J. Timothy King’s Blog

Web Site Mis-Marketing

I recently read on Yaro Starak’s excellent Entrepreneur’s Journey blog a guest piece by John Webster on analyzing web stats. I’m peeved. He didn’t actually say anything wrong, and he is the expert. In fact, I don’t take issue with any of his guidelines. But I do take issue with his advice.

First a humorous note: With poor proofreading, “you are probably loosing a lot of business.” I’m not quite sure what that means, but it’s probably bad. After all, we can’t have all that business running around loose all over the place. (By the way, there are at least 2 additional errors in this piece. Can you find them?)

Less funny, but more interesting, are the guidelines.

A well constructed and implemented web site should expect 50% - 70% of its traffic to come from the SEs.

He states this as though it were axiomatic, but it is not. What if my business model needs most of my traffic to come from networking, subscriptions, or brand awareness? What does CNN see? Do half of their web visits really come from people searching for the news? Maybe people searching for yesterday’s news. But if these percentages are off, I wouldn’t be too concerned. Because ratios like this assume a business model. First, I have to define my business model and brand message. Then I can look at search-engine visits and other stats as indicators within that context.

Basically, the more pages a site contains, the less frequently should the home page be the first entry point.

I wonder what Amazon.com has to say about this. Actually, Amazon.com doesn’t really have a home page. Or rather, every page is a home page. And probably few people do enter through their front door.

And if your web site is well designed, you probably will indeed get 50 to 70 percent of your traffic from search engines. If I get few search engine hits, this may indicate I’m not communicating my brand message effectively, or that my networking efforts are too narrow.

So what John says is all true. But it serves as unsound advice to the entrepreneur.

One of my favorite stories comes from Peter Drucker’s book Innovation and Entrepreneurship. In the 1950’s, the top New York department store Macy’s was inundated with people wanting to buy appliances. As Peter Drucker tells it:

I was told by the chairman… R. H. Macy, “We don’t know how to stop the growth of appliance sales.”

“Why do you want to stop them?” I asked, quite mystified. “Are you losing money on them?”

“On the contrary,” the chairman said, “profit margins are better than on fashion goods; there are no returns, and practically no pilferage.”

“Do the appliance customers keep away the fashion customers?” I asked.

“Oh no,” was the answer. “Where we used to sell appliances primarily to people who came in to buy fashions, we now sell fashions very often to people who come in to buy appliances. But,” the chairman continued, “in this kind of store, it is normal and healthy for fashion to produce seventy percent of sales. Appliance sales have grown so fast that they now account for three-fifths. And that’s abnormal. We’ve tried everything we know to make fashion grow to restore the normal ratio, but nothing works. The only thing left now is to push appliance sales down to where they should be.”

(This story is typical of Peter Drucker’s style, engaging and insightful.)

At about that time, Bloomingdale’s was a relatively minor player in the New York department store scene. They saw the same phenomenon Macy’s did, but Bloomingdale’s reacted differently. Bloomingdale’s re-examined the market, discovered that the market was changing, that there was a new kind of customer coming into its stores, a customer who was buying from a new perspective. They revamped their marketing and brand message to speak to this customer.

So what happens if you get too much traffic from non-search sources, and you try everything you can to increase search results, but your customers just ain’t searching on what you’re writing? What do you do? Turn away customers? Or re-evaluate your assumptions?

My big beef is that John’s piece assumes there are “correct” results and a “correct” strategy. But there is no one way. You need to define a strategy to serve your market, try it, measure the results, then adjust the strategy to leverage the results. Web stats are a key part of this cycle. But it is a cycle, a continuous cycle of tweaking and measuring. Whatever the results say, if they don’t line up with your expectations, that’s your customers trying to tell you something.

-TimK