Due to centralization and opaque rules leading to a "points farce," how to create the best airdrop strategy?
Author: Nancy, PANews
In the era of the internet, the points system is an important tool for maintaining user engagement and increasing user loyalty. Nowadays, this incentive method is being used by Web3 projects for early-stage fundraising and user acquisition, and under the stimulation of airdrop expectations, behaviors such as script-based point farming have emerged, reflecting FOMO.
At the same time, recent incidents such as the "shrinkage of staking points" blunder from the Ethereum liquid staking project ether.fi and the strict monitoring of malicious trading behaviors by the crypto exchange Backpack have exposed issues faced by the current points mechanism, including centralization of the points system and the rationality of its rules.
With points becoming popular in Web3, does this model have sustainability? How should project teams scientifically set airdrop rules to maximize inclusive airdrops?
Divided into POS and POW incentive points, transparent airdrop types will become mainstream
From Friend.tech, Blast, Linea to various staking/re-staking protocols, points seem to be becoming a popular "standard feature" for current Web3 projects. According to a recent report by The Block, there are currently 14 projects in the market that have issued over 111.5 billion points.
However, the mechanisms of these points vary. According to Simon, CTO of Web3 identity and reputation infrastructure Trusta Labs, who spoke to PANews, for example, Arbitrum uses a retrospective points system that quantifies the number of tokens a user receives through retrospective analysis of historical behavior on chain addresses; Linea employs a points system based on transparent interaction tasks, where user behavior is directly reflected in the points system, and even the subsequent mapping of points to Linea tokens is transparently displayed to the community.
Trusta Labs has rich experience in data-driven airdrop scheme design and user analysis, including providing user analysis and airdrop data services under different types of incentive methods for five popular public chains: Celestia, Manta, Starknet, Linea, and Arbitrum. Simon further pointed out in the interview that points are also classified into POW (Proof of Work) and POS (Proof of Stake) types based on the different types of rewarded behaviors. For instance, Linea focuses on user interactions on-chain, where users contribute interaction behaviors and a small amount of interaction gas fees, which belong to POW points; projects like Eigenlayer that focus on TVL are typical representatives of POS points, clearly stating that points are used to measure the total re-staking contribution of Eigenlayer (ETH * hours as the unit).
In fact, for Web3 projects, the points mechanism can not only help projects quickly achieve cold starts and build a pyramid of loyal users, but also effectively avoid regulatory risks compared to previous methods like ICOs.
However, with the announcement of more and more airdrop plans, points have also become a hot topic in the market. In order to bet on the potential value of these points, off-market trading and leveraged trading have emerged, and the overly speculative phenomenon has caused the points system to gradually deviate from its original intention of encouraging users to use products, making it easier for users to incur losses. As a result, many projects have urgently halted such speculative behaviors, including social media giant Reddit and blockchain media Coindesk, both announcing the cancellation or suspension of their points plans.
However, many projects still use points as marketing leverage to attract users and funds, being vague about airdrop rules and token issuance timelines, leading the community to question whether they are being manipulated. Simon candidly stated in the interview that since the project parties did not determine the points calculation logic of the incentive system during user interactions, retrospective airdrops are more likely to be manipulated. In the future, transparent points-based airdrops may replace retrospective airdrops and increasingly become mainstream in the market.
Caught in transparency doubts, suggesting the introduction of on-chain points and third-party points platforms
As an emerging incentive tool, the points war is intensifying, while its lack of transparency is being criticized by users.
Specifically, since the vast majority of projects have not disclosed specific information such as the total amount of points, only providing public leaderboards for users to check rankings, participants can only estimate by counting the points of top-ranked users. This means that users find it difficult to accurately assess the potential value of the points they receive, making it impossible to judge the value of participating in the project.
At the same time, this centralization risk can lead to malicious project parties using it to sell tokens or arbitrarily change the token and points exchange rates. Shen Yu, co-founder of Cobo & F2pool, commented on similar issues on social media, stating, "In the past, DeFi directly issued tokens, with data on-chain being open and transparent; now, after layering five times, various centralized servers give points, focusing on a blind box, relying on information asymmetry for gaming. Guess how many connections and insider trading there are, I'm exhausted…"; KOL @0xSeaside also expressed that the current points algorithm is opaque and entirely determined by the project party, with the possibility of them defaulting at the end; and community users joked, "The first wave of insider trading drank all the soup, leaving a few drops for retail investors, then said retail investors didn't claim points, blaming them."
Over time, the lack of user trust will make the points system difficult to sustain. Recently, ether.fi was questioned by users for secretly changing points due to incorrect point displays, with some members stating, "The promised 100% points for users have become a profit source for the project party." Although the official team has fixed the issue, it also exposed the inherent fragility of user trust established through off-chain points.
"Off-chain points are based on a centralized project party database for statistics and storage, which indeed allows for the possibility of malicious project parties engaging in insider trading with fake accounts and points. At the same time, the total issuance of points, as well as the subsequent token exchange ratios and methods, have never been determined and made public like TGE," Simon stated to PANews.
Some projects in the market have begun to focus on the transparency of points information. For example, the LRT protocol Eigenpie recently opened its "points calculation subgraph code" to all users, and the Ethereum re-staking protocol Renzo launched a "points details" feature to provide users with the source of their points.
Eigenpie opens points calculation subgraph code to users
In response, Simon proposed two solutions. First, use on-chain points, as the on-chain points implementation of Linea XP allows everyone to clearly see the addresses of over 500,000 holders and the number of points they have received; second, Trusta Labs is also working on building a fair and auditable third-party points platform, allowing project parties with points needs to publish their points on this platform.
How to create the best airdrop strategy? These textbook projects are worth learning from
Ultimately, the popularity of the points mechanism is mainly due to its airdrop incentive expectations, attracting a large number of airdrop hunters under the wealth effect. Although this method can help projects quickly gain traffic and popularity, the witch-hunt driven by interests affects the fairness of the reward distribution mechanism. With the influx of many opportunists, project parties have begun to increase scrutiny of witch-hunts to protect the interests of real users, but this "witch-hunt action" has also caused dissatisfaction within the community.
For example, recently, after the crypto exchange Backpack opened its trading points activity, its trading volume topped the charts for several consecutive days. However, the project officially announced that it would take measures against malicious trading behaviors, leading many community users to express feeling manipulated, as the project party did not declare this in advance, and those who contributed real money were treated unfairly by being subjected to witch-hunt prevention.
Simon suggested that a transparent points incentive system must proactively use proof of humanity as a participation threshold. The reason is that when the rules are transparently presented to the community, attackers can more easily forge fake accounts and use scripts for trading. For instance, since the Linea DeFi Voyage event, the project has only allowed participation and accumulation of LXP points through a specified number of proof of humanity systems. In this process, the on-chain reputation attestation and proof of humanity attestation provided by Trusta Labs are the two most important verifications, helping over 500,000 users claim over 1 million attestations, effectively keeping opportunists out of Linea's task activities. This method of issuing proof of humanity before allowing participation in activities can avoid complaints from community users who invest real money only to receive nothing, while also allowing genuine users who wish to participate to receive project incentives more inclusively.
He also emphasized that project parties should consider both POS and POW weights when designing the points system. The reason is that in purely POS incentive-based points like Blast and Eigenlayer, the top-ranked users are often large investors, which can exclude many genuine users who are willing to participate, and may not gain broad support for the project within the community.
In fact, designing an airdrop strategy that balances the interests of multiple parties has always been a challenge for projects. "Contribution and incentives are a game between users and project parties. Airdrop schemes that satisfy everyone may be an infinitely close dream. Between early users and continuously investing users, between so-called low-income accounts and premium accounts, and between real individual users and batch-controlled users, project parties must make trade-offs and cannot cater to all," Simon believes that the airdrop strategies of the three public chains, Arbitrum, Celestia, and Starknet, have valuable lessons to learn. First, Arbitrum's airdrop established the first relatively complete points system, effectively incentivizing user contributions through tiered token distribution, considering various behaviors from Arbitrum One and Nova; Celestia selected real users worthy of incentives through data analysis of full-chain Rollups, completing the first data-driven airdrop design; recently, Starknet's inclusive airdrop approach has created a new paradigm for airdrop incentives, aiming to cover more ordinary users and developers in terms of the number of incentivized participants, ecological coverage, and outreach to non-Web3 developers.
In summary, as the scale of points continues to expand, projects should provide a more open and transparent points mechanism, including on-chain points and third-party points platforms, to better face user scrutiny and airdrop controversies. More importantly, when designing points, projects should start from the perspective of incentivizing users to genuinely use the product, as this will directly impact the product's lifecycle.