Article > Opinion

Dissecting a Survey

By Jon S. Owens

The aftermarket loves conducting surveys and analyzing the results. But, more often than not, the data is biased and inaccurate. It’s time we begin focusing on the facts.

One thing we never seem to get our fill of in the aftermarket is survey information. We survey everything from buying trends to brand preference to inventory investment. Over and over, we ask and ask and ask. I commend our industry for wanting to know, but I wonder what we really get.

When I was in marketing for a large parts manufacturer, we used to say “output is only as good as the input provided.” Here at Counterman, we have our own in-house research department and I am very glad we do. When we conduct a research project, like our annual PARTS Supplement, I work closely with our editor Brian Cruickshank and our research expert Bob Roberts to construct the questionnaire and choose the parts categories to be surveyed each year. Consistency is critical, otherwise the data that’s been collected over the years becomes compromised and unreliable. An even more critical component is the way a question is stated or presented and the order in which questions are answered. Thus, the way you go about gathering the “input” for any research project or survey is the most critical step. 

Other key factors with regard to good survey work include (but aren’t limited to) response rates and data analysis. Without a high response rate (usually considered to be a 10 percent sampling or above), the input you’ve gathered is somewhat flawed because it is not representative of enough constituents or market participants. Accurate analysis of the data becomes a rather simple task, if the proper thought was applied to the construction of the survey up front. However, when the data reveals some things you would rather not know or aren’t happy with, then data analysis can become skewed and biased and real results can be excused away as “bad data.” In essence, sometimes the truth hurts!

So, why am I rambling on about data and research and survey results? Well, because quite frankly I’ve seen a lot of bad data dumped out in to the marketplace recently and it is quite alarming to me. In some conversations I’ve had about my concerns with this information, I’ve been told that it’s not the data that matters, but rather it’s how you package and present it. Kind of like putting lipstick on a pig!

I’ve also seen many examples of data that is being collected in a silo, with no consideration to all of the market factors that may impact or influence the results. Consensus building is difficult in such a competitive market, but an outright disregard for things that are generally accepted as normal operating conditions, general principals and widely known and accepted market behaviors is “silo” research and “silo” analysis at its best.

Here’s my advice to anyone in our market using data for strategic planning: Question everything about it, thoroughly and completely. The more informed you are about the methodology of the research and how the data was collected, the better. Otherwise, you might find yourself and your business plans waking up out in the barn next to some very pretty farm animals.

Advertise     Contact Us     Subscribe    
Babcox Media •
3550 Embassy Parkway, Akron, OH 44333
330-670-1234 • (FAX) 330-670-0874
Babcox Website Counterman: Home