There’s no single answer to this problem because everybody’s needs are different, but here are some general guidelines.
1. Ask if technology is really the answer.
Technology is great and allows us to do amazing things. But remember that when your only tool is a hammer, you want to bang on every problem, and when your only tool is technology, you want to program it all away. Sometimes the solution lies elsewhere.
For example, if you’re having trouble getting web visitors to sign up for your premium content, you may need to do A-B tests on your sign-up form, or you may need a better registration process, or you may need technology to target the offers more carefully.
Or maybe your content stinks, or your offer is horrible.
Before you sign on with some whiz-bang technology answer, make sure somebody on your staff takes a hard look at the problem the old-fashioned way. It’s too easy to believe that the new gadget will solve your troubles. Sometimes there’s no substitute for doing the hard thing.
2. Hire a marketing technologist.
Somebody on the marketing staff has to understand technology. I’ve written about this before, and at the bottom of that article you can see links to some rather awful videos where I reviewed my notes for a German friend.
3. Write an incredibly detailed requirements document and go over every single point.
When you decide that you do need a technology solution to a problem, download 100 white papers, product comparisons, blog posts and various mind dumps that deal with the issue, get on an Amtrak to some city that’s three or four hours away, and read every single word. Get one of those fancy pens with a highlighter on one end and a regular pen on the other and note issues, features, problems, difficulties, questions, etc.
When you get where you’re going, have a sandwich, review your notes and take a long walk. Let the stuff settle in your brain for a while.
On the trip back, sit in the dining car and get a couple beers. Start organizing your notes into an outline.
Yes, I’m serious about the beers. At this point you’re trying to think creatively. How will this thing fit in with your system? Where are the likely problems? What are you going to do about Alfred in IT who insists on programming everything himself? How will this interface with your other technologies and vendors?
Go home and sleep on it. Every day for the next week you’ll come up with new things you didn’t think about. Jot them down. Incorporate them into your outline.
On Friday, set the thing aside and try not to think about it all weekend. You’ll fail. When something occurs to you over Sunday breakfast, email it to yourself and go on trying not to think about the topic. On Monday, look it over again and prepare something close to a final document.
Now, and only now, you’re ready to talk to vendors. Send them your document and ask what you’ve missed.
They’ll try to snow you. They’ll try to sell you 10 times more than you need. They’ll tell you it’s easy, that they’ve done these integrations before, that they can work with everybody, etc.
Don’t believe a word of it. Get past the sales guy and talk to one of the operational people — the ones who know how the system works. Get them on the phone with your operational people. Talk through every single detail of your requirements document. Don’t accept “yes, we can do that.” Probe. Find out how they do it, because the devil is still in the details.
After you’ve done all this and you’re pretty sure you’ve got it all figured out, realize that the devil is one sneaky fellow. He’s still hiding in those details, and he’s gonna get you.
4. Have a workaround.
Despite your best efforts at figuring out every possible glitch, there will be glitches. If possible, have a way to work around them.
For example, if the data doesn’t flow from vendor A to vendor B the way it’s supposed to, make sure you can download and upload it yourself. If it turns out that widget #3 actually isn’t compatible with the server software you’re running, have a back-up option for that process.
Think creatively. Don’t get boxed in. And don’t believe anything anybody says.
What, … you thought it was going to be easy?
