Sex, Power, and Money: Dave McClure Tells Web Startups to Tap Into Consumers’ ‘Reptilian Psyche’

7/9/10Follow @gthuang

Dave McClure is a dynamic guy. I had the pleasure of hearing him speak to a small audience of tech startup founders in Seattle earlier this week. His impassioned message to entrepreneurs was, first of all, to take risks and be ambitious, and second, to build something that people really, really care about.

That sounds obvious, but when you’re a developer slaving over a piece of code, or a designer tinkering with a new look for a website, or a CEO trying to make money to pay your employees, it can be easy to forget that most people visiting your site (or otherwise perusing your services) won’t care about whatever new feature you’re working on.

Of course, every tech company from Amazon to Apple says it focuses heavily on its customers—and always has. But what McClure is talking about is something deeper, something primal.

McClure, a self-professed Silicon Valley geek, angel investor, fund manager, and startup advisor, has amassed a large following among techies and entrepreneurs. He has advised or invested in more than 60 companies in the past six years, including Mint.com (acquired by Intuit), KissMetrics, TeachStreet, Twilio, and WePay (which my colleague Wade just wrote about yesterday). Back in 2001-2004, McClure was director of marketing for PayPal. He has expertise in microfinance, social networks, consumer Internet strategies, and other areas.

There was some overlap in his Seattle talk with what he has said elsewhere—including at Dogpatch Labs in the Boston area in March—about the kinds of metrics that Web startups should pay attention to. But I pulled out a few more nuggets that I think all business leaders interested in innovation on the Web should listen to, even if they might disagree. (These are just my takeaways, they’re not representative of McClure’s whole talk or philosophy.)

First, McClure stressed the importance of building something meaningful, and getting the big picture right. There’s no sense in iterating small things about a product and testing different versions with consumers if you’re not in the right ballpark to start with. And that’s something that can be tested out by gauging just a few people’s reactions to your idea. If no one seems interested, look for a different concept.

Once you’re on what he calls a “meaningful hill,” you can start testing out different looks and features with a small number of consumers (say, 10-20 to start), to gauge whether what you’re doing really matters to them. Ideally, he says, you should talk to potential customers before you even start coding. This is related to “driving usage before trying to acquire users,” he says—a crucial distinction that means, basically, make your product matter before you try to sell it too hard.

And to make a product really matter, McClure says, entrepreneurs need to “tap into the reptilian psyche” of consumers. Make your service—and the visual elements on your website, if it’s Web-based—appeal directly to people’s primal need for things like sex, power, money, fight-or-flight response, protecting loved ones, and so forth. Simplify the interface and focus on one or two aspects, making use of people’s natural visual affinity for things like faces, buttons, and edges. (McClure didn’t give the following as examples, but I think of the “gamification” trend of services like Foursquare as appealing to power; companies like PayPal, Mint.com, and WePay as appealing to money; and some news media sites and porn sites, obviously, as appealing to sex.)

Lastly, as a developer or founder, when do you know you’re done with a given product cycle? It’s when you have a meaningful offering that’s “easy to find and better than the alternatives,” McClure says. “You’re done, market it!…Stop building [stuff]!”

Gregory T. Huang is Xconomy's Deputy Editor, National IT Editor, and the Editor of Xconomy Boston. You can e-mail him at gthuang@xconomy.com or call him at 617-252-7323. Follow @gthuang

By posting a comment, you agree to our terms and conditions.