Ever heard of the "Mac tax"?

It's the long running notion in technology circles that consumers are willing to pay more for Apple's technology over other tech brands. But now there's a new aspect to the story: Travel site Orbitz offers Mac users pricier deals than PC users, according to a recent Wall Street Journal report. By tracking people's habits, the site found that even details like the type of computer in use can help predict the types of purchases people will make.

Orbitz found Mac users on average spend $20 to $30 more a night on hotels than their PC counterparts, a significant margin given the site's average nightly hotel booking is around $100, chief scientist Wai Gen Yee said. Mac users are 40% more likely to book a four- or five-star hotel than PC users, Mr. Yee said, and when Mac and PC users book the same hotel, Mac users tend to stay in more expensive rooms.

It may sound like cutting-edge technology--and it is, in the sense of how much data a business can consider, the range of information available on consumers, and the speed at which analysis can be done, with marketing guidance offered in real time. With the ever increasing power of computers, the availability of cloud computing resources, and the analytic muscle of third-party software, such insight is available at prices that even smaller companies can afford.

But the concept is much older than it seems. Direct marketers have made use of the idea for many years. And they've learned lessons that entrepreneurs should take to heart. The current ability for online analysis can easily make some pretty dumb things look attractive.

Avoid differential pricing.

Charging two different customers prices based on what a company can realistically get from them is something companies have done in decades past. Doing so is even easier today, when you can match the Orbitz-type analysis with a pricing engine to maximize total margins. But don't be tempted--because it's a mistake.

Technology may aid the marketer in this case, but it also provides consumers a boost. It becomes way too easy for automated sites and social networks to make your differential pricing strategy transparent to buyers. Now you have to explain why you're charging one group more than another, and good luck with that. Instead, do what Orbitz does, which is keep the same prices but change the priority of how you show them, with the option for people to still look by price and other factors. That way, you don't leave people feeling like they got the short end of the stick.

Go from the general to the specific.

Patterns from large numbers of consumers can be revelatory. You can get a better handle on how segments of the population actually act, think, and decide. That lets you design or select products, structure services, and market far more intelligently. But you make a critical mistake when you take the fallacious step of assuming that because most people behave a given way, all people do. They don't.

Learn from an individual's behavior and let that trump the general schemes. When people have told you what they really want through what they look at and buy, why would you insist with general tactics that they don't know what they're talking about? There are undoubtedly people with Macs who prefer bargains, so make sure they get what they want as well.

Software doesn't create expertise.

The tools to do speedy and complex analysis are within the reach of most companies. However, the expertise isn't. While you can get a lot of value from big data analysis, it's easy to head off in the wrong direction by misunderstanding which factors influence others or to confuse things happening at the same time with causality. (For example, it could be that a third factor affects two different data points at or that the apparent association is just coincidence.) Consider getting some real expert help in setting up your analyses in the first place.

Keep these points in mind and you'll find that you can make broad analysis of customers and consumers in general work for your business without accidentally digging a commercial grave for yourself in the process.