Whenever we engage our panel to ask what improvements we—as the panel management team—could make to improve their experience, one of the suggestions we nearly always receive is, “Stop asking me for the same information on every survey.”
If you maintain your own customer panels or have conducted panel-based research in the past, you’ll likely understand our desire to confirm our profile variables (such as a participant’s employment status, marital status, income and location) as often as possible, since these variables are subject to change as people progress through life. Even if a mere 2% of our panel has changes on each of these profile variables from year-to-year, that would have a substantial invalidating impact on our automated panelist selection process. We know from internal maintenance and process improvement that a 2% annual change in any of these profile variables is highly conservative.
As researchers, we’ve found the path of least annoyance works best when confirming our participants’ relevant profile variables in screeners and longer surveys. Our strategy it involves two steps: timing determination and passive confirmation.
Timing determination. This basically boils down to: “How often do we really need to confirm this profile variable?” Depending on the profile variable being considered the answer to that question should be somewhere between “every survey” to “once a year.” Take gender and birth date, for example. These are very unlikely to change, so confirming them only once each year is appropriate. Now, consider location and marital status. People move, get engaged, get married and get divorced with relative frequency in our society (and our panelists’ updates reflect this reality), so we’ve chosen to confirm these variables every three to four months, instead of annually.
In other words, when sending out a screener/survey where respondent marital status is particularly important, and nothing else we’ve fielded in the last six months has needed to confirm this attribute, we make it a priority to confirm marital status by making it one of the first variables in our survey, then tying those answers immediately back into our participants’ profiles. This allows us to terminate respondents as necessary while—and this is important—at the same time letting them know that their responses have been tied back to their profiles in order to help improve their survey offerings in the future.
Passive confirmation. How many times have you needed to fill in your birthdate on an online form? A couple hundred? A thousand? If you imagine yourself as an active survey panelist, it’s much more likely the latter than the former. Each one of those instances involves filling in a date (either by typing or selection from a drop-down menu), selecting a month (another drop-down menu), and selecting a year (most likely a third drop-down menu). All-in-all, it’s maybe a 10-second task. Not a significant amount of time, until you multiply it by a few thousand instances—at which point you realize you’ve spent hours of your life performing what should be one of the simplest of online tasks.
When we put ourselves in our panelists’ shoes, we saw how much it might mean to them to have some of that time back. Now, they’re happier and more responsive as a result.
One of the simplest ways we avoid asking participants to supply their birthday each time they take a survey is by piping this information directly from their profile into the survey instrument and then simply asking them to confirm it. What was once a 10-second task for the panelists becomes a 1-second task with the addition of little extra programming on our end. Below is an example of the basic passive birthdate confirmation item.
- According to your Panel Profile, you were born on January 1st, 1976. Is this correct?
For any passive confirmation, answering “Yes” confirms that the panel information is correct and can be used, and answering “No” indicates that the panel information needs to be updated, at which point the relevant panel profile question is asked. Is this a little extra work for our team? Sure, it is. But is also lets our panelists know that we’re aware of (and using) their information on a regular basis, and it allows us to tailor our termination screens in a way that provides more information to our panelists as to why they might not qualify for specific studies, which makes them happier. And a happy panel is an active panel.
For more information about market research techniques, download our Free eBook, How Much Does Market Research Cost? Or feel free to contact us directly (319) 364-7278 and we’d be happy to help.