Dot-com Comments From a Cynic
Here are some thoughts on the state of Internet marketing from a cynic.
Portal deals suck, sometimes. A year ago, everyone got their financing and rushed to get a portal deal - no questions asked. "We have to have a presence on AOL or AltaVista," they all said, without thought to the kinds of deals they were getting. I spent a great deal of time helping clients like these renegotiate the portal deals they blindly signed up for.
Portal advertising should be approached with the same test mentality that we'd recommend for any direct marketing medium. If you want to affiliate yourself with Excite, Yahoo!, Lycos or any of the other portals, you should work with the portal representatives to test a response message throughout various areas of the portals, then negotiate your large deal based on response and conversion from specific testing areas.
But any broad portal deal, without testing, sucks.
E-mail works, sometimes. E-mail that targets the wrong offer to the wrong person is spam. I never heard anyone refer to an e-mail with something that interested him as spam. So it's simple: The right message to the right market with the right offer means a good e-mail campaign. And remember: You can do everything right on the front end and get a lot of prospects to click a link that takes them to a Web site that has nothing to do with what you're trying to sell. Be sure that when people respond to your e-mail, they arrive at a Web page that complements it. You'd be amazed at how many times this doesn't happen.
Banner ads never work, sometimes. Anyone who says that banner ads work is running a porn site. I've tested many banner ads for many of products and clients, and I have yet to see one pull a positive ROI over an extended period. Anytime we had a banner ad that pulled a good response, we saw that response trail off after the first 30 days, which is a lousy half-life for a control position. We now use banner ads more for limited-length promotions of no more than 60 to 90 days. You get the ads up, they do their job, you take them down.
Using the Web to build your brand doesn't work. What makes the Internet unique is that it's a response vehicle. If you use the Internet just to build brand, you leave your viewers unfulfilled. What's the use of telling them who you are if you're not going to tell them how to get it or what to do?
The Web is a very tactical medium - it's reactive. So you have to ask yourself, what comes first, sales or brand? Mr. Kraft probably didn't say, "I'm going to build a brand and then sell macaroni." Television, magazines and billboards are brand builders. They're broad; they reach everyone. They don't demand a tactical response.
Here's the bottom line when it comes to comparing traditional versus direct advertising: If I get them to know who you are, I can't guarantee you that they will buy. If I get them to buy, I guarantee you they know who you are.
"We want to build our brand."
"How much money do you have?"
"About $2 million."
"What state do you want to build your brand in?"
When you have limited investment money and you have to make every dollar count, ask yourself: Do I want awareness, or do I want customers? It takes a great deal of money to build a brand, particularly when your product or service appeals to a broad spectrum of people across the nation. And I'm talking about much more than $2 million to $10 million. When you have a limited marketing budget, you need to concentrate your efforts by targeting the best prospects and converting them to customers. That's a direct campaign, not a brand campaign. It all comes down to this question, which I've asked of a lot of clients: When are you going to quit telling them who you are and start asking for the order?
"Marketing plan? Yeah, we're working on it, but we need some advertising right now." If I've heard this once, I've heard it a hundred times. And then the person who says this usually tries to explain it by telling me that they're working on Internet time and can't wait for a plan. My normal response: "Go find another agency." If your marketing executive has been with you three or more months and you do not have a marketing plan with strategy and tactics, fire the marketing executive. If you're the marketing executive and you've been trying to get your strategy and tactics through company management for three months, resign.
Most venture capitalists know squat about marketing, so ignore their advice. They don't ask me to raise money, and I don't ask them to write marketing plans. And who's the idiot who gave a bunch of kids $100 million?
Are you a retail operation, or a library? Content's good, sales are better.
If I want community, I'll go to my local bar. If I want content, I'll read a book.
And if I have to have content, it had better be linked to product sales.
"Just in time" inventory systems will create a back-order nightmare. If you're out of stock, get it off the site.
Americans require instant gratification. That means if it takes a week to get something from your Web site, they'll bitch that it feels like two weeks. (Two weeks = forever in the Web world.) "Just in time" inventory is a great concept if you're a car manufacturer, but in a Web commerce situation, you'd better have some extra product inventory and you'd better be shipping within 24 hours if you intend to survive. If you run out of inventory, you're better off removing it from the site. That's the advantage of the Web over a printed catalog - you don't piss off your prospective customers.
Fulfillment operations must be under your control, preferably inhouse. Your fulfillment operation is part of your customer relationship management team. The warehouse should be working hand in hand with your CRMs. If you're using drop-ship operations, you simply cannot manage your customers' expectations or respond to your customers' needs as quickly as required in the interactive commerce arena.
Customer service is the heart of a retail Web site. Live customer service is an up-sell opportunity.
If somebody asks you about shoes, sell them a pair of socks.
Senior management should be required to read every customer service e-mail you receive. I don't think much of managers who sit at their desks waiting for somebody to tell them they have a problem. If every morning, when you open up your e-mail, there's a large random selection of customer e-mail waiting for you to read, you'll know exactly where the problems are in your organization, and you can get them fixed. Look for problems and you'll find them.
• Jon Roska is founder/CEO of Roska Direct, Montgomeryville, PA.