Marketing Research, Red Herrings and Data Quality Control

When it comes to making smart marketing decisions, it’s imperative we have utmost confidence in our quantitative marketing research data. It goes without saying that dodgy data is going to lead to a dodgy marketing campaign. To ensure data quality, several things can be done:

1. Make sure the survey isn’t too long, so we don’t bore or fatigue the participants.
2. Design fun, unique and interactive surveys to promote participant engagement.
3. Use language appropriate the audience to ensure everyone understands the questions.
4. Optimize online surveys for mobile compatibility.
5. Go through each and every line of data checking for straight-liners, speeders and non-common-sense answers.
6. Throw in some trick questions!

A blog post by Luc.id highlights the power of red herring or trick questions. These questions are intended exactly as they sound- they are there to ensure participants are paying attention.

In a recent Evolve Research survey we featured a battery of ten marketing messages, and respondents were asked to use a scale of 1-10 to indicate to what extent each particular message appealed to them. We randomized every message in the battery – well, almost. One message always appeared sandwiched in the middle of the statements, and that was “please select the number 4 for this question.”

During our review of the data, anyone who had supplied an answer of anything except the number 4 was removed from the dataset entirely. Perhaps a participant was simply distracted at that particular second – maybe their dog was going crazy at the TV or their kid wanted some attention. Or, maybe that person was supplying random answers to all questions in the survey. Either way, if the respondent didn’t read the red herring question, then there’s no telling how many other questions they answered without actually reading. It’s not worth taking the risk!

You Might Also Like:

Marketing Research vs. Polling