The survey has been taking a beating recently. According to the blogosphere, people are questioning whether primary MR will even still be around in 10 years, often under the assumption that Big Data will eliminate any further need for asking people questions.
It’s a strange assumption to make (could it be influenced by hype?) in a world where consumers are increasingly spending their time researching products before they purchase, asking questions and swapping ideas with each other.
Here are four assumptions the naysayers have made, and four reasons why the survey is here to stay.
4 pessimistic statements made about surveys
- No Vision. It’s been said that 80% of MR and surveys look backwards, with an aim of solving problems and defining strategies that are forward-looking. It doesn’t make sense, they claim.
- No Viability. In the 20% of surveys that are used to predict the future, results become very hit-or miss. Just look at recent polling disappointments to see the increasingly questionable ability of surveys to looks forward. Why use such an untrustworthy form of MR, they ask.
- No Value. As the hype train for Big Data boards, we’re hearing more an more about its ability to connect real consumers with everything they buy or think online. That seamingly limitless value is incomparable to surveys, they say.
- No Variability. Today, it’s getting increasingly difficult to recruit a sample of representative respondents for a reasonable price. If these samples aren’t random and representative, how can we trust the survey results, they demand.
The best ways to use surveys
The above assumptions, while based mostly in fact, ignore the vast potential of surveys for collecting data that is simply impossible to gather in any other way.
- Simulate the future. Using interactive simulation is so hot right now. High-tech methods simulate future events (eg. products, regulations, market changes). Then researchers use surveys to track what consumers would think about these changes.
- Executing experimental designs based on multivariate analytics. Techniques such as choice methods and MaxDiff rely on the visual interplay of attribute combinations or word pairings that work best in a simple visual setting that can’t be replicated in any form other than a survey.
- Filling in missing pieces. Predicting purchase behavior at an individual level relies on an array of independent measures, including brand attribute ratings, demographics, and descriptions of critical goals and targets. There are no better ways to collect this information than a targeted survey.
- Providing inputs for more complex models. Survey-derived probabilities are often used to define key parameters for the Monte Carlo and Bayesian polling models that are being increasingly used today.
The age of the survey has certainly not passed, but the uses and applications of surveys as a vehicle are definitely in a rapid state of evolution (as is every element in our industry).