By Carl Carter, APR
There’s no substitute for solid data about your customers, your prospects, your buyers, your sellers, and any other segments that can make or break your business. But bad data can lead to bad decisions that can hurt or even sink your business. So let’s talk about how to avoid that. These are some of the most common mistakes I see when companies try to do market research without understanding it.
Standardized quiz or test score sheet with multiple choice answers
Asking the wrong people. You’ll never get the right data from the wrong people. Obvious as that may seem, failing to properly define the population being researched is the single most common mistake I see. And the resulting errors can really hurt your business.
So who is your population, and how are you reaching them? This is a trick question, because if the answer is, “people who respond to the questionnaire,” you’re making two big mistakes. The people who respond are your sample, which you hope will represent a larger population (more on that in a moment). So define your population first, without worrying about how you’ll poll them. It might be people who register to bid. Or just those who actually buy. Or sellers. (Successful or unsuccessful?) Or restaurant owners within a 25-mile radius of Decatur, Alabama.
Recruiting in a way that precludes a valid sample. Now, let’s get back to that sticky matter of how you’re reaching them. There’s a huge yawning trap waiting on you here, because the easiest way to get responses is to have the respondents actively step forward and select themselves to be polled. We call these “self-selected” respondents, and they’re your worst enemy, because they already differ in some significant way from the population you’re surveying. Find a way to randomly sample your defined population.
Building bias into the questions. This whole business of market research can look very objective and scientific, but writing neutral questions is very much an art. Even if you have some training and experience in instrument development, it’s very hard to write questions that won’t get slanted answers. For that reason, it’s often best to hire a professional for some help.
Relying on inadequate sample sizes. If you know the size of your population, you’ll need a big enough sample to get a decent Margin of Sampling Error (MOSE) at an appropriate confidence level. If you don’t know these two numbers for your survey, you should not act on your data until you do. Let’s say you’ve got a population of 5,000 recent auction buyers, and you want results you can trust. You’d need a sample size of 357 to achieve +/-5 percent at a 95% confidence level. This means that 95% of the time, you’ll be off by no more than 5 percent in either direction due to your sample size. (Remember, this all assumes you have valid questions and a truly random, non-self-selected sample.) You don’t have to get a graduate degree in statistics. The Internet is awash in free apps that will calculate your sample size for you.
Relying on people when objective data is available. This one occurred to me shortly after I completed the a recent seminar, which includes a lot of good information on tracking results by using custom URLs and phone numbers, as well as other tools. It’s always frustrated me that many businesses are content to evaluate their marketing vehicles by simply having someone ask, “Where did you hear about us?” The person asking will usually offer a short list that includes things like, “Internet, newspaper, TV, sign.”
Which tells you pretty much nothing. And hardly anybody ever seems to drill down with follow-up questions like, “Where on the Internet? On your laptop or smart phone? Was it an ad, news story or Facebook post? Which site?”
You get the idea. To make things worse, we tend to ask these questions at the two worst possible times: When people are calling for information (not to be quizzed), and when they’re at the front of a line to register at an event.
Here’s a better idea: Learn to use custom URLs and phone numbers to track where your visitors and bidders come from. This will enable you to see where you’re wasting your money and concentrate your budgets on what’s actually working, rather than foggy or vague human memories.
*Adapted from an article originally written for Auctioneer Magazine, March 2016, used by permission.