Erstwhile researchers, be fair.

I was recently in a meeting with the insights manager of a leading telecom player. During this meeting, I was trying to sell a recent offering in which among other things, there are 25 image parameters on which respondents associate celebrities. During the presentation, the insight manager said, ‘25 is not enough; I need 150 image traits’. After the presentation was over, she repeated her need for 150 traits complaining that 25 were not deep enough. I was speechless wondering how can an ex researcher talk of 150 image traits being in a questionnaire? I do not mind conceding that 25 may not be enough for some clients so one may ask adding a few more. But 150? Does she not know what an interviewer goes thru in the data collection process? Do we all not know how difficult is it to get an interview? After all, how many of us would give one to a market research company? And of those who said yes, how many would go through a list of 150 image traits, among other things in a questionnaire? Did she never, at one point in time in her career when she was on the other side of the table as a research seller, defend the merits of keeping questionnaires short? Coming from someone who spent a few years in a leading market research company, left me wondering if she was really doing justice to her new employer. The meeting may have triggered this blog, but I have met many ex-researchers like her in client organizations.

Research or insights managers in client organizations are hired with the assumption that they know the market research process well and are therefore in an advisory capacity protecting client interests. Their job is to understand the requirements from the brand teams, convert them to research briefs, liaison and negotiate with the market research agencies and deliver back what was required. They are also supposed to educate the brand teams on what makes good research, what type of questions make sense and what makes respondents irritable. However we continue to see poor demands of deliverables (power point being used as excel sheets with 200 slides) or 40 page questionnaires with 80 psychographic statements or 150 image traits by the same ex market researchers. Whereas most of them are doing a good job, I hope, they are doing little to educate the clients, now their employers, on the best practices of the market research process. In turn, clients continue to be oblivious of the information in spite of having them on the payroll.

For instance, a true researcher knows (but would never admit) that if the questionnaire length exceeds 15-20 minutes, the quality of response suffers. I still am surprised at how little the questionnaire length plays a role in negotiating contracts. They all talk about reducing sample size to save costs but I have yet to meet a researcher who asks me to reduce costs by reducing the length of the questionnaire. A high proportion of interviews stop midway leading to productivity loss but who cares? How many of these ex researchers visit the field once in a while to get a feel of data collection? And how often? If they did, they would know how difficult it is to get an interview, let alone administer a long questionnaire. Rather than only negotiating the price of the survey, they should question how much are the market research agencies paying their interviewers for each completed survey? Or for that matter what is the value of the gift given to the respondent?

If you are a client and if your market research agency or your insights manager says OK to a questionnaire which in your opinion takes 40 minutes to administer, then you are being fooled. Period. If there is no provision for gifts, then half of the data is diluted. Ask yourself, why would someone oblige otherwise? Would you?
So here are 3 rules, as a benchmark if you really want to improve the quality of your market research data:

1. Try really hard not to exceed 20 minutes of interview time. Approve a questionnaire that you will yourself answer if approached by an interviewer.

2. Except for opinion polls, always keep a budget for gifts to respondents. This is customary in rest of the world. The quality of data will more than justify the increased cost.

3. If you are breaking rule number 1, then compensate it by increasing the value of your gift, from rule number 2.

And ex-researchers, please take efforts to educate clients at the risk of being unpopular. That is, if you really want this profession to have a long life.