« Another reason to network | Main | Marketing can be measured in inches not miles »

When the survey is worthless (Part 2)

Checkmark- Tom Vander Well, executive vice president of c wenger group, is a recognized customer service authority in the contact center industry.

How hard can it be to survey customers? It seems like such an easy thing. There are so many DIY websites out there. Ask a few questions, design the questionnaire and send it to the email list. The website will even calculate the responses for you.

DIY websites are great, and for many projects they are just perfect for the job. However, a company that needs information on which to make strategic or tactical business decisions, needs to be  careful. Many do-it-yourself surveys are enamored by the sheer numbers. Send out a survey to all your customers, offer a chance to win a gift card, and you're ecstatic to get 1,000 responses.  It seems like everything worked perfectly.

But it really didn't, and you had better be careful.

Let's say the 1,000 replies is 1 percent of the 100,000 customers to whom you e-mailed the survey. While it seems like 1,000 responses is a lot, the truth is that it's almost certain that those 1,000 customers are not representative of your entire customer population. You'll end up with good data about customers who are really happy, really angry, like to respond to e-mail surveys or who would really like to win a gift card, but it's almost a sure bet that they don't represent the other 99,000 customers as a whole.

Here in Iowa we are inundated with political polls leading up to next year's Iowa caucus and presidential election. On the news you'll see pollsters provide results for how America thinks by surveying a thousand or so people. It seems ludicrous to think that 1,000 people can provide an accurate picture of how a country of 300 million Americans are likely to vote if the election were held today. The truth is, they can. That's not to say that all polls are accurate, but if the survey is conducted properly you can actually get an accurate picture of how things are likely to shake out. But it has to be done properly by expert pollsters asking the right questions of a representative sample of likely voters. And that's where it gets complicated.

I am a firm believer in making strategic business decisions based on data. The data has to be accurate, however, or the decisions I make are worthless. Surveys are a great way to gather data, but be careful how you go about doing it. You may end up with a lot of impressive charts and graphs that have nothing to do with what your most important customers think.

Comments

The comments to this entry are closed.

« Another reason to network | Main | Marketing can be measured in inches not miles »

Technorati Bookmark: When the survey is worthless (Part 2)

This site is intended for informational and conversational purposes, not to provide specific legal, investment, or tax advice.  Articles and opinions posted here are those of the author(s). Links to and from other sites are for informational purposes and are not an endorsement by this site’s sponsor.