
Decoding Decisions: Making sense of the messy middle is a collaboration between Google and The Behavioural Architects to understand the nature of consumer decision making on the internet today.
We are proud to say that this work has gone on to win TWO prestigious awards: the MRS New Consumer Insights Award AND the MRG award for best research initiative!
The way people make decisions is messy and it’s only getting messier. Our challenge was to understand how consumers make decisions in an online environment of limitless information and abundant choice.
We conducted literature reviews, observed hundreds of hours of shopping journeys and ran large scale quantitative shopping simulations across 31 different categories, including CPG, Travel, Leisure and Automotive.
Through this research, we created an updated model of consumer decision making - the Decoding Decisions model. This reflects the reality of decision-making behaviour in the internet era, including:
- An on-going state prior to the purchase trigger called EXPOSURE, whereby consumers are influenced by prior experience of a brand - from a previous purchase experience, brand communications, word of mouth and generic product and category associations.
- The non-linear and evolving nature of decision making and how consumers cycle through multiple loops of EXPLORATION and EVALUATION as ideas, information and brands are sought and assessed. By observing people’s behaviour on and offline as they shopped a given category, we found that consumers loop between two mental modes: exploration, an expansive activity; and evaluation, a reductive activity. This is the ‘messy middle’ in the centre of the model. It illustrates how people can loop back and forth and can repeat the cycle many times before making a purchase.
- The role of heuristics (short cuts) and cognitive biases within decision making, and how powerfully these can influence brand choice in the ‘messy middle’. We found that people cope with complexity by using cognitive biases buried deep in their psychology. These biases existed long before the internet and our purchase simulations show how relevant they are to decision-making today.
You can read the study here, but please contact us if you'd like to learn more about the work.