories of subscribers. With more than 25 million people signed up, it's important to keep them all as happy as possible.

The novice sales consultant, calculating the commissions in advance, might think that a new chat software release might be of interest to all 3,936,384 subscribers to the Internet category and to all 2,958,777 subscribers to the Computer Software category. The sales consultant who's interested in keeping all of those subscribers will recommend sending the message to the 17,951 who signed up in the Chat category, a subset of the Internet group. Tracking your own unsubscribes is critical when sending out your own corporate newsletter. Company branding, customer retention, and good old online relationship improvement all come to bear.

Emails with Clickthrough Response

As with online ads, the next phase in email marketing is measuring how many people respond with a click. Your server logs make this a very simple task if, again, you include a traceable link in every message. You then need to ensure that your server log analysis tools are up to the task of letting you slice and dice the results to show which email messages are bringing in the most/best people.

Before working with a third party, you'll want to take a close look at their reports to be sure they fill the bill. YesMail does a fairly good job at spelling out the results of a campaign, as well as the financial significance of those results

Emails with Good Timing

When is the best time to send out an email come-on? Don't forget this little element when comparing the results of your marketing genius. A newsletter that includes the deer-resistant plant of the month from www.mydeergarden.com is better received over the weekend than the "Corporate Trust Connection," the quarterly newsletter from U.S. Bank Corporate Trust Services (www.usbank.com).

Promotions, on the other hand, work well when timed together. ConsumerMarketingBiz (part of Anne Holland's MarketingSherpa.com empire) reported on Eric Welter's progress as the vice president of marketing at Geerlings & Wade, the online and offline wine merchant: Welter worked closely with his print broker, Ballantine Litho-Sales Inc., to time emails around his direct mail campaigns. He says, "We've been a direct mail company for 15 years, so we already knew the optimal time to send postal mail. It's the timing of the email around and in coordination with the mail schedule that makes a difference. You'll have a piece of mail in your postal mail box, then the next morning you'll get a similar promotion in your email reminding you that you're two clicks away from placing an order."

Tracking Viral Email

It's such a wonderful idea. You send out an email to a handful of people and they each forward it on to a handful of people and they forward it on, and before you know it, the whole world is abuzz with your offer/game/contest/viral marketing brilliance. Only two problems. First, you have to come up with something so appealing, so compelling, so tantalizing that people actually want to forward it to friends and family. Second, measuring how that message gets passed along is not a piece of cake.

The first part is one for what we used to call the smoke-filled room. Pure creative genius coupled with absolute blind luck is the only way a viral campaign can be wildly successful. Oh, you can plan it, and schedule it, and budget it all right, just don't plan on spending the profits from your results. To say "results may vary" is a gross understatement. There's just no way to tell in advance if your shrewd marketing concept will be as widespread as All Your Base Are Belong To Us (which now produces 148,869 results on AltaVista and about 379,000 results at Google) or as unheard of as the thousands of intentional viral efforts that go nowhere every year.

Given an adequately powerful meme you can start an avalanche of interest by sending out just a few messages. Each one contains a unique link with a unique landing page. You can easily count how many people went to which landing pages, but you can't trace the viral network itself, only the source, the seed.

Let's say the seed message is sent to a hundred people. Even if only one of those people passes it on to a group of excitable viral communicators, you can still have a landslide of interest, but all you'll know is which of the original recipients was the Typhoid Mary of the gang. It may well be that the original recipient passed it along to somebody who passed it along to somebody who sent it to Mary.

The only way to truly track the viral progress of such a marketing campaign is to include some device that requires the identification of the previous sender. If, in order to register for the contest, join the discussion, get the discount, or download the game, participants must identify where they heard about the offer, they might be willing and able to give you a bit more information than you've been able to acquire to date- might. If I got it from Larry, and he got it from Jerry, and Jerry got it from Mary, then I can't tell you that Mary was the one who makes the whole thing viable. I can only tell you about Larry-if I have a mind to.

The best viral marketing requires the least amount of effort for the propagator. Forwarding an email message is the easiest. If that message has a link that takes the recipient to a page where they have to fill out a form, you've thrown cold water on the whole process. Every time you add another hurdle, you add another viral vaccine.

Email Auditing

Just as there are companies to help track the veracity of online advertisers, we now have services from ABCi to track email. ABCi announced their new "effort to stamp out concerns over list accuracy and delivery validity-and perhaps help produce some standards along the way," according to an article by Christopher Saunders from November 1, 2001, on Internetnews.com (www.internetnews.com/IAR/article/0,,12_915291,00.html):

For direct response service bureaus and email distribution companies, the Chicago-based ABC Interactive said it would study the systems used to create and deliver email campaignsanalyzing internal processes and reporting standards.

Meanwhile, ABCi also said it would provide auditing services for list managers and marketers that would review the results of specific campaigns. ABCi said it would be able to verify the accuracy of lists' purported size and demographic makeup, their compliance with stated privacy rules, agreed-upon merge/purge activity, delivery and results.

[The hope is to] help the email marketing industry by weeding out inaccurate procedures and players, and by making it easier for an apples-to-apples comparison of different vendors. Already, the group is working with associations like the Interactive Advertising Bureau to hammer out standards governing ad serving metrics and reporting, and it is huddling with the Advertising Research Foundation's Digital Media Measurement Council to create audience auditing standards. Now, it said, it plans to do the same for email.

Hope springs eternal.