This is the fourth in a series of articles on the topic of Respondent Engagement and Respondent Survey Satisfaction.
Over the years there have been myriad papers published on the topic of incentives as a lever to increase respondent engagement. Recently the GRIT CPR (Consumer Participation in Research) Study (published in March of 2017), showed findings that confirm previous studies: that incentives are highly motivating to approximately one-third of the survey participant population. These findings reveal, however, that almost 70% of participants are motivated by “some other aspect of the survey process.” In this ongoing series of blogs, we are discussing those intrinsic characteristics of surveys and survey invitation promises that drive the majority of potential survey takers.
But for this installment, we want to review the role of cash incentives (extrinsic to the survey experience) and the nature of the motivation provided by differing levels of monetary reward. While cash as a motivator is well documented, less is known about the function relating “amount of cash” to “degree of motivation.” Many studies have concluded that “more is better,” but is this always true? The answer turns out to be “not really.”
We measured the reactions to survey invitation content within a closed community, in which we examined the relationship between the amount of cash offered and the general levels of those stating the rate was “fair and acceptable.” In particular, we wanted to understand the marginal increase in those responding positively to a survey invitation based on the level of incentive offered. For the purposes of this analysis, we attempted to hold constant other possible deterrents to response by focusing on survey descriptions to similar audiences, of similar duration (12 to 15 minutes), difficulty, and affinity with the sponsor. (All 14 test survey descriptions were panel branded, not client branded and had an average n = 473 rating.) All the respondents had experienced the survey environment in the past.
What we found is that the relationship between level of incentive and marginal acceptability of the offer was discontinuous. Starting with the difference between nothing ($0 of incentive) and $1.00, the marginal acceptability rate showed a significant improvement. We interpret this as not solely attributable to the modest cash emolument, but rather the sign that “some” recognition of contribution, no matter how small, was a very important motivator.
So if $1 is great, $2 should be much more exciting, right? In reality, we saw statistically the same improvement in the marginal response rate when we doubled our paltry amount. So how about tripling the incentive offered to $3? Here’s the function we observed:
Chart 1: Incentive Amount versus Marginal Acceptability
The function we see appears to have naturally occurring jump points. $1, $2, $5, and $10 rates all show upward ticks in the marginal increase in the acceptability rate, but everything in between does not have the same level of effect. This means that if you’re thinking of trying to increase response by offering $6, versus $5—save your money.
Another aspect to be aware of is that at the higher rates, the acceptability becomes less and less significant. So the economics of offering large cash incentives appears to offer rapidly diminishing returns. In the following chart, we see that the marginal success in communicating a “fair and acceptable” incentive flattens out after $10 (at which 85% said the incentive was acceptable), with very little to be gained from adding more cash to the offer. (Note: these studies did NOT include high-value respondents like doctors or lawyers.)
Chart 2: Cumulative Acceptability versus Incentive Rate
Also note, that when we asked people what would be a fair and acceptable incentive for the 12 to 15 minute survey, 38% said that receiving no incentive would be acceptable. This leads us to believe that if a third are highly motivated by cash incentives—MORE than a third are motivated by other, non-monetary factors. We assume that the remainder are motivated by a combination of things that are probably unique to the individual.
So while some cash may be important, we conclude that it may be more highly related to recognition and confirmation of the appreciation of the sponsor, rather than a mere mercenary motivation. Certainly, cash does not produce a uniform positive relationship with variables related to participation results.
As we continue to lay out the arguments for intrinsic values of the activity to drive survey participation, we’ll agree that some level of extrinsic incentive is needed for engagement…just not as much as we might have thought.
Let me know if you have observed something similar!
Bill MacElroy is Chairman of Socratic Technologies, Inc.