Trails of experience

The weekend before last my two friends and I were resting in a warm and cosy lounge reflecting on the previous four hours hiking over the South Downs National Park. As we were feasting on homemade soup, bread, cheese and obscenely strong hot chocolate, our conversation was rudely interrupted by the hiking app announcing we were off course. “Well yeah… thanks Columbus!”

It reminded me of two things; the importance of wholistic contextual understanding and how digital experiences feel jarring when there is an absence of the former.

Jeremy’s recently published post Research poked at this. In it he compared using quantitative and qualitative research methods for understanding the what and why of a situation. Jeremy argues that focusing on quantitative methods alone will limit your learning.

…it’s going to be tricky to figure out the reason just by looking at the wound.

The quantitative data points in our hiking app example are the GPS trail lines overlaying the map; a green line representing our intended route, a blue line representing the actual path we took.

Looking at the green line provides an objective view of what we did. Granted, there might have been a few inaccuracies, perhaps caused by a laggy GPS refresh rate or the device signal dropping momentarily. But on the whole, you can’t argue with the data, that’s what we did.

But what the green line can’t fully explain is the why behind its shape. It can’t fully explain why we deviated from the path. It can’t fully explain why we slowed down, sped up or why we stopped. More importantly it can’t explain our experience. We can infer why but without inviting the three of us to explain our experience we’ll never get to a rich and nuanced understanding of why we did it.

This echos with the challenge our team is facing at Waitrose. Designing and prototyping personalised experiences with data is like working with an entirely new design material. What we do with this data underpins every design concept we put in front of people. If the data isn’t designed in a way that meets people’s needs… forget about it!

Designing with historic behaviour reflects just that, past behaviour. We should consider that what people did in the past might not reflect what they intend to in the future, nor reflect their decision making process.

To quote Tom Scott “What people decide, isn’t how they decide”.

It’s no wonder businesses lean into quantitative research methods. They provide safe, binary outcomes and numerical metrics that make go/no-go decisions far easier. But let’s face it, the world and people are far more complex than yes/no.

If we want to be truly human-centred we need to immerse ourselves in the world of the people we are trying to support. That means becoming receptive to the complexity of their contexts, their differing mental models and developing an equal understanding of what they did and why they did it.