There’s been a minor backlash against focus groups in recent days.  The criticisms are specific – on what focus groups say about the state of the government right this second – but I thought it an important moment to talk about focus groups more generally, and why policy makers need to pay attention to them.

I don’t run focus groups (though my company is well known for them). I have too much respect for the people who are brilliant at it to do a cack-handed ersatz version. I have no poker face, and I get emotionally invested in policy areas and solutions. I don’t have this strange ability to emotionally connect and extract the most from people, while also being a blank slate.

Of course, you could say that’s the downside of them – the emotional connection. Nothing like clean data to remove your priors (which is why you need polls too.)

But here are my top half dozen reasons why you should run focus groups alongside polls (and why you need to make sure the people running your focus groups really know what they’re doing):

• People in focus groups use words that people in politics don’t. Most of the best political slogans and messages I can think of emerged from focus groups (“Take back control.” “Get Brexit done.”) You know what you don’t hear from focus groups? Unleashing entrepreneurialism/aspiration/the market.
• It is easier to pick up from focus groups, by reading all the thousands of cues that are non-verbal and which form a huge proportion of human communication, how much people care, how comfortable they are with the question and area, how much they’re affected by social dynamics…all the things we subconsciously pick up all the time in our own interactions and affect how people behave.
• They enable two-way conversations on complex policy issues that aren’t possible in polls. Clearly, you must be very careful not to artificially inform participants beyond what could reasonably be expected in ordinary life, but people do become more informed on issues all the time – and it’s therefore useful to bring them up to a level that they might become in, say, a couple of years and to find out their views. An obvious example is new green technology for the home, which we know is coming very soon.
• They help overcome the ignorance and biases of the researchers. It’s widely assumed that unintentional bias is the bane of focus group research – that researchers lead participants and “hear what they want to hear”. Not if done properly. Focus groups can be better than polls in mitigating this because they give the opportunity to participants to say whatever they want, while polls only give participants the opportunity to say what pollsters allow them to say. We have lost count the number of times that focus groups picked up emerging trends – trends that effectively came out of nowhere – that could never have been thrown up by the polls because pollsters didn’t know to put the options in.
• If the people who are running focus groups are really experienced, they are brilliant at understanding whether things are really changing, why, and what they’re likely to change to. Clearly, this doesn’t mean that you can’t have new entrants into market, or indeed to train new moderators, that would be absurd; nonetheless, it helps when you have people in your research agency who have been doing qual for a long time because they are able to notice small changes which might become big changes, or indeed to put apparently big changes into perspective.
• Focus groups are fantastic at getting customers (political or non) to implement campaigns with fidelity. Because we are all human beings, most of us remember and react to other humans’ stories more, and for longer, than data points. This is true when we’re at school (the cognitive psychologist Dan Willingham has written a great book talking about the importance of storytelling in teaching) and it’s true as adults. When people sit in on our focus groups, it changes how they think about their campaign, what they’re doing, and how they should approach it. Polls get forgotten more quickly in my experience.

None of this is to dismiss polls, which are essential and do lots of things that focus groups can’t. Most good research projects will use both – it shouldn’t be a binary choice. Qual (e.g. focus groups) lets you design better polling questions, and dig into unexpected quantitative (poll) findings.

Nor are focus groups the only form of qualitative research – we experiment with different methods a lot, all of which have pros and cons: some are better at getting to groups of people that neither polls, nor traditional focus groups find; others are better if you want a form of public decision making, rather than public opinion.

This is also not designed to dismiss the idea that, in politics, you should decide what is important yourself and figure out how to persuade people of it (though you also need to know what problems people have).

But I think they’re vital for anyone engaged in what governments do, and how to make policy succeed.