Friction as a lever

I mentioned yesterday how by using friction we can influence how users behave and how the business is organised. That’s because friction is a great lever

It turns out friction is a great lever you can use when building data platforms to encourage the behaviour we want to see, and discourage behaviour we don’t want to see.

For example, let’s say we want to reduce the amount of breaking schema changes that happen upstream, because of the impact that is having on revenue generating data applications downstream.

We can add friction through the use of CI checks, implement policies for migration plans, and so on.

Adding friction signals that this is something we want to see less often and with more care.

Data producers then try to avoid this friction by spending more time upfront designing their schemas, preventing breaking changes when possible, and engaging with their consumers to understand the impact of changes.

We’ve changed their behaviour.

We can also reduce friction to drive behaviour change. For example, allowing data producers to create and manage their data with autonomy, removing central reviews and bottlenecks, signals that this is something we want to see more often.

Friction is a powerful lever you can use when building your data platform to influence your users and your organisation.