Using Incentives in Communities

This post is going to be much more crude and short.

Crypto puts your money squarely into where your mouth and mind is. We call that incentive alignment. In my opinion that alignment is one of crypto’s killer traits and at the same time one of core reasons why many people find it repulsive.

Obviously I ultimately believe such alignment scheme will prevail. However it will take a long time for most people to get around the concept and be okay with it. I still struggle day to day in grasping its implications.

Recently I have been thinking more and more about how that incentive alignment can be applied in other areas of life like social media, community management, and combating misinformation.

For instance, let’s say that I need to “stake” or spend 1,000 in-community point (similar to Reddit’s Karma points) to write a post in my favorite online community. Then the community’s moderation bot finds that it contains misinformation. Subsequently readers flag again that it’s indeed the case, betting their own points. As a result, not only my post is removed, but also my 1,000 points are slashed.

For positive feedback loop, you can structure it in a way where when you share some revealing scientific truth, people like it, you get more points and gain more viewership. (Tiktok + Community)

Of course, I’m simplifying lot of steps here and the devil will be in the details. But what’s important is that the platform and community have a non-trivial incentive and penalty system that can moderate contents. Only problem is how to design that system.

I believe when incentive is aligned towards truth, kindness, and coalition, rather than misinformation, hatred, and division, it can be a way to solve our social media problem.

It’s not an easy problem, but essence of shared social space is moderation and doing that well. I’m pretty comfortable in saying that a lot of it should be the job of digital platforms with active contribution from their own communities. That’s what platforms are meant to do, not become some soulless protocols without any active interventions. And that’s why they are so hard to get it right and at the same time such a good business. Because very few will be able to strike the balance.

Every system will be eventually gamed and it’s not about how you game it, but how you change as you try to game the system. As long as that change is positive, I believe such systems are worth experimenting.



Leave a comment