The point of it all
A quick sojourn through the past to figure out how we got to points.
“Show me the incentive, and I’ll show you the outcome.”
The summer of 2020 is fondly known as DeFi summer, started by the lending market Compound launching its token, $COMP.
Instead of being sold, the token was instead dripped to its lenders and borrowers—the very users of the protocol. The value of this extra token, alongside the existing lending rates, set Compound leagues ahead as the best place for lending yields.
This sparked what came to be known as yield farming, catapulting Decentralized Finance to new heights. Almost every other day, there would be a new project releasing a token ready to subsidize their platform. Yields across the entire ecosystem were at minimum 20-30%.
But this reward mechanism skewed being early—be the first depositors, be the first token buyers to reap the best token-specific rewards. There were criticisms, there were 2.0’s, but most importantly, there were prices that crumpled in the bear market.
With a dearth of tokens to prop up yields, it was time to reformulate strategies. Airdrops became en vogue, especially with the success of Optimism and Arbitrum.
For what it's worth, Optimism and Arbitrum were large marquee chains that had attention and uses, even without incentives. But for others in the bear market looking to stand out, incentive structures were necessary to grasp the spotlight.
Enter: points.
Incentives, compared
A spectrum of all incentive options thus far would look like this.
On one end, there's yield farming. Explicitly defined, linear rewards usually enforced by a pre-defined smart contract.
On the other side: airdrops. Implicitly encouraging aligned behavior, with rewards just suddenly being handed out.
And points sit in the middle. The method of accumulating points is instant and explicit, like in yield farming, but how and when those points are translated into tokens are as surprising as airdrops.
Points ultimately offer some of the most variability, out of all the methods in the spectrum.
One example was GMX, the decentralized perpetuals exchange, and its points program.
In 2022, users were incentivized with loyalty points to keep providing liquidity for the exchange. These points offered extra boosts for a share of the yield, as well as locked GMX. The user could choose to unlock those tokens over a year, or keep it locked but staked to earn liquid rewards from the protocol.
Explicit directives, explicit rewards, with some loyalty mechanics.
But most point programs today borrow a lot from Blur. It emerged in 2022 to challenge the NFT marketplace leader at the time, Opensea.
It was the first time that there was an explicit social component to crypto incentives. The platform was closed to referrals to begin with, and then upon launching with points, it also a introduced a leaderboard and bonus multipliers for the top traders.
What was not clear was how those points would translate into tokens. But it turned out to be a rewarding airdrop, teed up with a Season 2 immediately after. With the way it gobbled up marketshare, it’s understandably been the inspiration for many, for better or for worse.
On points
Points do offer some improvements over traditional yield farming.
1. Points facilitate price discovery.1
2. Points allow for more flexibility.
One of the most necessary components to incentivize in yield farming was the Pool 2, the liquidity for the token itself. This usually came with the highest rewards, as well as highest reflexivity.
But at the start, the project would usually launch with a limited liquidity pool. In the beginning, with barely any distributions, it was common for the price to be driven up as people due to the supply-demand dynamic.
So as the project matures, while its metrics and functions could improve, it could all pale to the cardinal sin of never meeting its unreasonable starting price.
It’s understandable then why points are preferable. This allows the protocol to incentivize users and grow the project's momentum before releasing a token.
In the best scenario, it starts this dance where as the project's value becomes apparent, this drives up the perception of the future token. As more users converge to partake in the program, the better the metrics, the better the expected value: repeat.
Points also enable crypto builders to iterate.
Traditional yield farming was constrained to locking tokens into an onchain smart contract. Course-correction would have been expensive, versus just tweaking parameters for an offchain points program.
But now, the main con:
Points can be predatory.
In every other token distribution method, the users know what they’re signing up for. It’s a purchase of tokens at a certain price, or this much tokens at every block for this much capital. Or surprise, enjoy an airdrop.
With points, it’s just a black box. The points are visible, but when do the points end? How do the calculations go? How are users going to be compensated for risk?
At its worst, points are lulling the crypto space to accept blind risk-taking.
And it’s a reckless precedent to accept. In yield farming, there was still the opportunity to sell unlocked rewards over time. With points, an exploit would be twice as bitter: lose the funds, and lose the value of future tokens.
So what’s the point
I’ve been a big fan of points so far. I was such a fan of GMX’s program, such an underrated incentive structure. I still firmly believe that points are amazing because they allow for flexibility.
What's dimmed my optimism is that a lot of programs never utilize this flexibility. Instead, everyone just puts out the same program with no adjustments throughout it all.
And alas, the points do work! Slapping a number-go-up simulator and a referral bonus does drive deposits and activity.
But this only manufactures a sense of product-market fit for the short term2. In the long term, though, of course everyone gets disappointed. The points are specifically implemented to get a token out the door, which attracts users interested only in the carrot, not in the protocol itself. Mercenaries all the way down.
In the absence of a large exploit, though, there’s probably one last iteration of points before it overstays its welcome. Projects still need a way to garner attention3, outside of paying people for it upfront.
So here’s what makes a good point program:
Identify a flywheel for a core metric.
In DeFi, it’s usually liquidity. What most don’t realize is that idle liquidity isn’t a good metric, and it’s necessary to boost trading volume, for example.
It can be tricky to identify a system, but it’s important to incentivize around it. If the points are taken away and everything grinds to a halt, then it’s essentially burning money to boost nothing that will last.
Have a clear plan for when it ends.
Base it on a Total Value Locked or other milestone
Future rounds tend to dilute initial participants, so if necessary, place a lower weight on future seasons
If tokens are for distribution, then actually distribute it.
What would power users look like? Then implement, like you can earn loyalty points or multipliers for LPing week-by-week (inspiration: GMX4). This curates for better user-ambassadors
Convert the points to tokens, but set aside extra and give top leaderboard placers the right to purchase
Because at the end of the day, points are a neutral tool. They can be crafted to catapult a project’s momentum or be hastily put together for the quick need to tokenize.
There are enough perverse incentives in this world. Crypto doesn’t have to add to it with meaningless points.
Credit goes to @chudnov for putting it into words - points are futures contracts on top of airdrops. Yes.
Don’t want to specify names, because even after egregious outflows, some still end up with more TVL than before. Points work, unfortunately.
There’s this little tension where everyone is unhappy with VC favoritism. But ironically, people rely on VCs and large raises as validation that a project is worth something.
Decentralized fundraising doesn’t quite provide the same luxury of attention right now, though. It will be interesting to see how echo.xyz impacts anything, if at all, and how this interplay evolves.
It was popular to lock up tokens for 4 years to gain a “points” value, which would then decay over time. But people spun up pegged assets, which ended up making made those locked tokens very liquid. When it was popular to choose to lock and leave, GMX was about choosing to stay.
This mirrors points today. The worst ones ask you to lock in then leave with your tokens, but the best programs convince you to stay.