What do opportunities for the market research industry have to do with decision making or cakes? As part of the Festival of New MR I’ve written a blog about just that, which you can find here.
Last week Tameside Council ran this ad in our local paper – a nice example of using social norms (‘most people do this’) to encourage behaviour change.
Will it work? Well, it’s similar to the tactics used by the government to increase tax returns in 2011, which reportedly did. They sent reminders saying most people in the area had already completed their assessment, which apparently resulted in a 15% increase in completed forms. So maybe it will.
So the Behavioural Economics course has finished (and I got my certificate, wooo), and it’s about time I reflected on it to see what I think market researchers can learn. Maybe there will be a follow up blog post, as there are loads more things I think we can take from the discipline, these are just a few thoughts.
We rely on contextual information to make all sorts of judgements and decisions
In-situ research (facilitated to a large extent by smartphones), combined with observation, self-reportage and reflection is probably going to be key to getting closer to people’s decisions as they make them. That’s because understanding the context in which a decision is being made is much more important than traditional research has given it credit for, but also much more complex to research overall (what parts of the context do you look at – which parts might be relevant? How do you measure the contextual stimulus, defaults, or relative choice sets? If you do measure them, how do you aggregate and analyse them?).
We need to re-think the way we approach research design to help get us nearer to the ‘truth’, and figure out new ways of analysing the information we collect to find insights. It’s worth remembering though is that there are some things we’ll never be able to really predict about people’s behaviour, but we can certainly get cannier about how we try to understand as much as we can.
Experiment with experiments
One thing I’ve reflected on since the course is that much research is about asking people to predict their own behaviour, while much of the data cited in the course is based on experimention and testing hypotheses in ‘what if?’ scenarios, to see what actually happens. One of the criticisms I’ve heard people make about the course is that it wasn’t very commercial. So why don’t we test some of our hypotheses in a commercial environment?
Do we experiment enough in the real world to see, for example, whether people do actually notice the new packaging of our product on the shelf in the hustle and bustle of an actual supermarket with kids in tow, versus completing a survey with a pack stand-out test when they’re having a cup of tea at home? Which way of testing is likely to be closer to ‘the truth’?
(Okay, so both methods might still be survey-based – though I didn’t say they have to be – but something like eye-tracking kit could be a useful additional data source in store, and in store you’re at least getting people in the right emotional ‘state’.)
Making customers work can be a good thing (in moderation)
The IKEA effect theory suggests that if people put effort into things they own, they like and value them more. It’s like IKEA furniture, because you’ve had to put effort into creating it, so you appreciate it more than you would otherwise (apparently; I don’t speak from personal experience, DIY is not my thing). There is a reasonable tolerance here though; too much effort – and making customers work too hard – will count against you, so it’s about striking the right balance.
With that in mind, allowing a degree of personalisation to your products, requiring just the right amount of effort from the buyer to feel they have a personal stake in it could mean when people buy your wares, they become more attached to them than if they’d just bought an off-the-shelf version of your product. Linked to the idea of personal attachment to things is loss aversion (the posit that we over-inflate the impact of potential losses more than we estimate gains); if convincing consumers to switch to whatever you’re selling might make them feel they’re losing something by switching (even if they’re gaining more), you’re going to have to sell the benefits of switching really well to make it seem appealing.
Money is only one aspect of motivation
Okay, this isn’t new news if you’re a decent manager, and it isn’t even specific to research, but I found it interesting to think about how important meaning is to people in what they do and how this translates into a work context – and of course it makes sense when you think about it.
Making sure people’s efforts at work are recognised, even if that report they’ve spent ages working on doesn’t get presented to the client, or even if the project is cancelled and the brilliant workshop they’d planned never happened, recognising the effort and work put in by people can make all the difference to how people view a situation – as could re-deploying some of the original work so it’s used for something else. Certainly something I’ll be bearing in mind in future.
Google does a great job of this with its Labs project where employees are allowed to spend some of their working week devoted to their own creations. Apparently many of Google’s products came from this project, and it means people get to work on stuff they enjoy, stuff which has meaning for them. An interesting way of engaging people in a business and something I recognise in my own role, where I’m able to explore new ideas I find interesting. And something I need to bear in mind as a manager – that I need to make sure the people I work with do things which have meaning for them. The BIG conference I attended this week suggested meaning is different for everyone, so this isn’t necessarily going to be an easy nut to crack, but if you can get it right at least some of the time, it’s definitely worth trying.
I’m just coming to the end of a six week online course in Behavioural Economics (BE), lead by Dan Ariely (author of Predictably Irrational, amongst other things) which I’ve been taking through Duke University and Coursera. It’s something I’ve done in my own time although I’ve really been keen to complete it because of my new role at work, where I’m heading up a newly formed ‘Innovations’ team. One of the team’s key remits this year is to develop the business’s understanding of consumer decision making, so the course, run by an established Professor in the BE area, and on a topic increasingly of interest to marketers and market researchers seemed a good fit. It was also free, which definitely helps! (Our irrational attitude to ‘free’ stuff is actually covered on the course.)
It’s been an interesting experience, not least in my own procrastination (something else the course covers), although I have mainly kept up to date with the coverage (just). It’s covered the psychology of money, motivation, emotion, self-control and has provided lots of examples of how irrational we (i.e. people) really are, and how policy makers, marketers and the like can influence people by nudging, guiding and coercing people into making certain decisions or choices, or by understanding how the choices we make aren’t always as rational as we might think. For example, why are organ donation rates so high in countries where opt-in is the default? Why do credit cards reduce our rationality when it comes to spending? Why do we respond better to charity adverts where we see a single, named child, than we do when we see the scale of a horrific situation on scores of families?
Apparently tens of thousands signed up for the course, significantly fewer (although still thousands) completed the written assignment, and it will be interesting to see what proportion of those who started actually complete the whole course (or gain a certificate). As an experience, I’d definitely recommend doing an online course like this if you’re considering it – it’s been fantastic that such a ‘social experiment’ in education is offered for free, as the quality has been impressive. Because there’s no monetary cost, you can dip in and out as much as you like, and just watch the lectures if that’s all you fancy doing. I suspect there’s a big future in online courses like this, which are accessible to anyone with internet access and a device to access it on. I’ll certainly be looking around for others. [I’m now waiting for a design/photography course to come up!] This one has no definite plans to run again, but it’s a possibility so it’s worth keeping an eye on.
To gain a signed (by Dan Ariely) certificate on completion of the course you need an overall grade of 85%, achieved through a combination of weekly quizzes (which have hard deadlines, after which you don’t get a credit for them), peer- (and self-) assessed coursework and a final exam. I’m pretty much up to date with the quizzes, and I’ve done the coursework, just the exam to do next week. So is it about getting the certificate? It wasn’t, but it’s starting to be (my own irrationality about loss aversion and not wanting to ‘lose’ the certificate I haven’t even gained yet aren’t lost on me), although it will all come down to the final exam now. I’m not sure I’ve done enough wider reading to reach the overall 85% but either way, I’ve gained a fair number of stories and theories about decision making and a much greater awareness of how irrational people actually are, which was the whole point when I started it, and is already proving useful to me at work.
Here’s my coursework if you fancy a nosey. Thought I might as well make it about a real-life problem at work, although I’m not sure it’ll be implemented…
At work, we have a timesheet system which all employees are expected to complete each week to account for the different projects they spend their time working on e.g. Monday 2 hours working on Project A, 5.5 hours on Project B, Tuesday 4 hours on Project A, etc. Timesheets are recorded in a spreadsheet on the individual’s computer and then emailed to the Finance team at the end of each Friday. It takes around 15-30 minutes for individuals to complete their timesheet each week; it’s a fairly minor task. Timesheets are used by the Finance department to produce financial reports each week. If individuals submit their timesheets late, the Finance department are delayed as they cannot produce complete reports.
Many staff routinely submit their timesheets late, causing problems for the Finance team. Timesheets are considered a low priority admin task by many staff, which means despite Finance sending regular emails and making phone calls to remind late individuals to submit their timesheets, they still consistently come through late.
The existing research
People make different choices when they are ‘nudged’ towards something (video lecture 1.2) and when there is a default option, so to get people to change their behaviour, we might want to try setting a default with the desired behaviour. Johnson & Goldstein (2003) and Dan (video lecture 1.2) show that in organ donation programmes where opting in is the default, there is a significant increase in organ donation, as people stick with the default, the path of least resistance. The default pushes or nudges us towards a particular choice.
To consider another theory which could apply to the problem of timesheet completion, Dan discusses the pain of paying (in Predictably Irrational p248 and video lecture 2.3) and suggests that it can be increased if the saliency of paying – i.e. the amount of attention given to paying – is increased, and can also be influenced by the timing of paying. He suggests that if people receive notification whenever money is spent, the pain of paying is increased. He gives an example (in the same lecture) of people eating pizza and how they would enjoy their meal less if the waiter counted how many bites they had (and how much it was costing them) than they would if they were just charged a fee at the end of the meal.
My proposed solution
I propose two potential routes for solving the timesheet problem.
The first is to think about defaults. Since people make different choices when they are ‘nudged’ towards something, I hypothesise that if the timesheet automatically opened on people’s computers each morning (i.e. became the default) it would nudge them into completing their timesheet because it is easier for them to do so (i.e. they don’t have to remember or think about opening it up, and they take the path of least resistance and complete it while it is already open).
The second potential route is based on increasing the pain of paying to increase timesheet completion levels. This could be thought of in terms of money e.g. dock someone’s wages by a certain amount for every day it is late, but this seems very extreme and would likely cause a lot of bad feeling in the office if people’s wages were docked! Instead of money, we could impose a variation of the pain of paying, for example on a Friday afternoon if all timesheets are completed on time, a fridge full of drinks (e.g. beer) is opened and everyone benefits from a free drink. I hypothesise that any late submitters will feel bad about being the person preventing everyone else from benefitting from a free drink, so the pain of paying comes from having a negative impact on the group. This could be even more effective if the system (e.g. an electronic sign on the fridge) displayed the names of anyone who hadn’t submitted their timesheet, to ‘name and shame’ them into completing it for the good of all of their colleagues. It may also invoke loss aversion (and be an even more effective strategy) if people worry that they will lose more than they gain by not completing their timesheet e.g. they might lose friends at work who are angry at not getting their free drink, because a relatively minor admin task hasn’t been completed on time. [This strategy probably won’t work with people who don’t like their colleagues!]
I expect that the pain of paying route would be the more effective of the two I have outlined above, but would have to test both scenarios to establish evidence to prove (or disprove!) my hypothesis.