For our 25th (25th!) Google Firestarters we took the theme of 'patterns of behaviour'. We'd not done a Firestarters on behavioural economics/behavioural science before but it's a popular subject (check out the brilliant Nudgestock – next year I must get along to that). But as well as focus on the social psychology of advertising I also wanted to take a broader perspective and consider ways of understanding and modelling behaviour. And so for this one, rather than a series of short provocations we had two speakers giving a more in depth view of both aspects. And we ended up with two excellent, dense, and very rich talks which will take some time for me to digest.
First up Richard Shotton from MGOMD talked about some of the lesser known and under-appreciated biases relevant to advertising and marketing. He began by noting that the scope and variety of advertising (from lost cat posters to government behaviour change campaigns) means that we need a range of ways of understanding how it works, but the skill comes in being able to assess which model to use for which scenario. There's been quite a bit of focus on relatively well-known behavioural concepts such as social norms theory and survivorship bias, but planners need to draw from a wider set of tools including some lesser known insights from social psychology.
Richard talked about how some of the simpler creative techniques in advertising (notably rhyming) can be powerful in driving recall and association, yet most of the examples that we know ('Beans means Heinz', 'you only get an "oo" with Typhoo', 'we all adore a Kia-Ora') are thirty years old. In fact he showed that there has been a distinct decline in the use of simple mechanics like these (he used another example based on a study done at News Corp of the Sun newspaper to show that the same had happened with headlines in that paper). The point being that we forget some of the elementary lessons of our industry at our peril.
He then went on to consider the 'pratfall effect', the tendency of people to perceive of something as more attractive or human if it is not quite perfect. Sometimes called the 'blemishing effect', he talked about how imperfect reviews can lead to more purchases, and how some of the most famous campaigns of all time (VW, Avis, Marmite, Guiness) had played up and made an asset out of their imperfections.
Consumers cynicism with advertising means that this technique enables to brand to better focus where the flaws in brands are expected to be. The pratfall effect can be a way of better aligning the interests and positioning of a brand with the expectations of customers.
Richard finished by focusing on 'wishful seeing' (biased interpretation of research or data) with a brilliant dissection of the whole idea of brand purpose, taking apart popular research such as that which formed the basis of Jim Stengel's book 'Grow – How Ideals Power Growth and Profit at the World's Greatest Companies'. He adeptly showed how problematic the conclusions were that were derived from that study and showed that actually a more balanced, thorough investigation of the same brands/businesses shows that brands with an supposed 'purpose' do not outperform those that apparently have none. Too often, he says, we look for the data to prove the conclusions we want to believe in rather than what the data is actually telling us. He left us with the words of Bill Bernbach, making the point that it is as important to focus on the things that don't change as it is the things that do:
It was a fascinating dive into the social psychology of advertising.
Next up, John Willshire gave a tour de force of his thinking around models and frameworks for understanding culture and behaviour, and how to navigate through the 'pattern problem':
John spoke at one of our first ever Firestarters on the theme of Design Thinking. Linear processes (like most design processes), he says, are built for comfort. Most realities are far more messy. He talked about John Boyd's OODA loops as a way of showing how strategy and orientation happens in loops rather than linear lines. We orient ourselves by using new information, analysis and synthesis, previous experience, and cultural traditions (present and patterns). But these change all the time, and the very act of going through the process means that when you go through the loop again things have changed.
So how do we break the cycle of endlessly repeating the same process in what could be very different contexts? Charlie Munger once said that "eighty or ninety important models will carry about 90% of the freight in making you a worldly-wise person". There are a limited number of models that we can apply and reapply in different ways and versions – loops help us remake our models in response to reality. He then talked about Stewart Brand's pace-layering, ('if loops are behaviour, layers are the environment') and the idea of gearing (Stewart Brand said: 'The quick processes provide originality and challenge, while the slow provide continuity and restraint'), and his people/space matrix and what happens in the extremes of the combination of these two things. His dissection of the Pokemon Go phenomenon as an example of layers was fascinating.
So how do we design tools that break patterns (and break the pattern problem)? John talked through two examples of tools that he has created for Smithery and uses with clients - ‘The Obliquiscope’ and ‘Zenko Mapping’. Both of which are designed to grow and change as you use them, so that effectively they’re never the same tool twice. The former draws on John Kay's thinking in Obliquity as a way of interrogating something (like a problem) that you're looking at. The latter utilises the space/people grid to map fast/slow. Again, it was a densely rich talk that carried with it some very useful concepts and ways of understanding behaviours.
My thanks to Richard and John for two such thought-provoking talks, as always to Google for hosting, to Scriberia for such a fantastic job of visualising the event, and to those that came along and were part of it. It was a lot of fun, and fascinating with it. Look out for the next Google Firestarters in November.
Leave a Reply