A View on the Future of Asking Questions for Market Research

 

I wish you a Healthy & very Happy 2018, dear readers; this is the second of a series of four blog posts as described in the first piece, posted in 2017.

Let me say from the beginning that in this blog post I will be dealing with quantitative and also qualitative research i.e. asking questions in both surveys (mainly closed ended) and focus groups (mainly open ended).

1. Surveys

I don’t think there is much dispute when it comes to the “rumour” that face to face and telephone surveys are on their way out. There are only a few pockets left on the planet where internet penetration is low, thus telephone or face to face questionnaires still make up most of market research surveys, but only for a couple more years. The way smartphone penetration has grown exponentially in the last few years makes fixed internet line connections in the home obsolete. Soon, the only way surveys will be conducted will be online.

The next decade will of course bring many more changes to surveys, as there are a few inherent problems that technological progress allows us to fix, such as:

  • Reliance on respondents’ memory – bad idea
  • Sampling methods, especially respondents recruited from access panels – really bad idea

Market research practitioners who still run tracking surveys often say in their defence – recognising some of the weaknesses of this approach - that this is the best method currently available to gauge customer perceptions. This is the “better than nothing” type of excuse. Well, to be honest I am not so sure that “nothing” in this case isn’t better than bad/unrepresentative/misleading data. After all, what should be the price of bad data? It has been proven time and time again that human memory about not so earth-shattering things (for example why you chose the brand of soap you bought last time) sucks. We also know that many members of consumer panels only participate in surveys for the monetary incentive involved, so it is in their interest to complete them quickly and move on to do something more interesting, or even take another survey for the same reason.

Is there a solution to this conundrum? Well, as the title already suggests I do have a view about the future of “asking questions” for market research, so the answer to my rhetorical question is: “yes of course there is a solution”. We call it SIS = Short Intercept Surveys. You must have noticed that we quite like coining new terms and acronyms; I hope this too will stick like the last 15 or 20 that you all know and love (true or false ?).

The underlying science for SIS is quite simple really: if you ask someone about something they just did or have just said, then chances are that they will remember what it was and why they did it! This is now possible not only due to the shift from in-store to online shopping, but also because of smartphones, GPS, RFID beacons etc.

It also helps if you only ask them one or two questions… up to 5 if you must do more. Since you will not be paying them to sit around for 20-30 minutes being bombarded with all sorts of boring questions, they will no longer be motivated to hurry up, cheat, lie etc. In other words, it would not be farfetched to say that current survey methods bring out the worst in people!

2. Focus Groups

Moving on to focus groups, the idea is the same: online focus groups are catching on. There are so many more ways to engage online and ask questions. We have bulletin boards, chat groups, video & photo diaries, walls, video chats, in-depths and the list goes on. You can even mix and match data collection methods within the same project; start with a poll, then follow with two open ends in a bulletin board, then a task asking participants to share a photo and/or record a video clip, concluding with a chat group for the top 6 contributors. This allows insights experts to connect the dots and synthesize superior insights that would otherwise remain invisible. The benefits of online already outweigh the benefits of physical focus groups:

  • asynchronous engagement is now available and very powerful
  • faster – transcripts are available at a press of a button and they are way longer (qualifies for better)!
    Quiz of the day: why are online chat group transcripts longer than those of physical focus groups, for the exact same number of participants and duration? Tweet your answers to @DigitalMR_CEO.
  • providing opinions by also sharing rich media (also qualifies for better)
  • cheaper (some people say that clients who ask for all 3=better, faster, cheaper are unrealistic)
  • participants from anywhere in the country (or even in the world) in the same discussion
  • on demand or agile or continuous when used on online communities

An online community is like a wrapper around all the data collection methodologies described above, providing cohesion and most importantly genuine and credible respondents.

All this and we have not even said anything about how using artificial intelligence can add tremendous value to online communities. One way it is applicable is in analysing large amounts of text in any language, in an automated way, and in real time.

Beat that, “face to face research”!