Killing the Goose (Part 1)

Remember email marketing?

Well, you don't just remember it; more than likely you are doing it right now. And it's a useful tool. But do you remember what it was supposed to be - a revolutionary new approach to marketing that was going to change the way businesses related to their customers, promising profitable long-term relationships with customers for those who mastered to new way of thinking.

It didn't happen.

Okay, it was the late 90s, so maybe the promises were extreme. That was also when businesses didn't need to make profits, or have any hope of ever doing so, in order to justify their sky-high market valuations. That was when Kozmo.com seemed like a reasonable business idea. So perhaps the initial buzz about email marketing was part of the hype.

Even so, email marketing was supposed to be something very different than it is today. Consider the idea and the current reality.

The idea: businesses would leverage database technology to learn about their customers - what they bought, when they bought it, what they did not buy, why they made certain purchasing decisions, and most importantly, what kind of relationship they wanted to have with the companies they bought from. Based on this information, businesses would market intelligently to customers in ways that built relationships. During each interaction, marketers would ask for a little more information from customers, and ask for permission to send them email: email that contained useful information, based on that ever-growing database. The benefit for customers was that they would be dealing with intelligent vendors who only bothered them with communications that were personalized, targeted, useful, interesting and which made it easier for them to make the purchases they want to.

The guru of this concept, Seth Godin, saw the writing on the wall for traditional "broadcast" marketing. With consumers bombarded by marketing messages from all directions - from the usual television, radio, and print media to signs plastered on every bit public space, including the wall above a urinal - trying to build brand awareness that way was an increasingly expensive proposition, with ever-declining return on investment. Godin was right. The idea of instead targeting your best customers with intelligent information was revolutionary.

So what happened?

The reality of email marketing in 2004: businesses try to get permission to send email. Since this is hard to do with a population that overwhelmed by email, both the legitimate and the spam varieties, they play games to make people "opt in" in ways that range from questionable to downright sneaky. Then, they ignore everything they know about customers, and send them generic messages, which customers as often as not simply ignore. As spam filtering technology improves and starts to block these messages before they are even received, marketers look for ways to evade filters of get uninterested consumers to add them to "white lists" so that messages are delivered to inboxes (before being ignored).

Really - what happened?

There are several reasons that the term "permission email" has become, for the most part, a bad joke. Some are legitimate; some are not.

First, permission marketing is hard. It requires you to gather a lot of customer information. That's the simple part. Then you have to understand that information. That means investment in database technologies and applications to parse this enormous pile of data. That's expensive and challenging. Then you have to make decisions about what to do with that information. What do you say to whom? The world of simple marketing campaigns became a migraine-inducing landscape of campaigns that splinter into sub-campaigns based on customer information, so what starts with prospects clicking on a banner ad can lead to a hundred different variations.

So we had a challenge that would take money, creativity, and brains to tackle. No problem - who doesn't like a good challenge? Who didn't want to be part of the permission revolution? Who didn't want to be a smarter marketer than Coke and McDonalds?

Around 2000, we had a new wrinkle to this. The dot-com economy went to hell. This affected a lot of people. First, if you were a marketer for one of those businesses, your budget probably disappeared along with the value of your options, and your boss, who was telling you that you had to build the brand, pull out all the stops, suddenly wanted to know how many sales came from every dollar you spent. Reality is a drag, man.

Some of those tech companies were the people making databases and applications that would help us all deal with that mountain of customer information. Things got rocky for them, which meant fewer products and services to help us do permission marketing.

All of this would have slowed permission marketing via email down, but it wouldn't have led to the current situation, in which what is called "opt in email" or "permission email" today is mainly spam with a loophole. How did we get to that? Well, lazy marketers, mostly. When we look at declining click-through for email marketing, we have only ourselves to blame. What is going out by email today is generally not permission marketing.

First, let's look at the opt-in process. What this should be is an interaction with the customer or prospect in which you, the marketer, tell them that you want to send them some specific kind of information, and that you promise not to abuse their inbox. You tell them what they will get, then you deliver exactly that. If the customer wants it, they complete your web form, and you've got a new opt-in customers in your database.

The Seth Godin page I linked to earlier in this article is an excellent example. "Give me your email address and I will send you a chapter from my book."

But how does this really play out? Nearly every web form there is asks for your email address. Some give you an box to check to opt in. That's fine, though they often are very vague about what they'll do with that email address. Many more give you a pre-checked box - so you have to un-check it to avoid getting on the list. Not so good. And many don't tell you anything.

Even worse are the sites that hide the opt-in box in a dark corner of a form with microscopic type explaining what's going on, so that most site visitors have no idea they've opted in - until they get email. They consider this email spam. In legal terms it's not, but who cares? If the customer thinks it's spam, the marketing value is zero, and you've done nothing but alienate a potential buyer.

A really extreme bad guy here is Yahoo. Recently I was filling out a Yahoo form (providing feedback on the functionality of one of their services). It had a pre-checked opt-in box, which I unchecked. I left a required field on the form blank, so when I clicked "submit" I was bounced back to the form to complete it. When that happened, the opt-in box was re-checked. Basically, Yahoo was trying to trick me into "opting in." This is useful information for anyone who buys media: any Yahoo opt-in databases should be viewed with great suspicion.

So the opt-in process is a mess. But, if you get that right, you've got an even bigger challenge: content. The key to effective email marketing is to send customers information they want to see. This is usually not an ad. If you are going to ask people to opt in to receive your widget newsletter, you'd better have someone there writing interesting articles about widgets - how to choose the right one, the latest widget innovations, and how the right widget can improve your love life. Otherwise you'll be sending some very dull email that no one will read.

Remember all that personal data? You need to use that too. I'm always amazed when I get an email from an airline with lots of discount fares, none of which are for flights departing from Houston. They know where I live. True, I might want to know about other departure cities - but they could ask me, couldn't they? Instead I get a long list of flights, my eyes unfocus as see things like "Dallas" and "Oklahoma City" and "Little Rock," and I click delete.

If you've ever managed an email database, you know that every time you send something, people unsubscribe. The more junk you send, the more they will go away. But it actually gets worse: a lot of people aren't even opening or reading your messages. People have been trained by spammers to assume that unsubscribe instructions won't work, so they just ignore them, or (if they're more tech savvy) train their junk mail filters to just delete it for them.

One of the recent hot topics in email marketing has been dealing with filters. Large ISPs like AOL and webmail providers like Yahoo and Hotmail regularly dump our email marketing messages into spam boxes. The current version of Outlook is set, by default, not to load graphics in messages from any sender who's not in the user's address book (which blows all your reporting). These features are a headache to marketers, but they are popular with users.

Why? Because users consider your "opt in email" spam. It doesn't really matter if they couldn't successfully argue that in court - they think it's spam, so it is, and it's not selling more widgets for you.

Thus, the steady decline in clickthrough, the increasing difficulty of even getting anything delivered, and the end of the promise of permission email. Email is now just another broadcast marketing approach, like telemarketing or direct mail, that offers some degree of targeting, very crude interaction with users, and not much else. It's the dimensional mailer of the 21st century. It's a useful tool, but the revolution is over, and it failed.

(Note that there are some companies doing it well. Amazon.com is the star here, because they ask questions, send relevant email, and it gets business for them. If you want a good model to copy, go sign up for their email alerts, study the process, and do something similar on your own site.)

Why have I spent all this space talking about this? Two reasons.

First, Seth Godin was right. It is possible to use email to revolutionize the way you build customer relationships. It was hard when he wrote about it because it was new and required tools that didn't exist. Today it's hard because it's been done so badlyk, and the waters have been polluted by lazy marketers; the general population has developed resistance to email, so the hurdles are much higher.

Second, because there are always new golden gooses, and we can kill them just as fast as we did email marketing. In my next blog post I'll talk about the new golden goose - search engine marketing - and why we need to tread lightly if we want it to live up to its potential as a marketing tool.