Known Respondents Yield Quicker, Less Expensive Survey Results


If you were heading to the mall to get a new smartphone, would you go randomly from store to store asking whether they sell smartphones? Or would you want to know which store sells them and to go straight to that store?

Chances are you don’t want to a) waste time going from store to store, or b) look like a fool by asking the clerk at the women’s clothing store if they sell smartphones. You just want to get straight to the smartphone store, right?

It’s the same thing with surveying people. If you wanted to get feedback from people who play basketball, you would want to start with a sample of people who you know play. Seems simple, right? But that’s not how most surveys are done.

The vast majority of people who answer surveys come unscreened and unknown, from what are called river sample sources. To understand anything about them—like whether they play basketball—you have to ask them. About 7% of the U.S. population plays either pick-up or league basketball. So that means that for every basketball player you get to do the survey, you have to reject 13 other people, after asking them the question. That’s a terrible respondent experience.

Most people willing to do a survey face a barrage of questions that may or may not lead them to a survey they qualify for. It is very common for people to be asked their age, gender, education, region and other demographics, before they start getting asked things like whether they play basketball or play backgammon or own a small business, or any number of things that they are likely to say “no” to. It’s not exactly a great way to signal to the respondent that we value their opinion and time.

We know that a poor respondent experience leads to poor quality information, which leads to bad decisions. That is exactly the opposite of what research is supposed to do.

In contrast to standard practice, we recruit people to join our market communities. In doing so we profile them extensively and then keep all that information so that we don’t have to re-ask it. We can seamlessly fuse it with survey data. In addition to screening people when they join, we also screen, on a monthly basis, for much more detailed things, like playing basketball, playing backgammon or owning a small business. This allows us to develop a rich profile of each respondent, one we can draw on anytime we need to find, say, basketball players.
The profile for each respondent also extends to their answers to individual questions within a survey. This can be incredibly helpful, saving time and money and making sure we are only talking to the right people.

For example, our MoneyScreen Payments testing benchmarks and tracks payment innovation across all aspects of the industry. Offers for credit cards and other payment methods are evaluated in an extensive online survey where consumers are asked to evaluate one offer at a time across a series of diagnostics. We often return to people who we know responded well to a particular offer, so that we can do follow-up message optimization research. It’s a powerful advantage when you know you can ask the right questions of the right people, quickly and efficiently.

We recently did a study with Canadian Tire Retail (CTR) where we took advantage of this knowledge of respondents. They needed to find a very specific and rare type of consumer. Cedric Painvin, Associate Vice President, Consumer Research Customer Insights & Analytics at Canadian Tire Retail explained it like this: “Like all companies, CTR is always looking for ways to cut costs. Finding efficiencies in how we do market research is no exception. Targeting hard to find customer groups can be very expensive. In a recent study, CTR was able to work with the [Angus Reid Forum] team and utilize a ‘back to sample’ approach. This allowed us to target a low incidence segment and know (with greater confidence) that their behaviour would match the demographic we were interested in. What would have been a costly and lengthy project, was completed quickly and at a significantly reduced cost vs. traditional methods.”

By using known respondents we are able to save time, reduce costs, have happier respondents and get better quality data. That’s a winning combination.

To learn more about our market communities and the value of known respondents contact us.