Go back

On notifications, and asking more questions

Bastiaan Terhorst, 29/1 '20

Do you remember the last time you installed an app on your phone and it didn’t ask for permission to send you notifications?

Difficult, right?

Install a recipe app. It wants to send you notifications. Install a map. Notifications. Install a meditation app. You guessed it. Install a farting app. Yup. And before you know it, your phone is blasting notifications at you non-stop, which is unproductive and unhealthy.

Why is this?

The simple answer would be that engagement has become the way to measure success. It has replaced the vanity metrics of the 90s – like Signups, or Number of Installs. Because just counting Signups tells you nothing about if these users are actually valuable. So instead we measure things like Lness – the number of days a in a given time frame a user is active. For example, L5+/7 means a user is active 5 days out of 7. Pretty good! But can we get them to 7 out of 7? Perhaps they just need a little nudge on those inactive days?

But if I take a step back I believe this effect, and not cause. The real cause is that most people developing software products take little time to consider the consequences of the choices they make in developing their products. And while I used notifications as an example above, this extends to any and all aspects of a product.

When creating a product, people generally have good intentions. They believe their product will improve people’s lives – and there is nothing wrong with that. But beyond good intentions, it is imperative to actively consider a product and its features in the complex real-world environment of its users. As creators, we have a responsibility to do so.

This means asking a lot of questions that are currently uncommon to ask. Questions like: “What are the adverse effects of this feature?”, “How can this feature be abused?”, and “What if all products did this, would it still be helpful?”.

In order to create technological products in a responsible way, we must ask ourselves these kinds of questions. And we must be willing to sacrifice short term benefits for the long term wellbeing of our end users.