Ask any panel company the right questions and they will reveal that the more survey invitations (and re-invitations) an active respondent receives, the less likely they are to complete any of them. This somewhat paradoxical finding is at the heart of the market research industry’s response rate problem. The process, in a nutshell: panels identify respondents willing to help them; those who do so are naturally sent more invitations; and are soon inundated from all sides; at which point many follow their instincts and run away; leaving an ever dwindling number under ever intensifying email bombardment. Thus we scare off our best customers, in progressive and almost cannibalistic fashion.
Yet still we read umpteen daily articles demanding shorter, simpler, surveys – which surely also implies more surveys.
Making the broad and perhaps bold assumption that the number of survey questions the market research industry asks is not, has not, and will not fall significantly then we obviously still need to ask those questions, somehow. Under the ‘shorter surveys’ model of thinking, fatigue and boredom are reduced and respondents (who apparently all have microscopic attention spans) become simultaneously happier and more engaged. That is not an analysis I agree with and in any case it isn’t going to happen. Firstly, clients resist being asked to cut-down in this ‘more for less’ business culture and, secondly, as researchers we often can’t truncate the survey and fulfill the objectives all at the same time – good design simply won’t allow it.
In which case, we can expect the called-for rise in the number of shorter surveys to correlate nicely with an increase in the number of survey invitations received by willing research participants, which in turn will correlate nicely with a reduction in the numbers of willing participants available to us – as they become utterly fed up with the volume (and quite possibly also the dumbness) of what they are being asked to complete.
Why not focus on writing better surveys and not worry too much about how long they take?
Yes, I agree that anything over 30 minutes is pushing our luck (but then again, respondents can and should be able to resume a longer online survey after a break at the point they left off). And, yes, we need to be up-front about how long a survey might take.
But my strongly held hypothesis is that participants care more about their experience than they do about watching the clock. As researchers we should push back first against a client’s tedious and boring design and draft something better. Forget about how long it might take, at least initially. In doing that we appeal to and showcase our core skills and have a better chance of engaging respondents, by virtue of the inherent quality of our work.