Driving Human Interaction with Design

Over the last couple of years I've made a transition from a graphic and an urban designer, into a product designer. And throughout the last year specifically, I've focused on design for Product Growth — a fast-paced, iterative process of product experimentation, testing various hypotheses aimed at moving the North Star metric. For product growth, it is defined as the rate limiting step of the startup's growth. In our case, we tried to get more users to invite others.

And so my role on the team has been to produce really quick feature designs, be that on the web or mobile app, so as to quickly check if our hypothesis can be confirmed. The nature of the product growth process is one that assumes that most of the tests will fail. However, in the process, the team should be able adjust its course by noting down the learnings from failed experiments. That way, this process should eventually lead to some big wins. And we had a few particularly prominent ones.

One experiment that stood out to me has been the optimisation of the app's main sharing funnel. I have long thought that our invite funnel was too lengthy and included unnecessary choices. And as a result, I had an idea that we could make it shorter and alter the choice architecture for more users in this funnel to choose to invite friends. So here's our funnel before:

And this is our test invite funnel, after:

Thhe text in the second screen would be typed up by a console. I wasn't a big fun of the solution but the team insisted and it was OK for the sake of a test.

The result was pretty solid: we got a 25% relative increase in the share of users who end up inviting friends following this funnel. And this got me like REALLY impressed. I made fairly minor visual changes to screens, but altered the architecture of the flow — essentially, the updated design made more users interact with the app in a way that we wanted them to. As a result of their interaction, the key metric was moved, by which we measured our success.

And I started extrapolating this idea further. Aren't books designed to be read? Aren't chairs designed to be sat on? And, ultimately, isnt' Instagram designed for users to be infinitely scrolling through their feed? Of course, one could break down these interactions even further, or indeed add more layers to what design should do, but I would still claim: design is about and should focus on fostering an interaction, with those who are considered to be the audience of this piece of design.