Customer Development and Research

“Come on-a my house, my house, I’m-a gonna give you candy” – Should you incentivize respondents?

The question of incentives has always gave some headache to researchers and companies alike when it comes to collecting customer feedbacks.

I considered writing a few thoughts about this topic after stumbling upon a former article on the Customer Development Labs blog (that I have already mentioned before, unfortunately quite inactive for some time now). In this article, they were proudly presenting a case study about a startup (LikeBright) conducting 100 customer interviews within just a few hours using Mechanical Turk.

To cut the story short, they practically created a “HIT” on the site, offering target group customers (moms sitting at home, actually) about a buck for talking to them for around 5 minutes.

You just start wondering when you hear stuff like that. On one hand, this seems like a clever way to ease the pain of recruitment and motivate people to talk with you about their problems.

On the other hand, there are several doubts: Is it realistic? Are people really giving sincere responses when they feel that they are paid for that? Or does it turn into a “job” that they just wanna get done and don’t try to think about it too much?

Lean gurus like Ash Maurya (author of the well-known Running Lean book, source of the quote below) usually don’t really suggest payment/incentives for your prospects.

“Don’t pay prospects or provide other incentives. Unlike usability testing where it is acceptable to provide incentives for participation, your goal here is to find customers that will pay you, not the other way around.”

I generally believe that this rule should be followed, yet my Market Research experiences made me see things in a more complex way. Note: incentives are actually quite common in Consumer Research, yet “professional respondents” (taking part in every possible project just to harvest the rewards) are persona non grata for their assumed lack of credibility. 

So, to sum up pros and cons, my warnings and advices are the following:

– As a rule of thumb, beware of “comfy” ready-made solutions like this. You might feel that you “ticked the box” faster, but it’s not just about numbers (“I have to get 20 people, I have to get 20 people”), it’s also about the quality of respondents and learning where to find them in a “natural” way (either online or offline – where is their habitat, where do they form a community etc.)

– Note: even if you are so lucky that your target segments are INDEED present on sites like mTurk (like LikeBright assumed), it’s still wise to only use such platforms as an additional source. You should still try to actually go out and try to find people, observe how they act, react, behave and look them in the eye (if you can) – this cannot be substituted by assembly line like recruitment methods.

– Some sort of incentive might be okay for methods where numbers ARE actually important (see surveys, another horcrux in some gurus’ view but still useful for specific objectives…) AND your target segment is so wide and compound (different sub-groups with different attitudes) that it could need some boost to get people hitting on that web invite. (And yes, I am no saint, I did that before as well and I do believe that if you are looking for quant validation of hypotheses and want to ask masses, masses of people, this can make your job a bit easier.)

– BUT: this mainly works if you are trying to explore PROBLEM areas AND there is enough “emotional distance” between the respondent and the researcher (so, mainly quant and online methods).

– BECAUSE otherwise, some harmful mechanisms may erode the credibility of the findings. For instance, if you pay people, they might start to feel that they have to be nice to you. They might feel the pressure to answer the way they expect you expect them to do so (like at a workplace or like being nice to a neighbour when you want them to keep an eye on your house while you are away).

– IN ANY CASE, try not to incentivize every single respondent with money (aka, don’t pay) as this might even scare off those who might just really be genuinely interested in the problem area (and make them feel like materialistic answering-robots).

– INSTEAD OF THAT, try to come up with something creative, maybe something that is related to your service (this may actually motivate people who really feel that the problem/solution might be relevant for them). This can be directly connected to your product or just regarding the broader “lifestyle” (e.g. travelling-related if you are doing a travel startup, experiences, ebooks etc.). Or maybe you can offer a prize, but I only advise that if you are targeting “ordinary” (B2C) segments and other forms of immaterial incentives (pride, expertise, interest, “social stroking” as anthropologists say) won’t take you far.

– Further note: definitely don’t try to do that during your MVP-test or Solution Interview phase as the chance of “empty nods” will increase (“Yes, yes, that’s purrrfect like that. Where is my fiver?”) UI-testing and looking for bugs is of course an exception, owing to the character of the examination (also according to Lean gurus like Maurya).

– Final advice: just as always, think about your target segment and what makes them tick. How could you catch their attention and get them motivated to take time for you? Maybe some sweet talk will be enough, otherwise, you can still get creative and offer them some candies, just don’t give them the money to buy some candies from another candy man.

One thought on ““Come on-a my house, my house, I’m-a gonna give you candy” – Should you incentivize respondents?

  1. Pingback: How To Establish Better Interactions By Learning From "Market Research Online Community" practices | TheCollaborativeStartup

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s