There have already been a lot of good thoughts and observations upthread, so we’ll focus on a few in particular. In just 2024, GFX sat on grants and procurement committees commanding around $300m, and we have a wide range of past grants experience as well. A hard-earned lesson not yet fully addressed:
Grants really need to be thought of as having a lifecycle. Disorganized, “put up a proposal on the forum and get to a vote” kind of grants not only tend to be poorly designed, but they don’t take into account that a grant is more than just disbursement of funding to the grantee.
This discussion already began with an intuitive understanding of that, with milestones being prerequisite to full grant funds disbursal. But there’s also an ongoing monitoring and then closeout process. Monitoring may actually be light touch or minimal if it’s all onchain activities, the grant is small, or the grant plan is designed to be short. But a closeout process is important where grantees produce a final summary of how funds were spent, were the goals met, and why or why.
Grants programs aren’t simply a way to fund promising initiatives. They’re an opportunity to gather knowledge about the grantor’s ecosystem of users/developers and their own product.
Realistically, there will be failures, and when a grantee fails, you want to know why - was it poor execution, the grant plan was just designed poorly, incorrect assumptions about how a technical or market force would work, or something completely outside the control of Morpho and its grantee? Conversely, there will be successes, and you have to ask similar questions: was this luck or good execution, what was the cost vs the perceived benefit, and most importantly is this repeatable and scalable?
And then you also need a mechanism – such as a grants program that runs on cycles instead of being always open – to take those learnings and iterate.
Unfortunately, a good grants program will certainly require at least a part-time administrator. This is because grantees will inevitably have unaddressed questions in real time that need to be figured out, the grant plans approved will need to periodically provide updates, and someone has to think about how to improve the program and its grants giving regularly and not just one afternoon in a Zoom call. Also, if something goes wrong, having someone who owns that failure (whether it’s directly their fault or not) also means having someone motivated to prevent things going wrong in the first place.
Lots of great thoughts on this thread so far, but we wanted to emphasize that a more formal program tends to be more costly to operate, but it also tends to have better spending discipline and accumulate knowledge and expertise that persists after a user incentive program is done or a widget is already built.