Research suggests that the average adult makes upward of 35,000 decisions each day, and 226.7 alone on just food. Think about how you pick which restaurant you want to eat at, what stock to invest in, which of the thousands of options on Amazon you’ll compare before purchasing. The problem we face today is not the lack of options, but the abundance of them and along with it, an overwhelming amount of information that causes decision fatigue when we try to make the right decisions, regardless of how big or small they are.

What you might not realize is that many of the decisions we make are sped up by reputation systems that are either naturally occurring in our social network through close friendships, or set up by companies that act as an intermediary trust layer. Think Yelp, Amazon, Google as examples of companies that facilitate all types of human transactions and interactions through their rating and review systems.

At our society’s core, social attestation is the foundation for a reputation system and acts as a necessary trust layer for most decision making. Reputation systems not only improve the quality and safety of our decisions, but also allow for efficient decision making.

Reputation as a product/service in web2

There is a slew of companies in web2 that provide reputation as a product or service. Traditional banks are centralized financial institutions whose core value proposition is trust. They give us confidence to make financial transactions because we can transact faster with a trust layer, but also because there’s almost no other safer, better choice.

The meaning of the financial term “credit” signals trust between two parties, which allows one party to provide money or resources to another party, who promises to repay or return those resources at a later date. Credit score is essentially a reputation score.

Many successful companies leverage the very idea that people need social attestations to serve as a trust layer for decision making. Each company provides a trust layer for different types of decisions.

• Where to stay when traveling: Airbnb

• Which restaurant to try out: Yelp

• What information is relevant: Google (PageRank is one of the earliest ranking algorithm)

• What product to buy: Amazon

Of course, many of these companies provide other services and promises as well, e.g. Airbnb has homeowners’ identification information and is responsible for user safety, Amazon guarantees two-day shipping and product quality. But think what and where these companies would be without a community review and rating feature.

Let’s also not overlook the fact that communities have been born out of this exact feature on Reddit. Quora connects people with questions to people with answers and does so effectively with social attestation features like upvotes. Stack Overflow is the Quora for developers with technical challenges and encourages quality, helpful answers by assigning reputation to individuals as a “rough measurement of how much the community trusts you.” Product Hunt enables search and discovery of new applications and products through community-based comments and voting system. All of these platforms provide value to users through reputation systems enabled by community review and rating features, which is perhaps the most explicit form of social attestation.

The caveat with web2 companies using reputation systems, however, is that the system is centralized, and the data and algorithm can be manipulated.

“In web2, you're influenced by all the algorithms controlled by the big companies. They feed you whatever they think that you want,” says Erazer Li, founding partner at SevenX Venture.

Advertising is an essential revenue stream for web2 companies, which inevitably corrupts transparency and intention of its reputation rating. Take content suggestion as an example, it is well-known that most social media companies’ algorithms are built with retention as a main goal: the longer they keep you scrolling on their platforms, the more advertising revenue goes into their pocket.

“People say, when you don’t pay for a product, you are the product,” says Hans Xiang, senior investment associate at Sino Global Capital.

A product’s reputation is so important to its sales and popularity on Amazon that there are underground markets for fake Amazon reviews where sellers will send free goods to real people in exchange for glowing write-ups.

In 2019, Airbnb was accused of manipulating its search results to favor higher-priced listings and to hide lower-priced options. The company denied the accusations but still made changes to its algorithm to address concerns.

There have been reports of Uber drivers creating fake accounts to give themselves high ratings or of businesses paying for fake positive reviews on Yelp.

These actions go against the principles of fairness and transparency that underlie user generated content and damage the trust that users have in these platforms. It’s impossible for web2 users to keep companies’ algorithms and ranking systems transparent, unbiased and fair to all users because they are centralized institutions. Although user data is what supports the reputation system and gives companies profit, they do not have a say in how the system should be run or how data should be processed.

Web2 uses a basic reputation system for the sake of efficiency, but simple isn’t always smart. Although that may have been the lure of web2, we must strive towards higher standards for reputation systems in web3.

How we make decisions in web3?

Web3 is expanding rapidly and with that comes the explosion of options and on-chain data that make it virtually impossible for people to navigate the space and make social, financial decisions safely and confidently.

“Even though crypto and blockchain today are decentralized and transparent, there are a lot of noise and fragmentation,” says Xiang.

Some estimates say there were about 500,000 NFTs in 2020. Today there are 111M NFT on Open Sea alone. Around 5000 DAOs exist in web3 now with more than 978,000 members. According to Shier, co-founder of NFTScan, their latest data shows around 5000 NFT smart contracts deployed on-chain and 2 million NFTs minted every day.

Our need for reliable sources of information extends to the world of web3, where a lack of established social connections and credible information sources can make it difficult to navigate the space effectively.

It’s not uncommon in crypto for users to get roped into sham projects or be majorly misled when learning something on their own. We’ve seen one too many times investors, both new and experienced, lose their savings to rug pulls when decisions are made based on online information, Twitter hype and even their own biases.

“A social graph of trust is probably a small graph,” says Christian Montoya, senior product manager of MetaMask Snaps. “When you think about your interactions in web3, you probably have a small handful of people that you really trust.”

This creates challenges for both new and existing users seeking trustworthy projects and opportunities to invest in. Without a reliable way to assess credibility, many of us are only able to access a small fraction of the vast crypto space, hindering us from truly widespread adoption.

Web3 is missing a permissionless information ranking and suggestion algorithm that users can trust, that utilizes on-chain data to recommend and rank trustworthy and relevant information.

How will reputation system in web3 be different

A peer-to-peer reputation system that acts as a trust layer to coordinate in a decentralized, pseudo-anonymous environment might be the most resilient choice. Different from web2, the reputation system in web3 can operate on an open data layer and open-source honest algorithms, which can drive composability. This system enables and encourages transparency in the choice of algorithms and reputation signals across various contexts.

The problem is that reputation in the decentralized context is hard.

“As a product company, we face product decisions everyday on how to help users signal value out of noises, and we have to keep the balance between what we need to decide on our own and what users want themselves to decide.” says Evgeny Yurtaev, co-founder and CEO at Zerion.

The main questions that needed to be answered were: how to bring the system on-chain and how to solve reputation through a core infrastructure approach. After all, as Li mentions, “Someone's noise may be someone else's value, and someone's value may be someone else's noise.”

Another important principle of a reputation system for web3 is that it has to be openly verifiable. Having a centralized computation backend restricts developers in many ways — centralized liability and point of trust perpetuates lack of trust from the community. Using a public reputation infrastructure, developers will be able to scale web3 apps quickly without having to manage an in-house data infrastructure or build proprietary algorithms, avoiding the risk of restricting themselves to centralized liability.

“For MetaMask, ultimately what we are doing is designing consentful user interfaces. We want to ensure not only the users know what is happening but also the information we give them is accurate and not overly simplified. We also explore ways to help users self-organize and interact with people they trust. That is why we introduce Snaps and try to build in features for users to contextualize data by themselves for reputation signaling,” adds Montoya.

The contextual nature of reputation demands that web3 reputation system be customizable. Developers should also be able to customize the open-source, generalized ranking computation based on what unique heuristics their application or community care about.

“The reputation/identity mechanism needs to be empowering to the users instead of just gating. Whether it's a scoring or ranking mechanism, it almost always needs to be contextual to either the web3 space, your Bankless Academy, your MetaMask network, your Gitcoin Passport, to wherever it is that you're participating,” says Kyle Weiss, executive director at the Gitcoin Foundation. “The tools that are able to adapt to that and focus on communities in a macro sense are going to be the ones that are most successful.”

Signal value out of noise

A reputation system must empower users to self-coordinate in the absence of gatekeepers. It must make it easy for developers to use public infrastructure that helps use contextual, relevant data and powers a verifiable computing system.

“The idea for Karma3 Labs was born because the we were web3 builders and developers ourselves and realized the need for reputation system in web3 as we were working on a layer 1 protocol,” says Sahil Dewan, co-founder of Karma3 Labs.

They built strong conviction studying how early design decisions around search, rankings and review-ratings systems made web2 companies acquire, engage and monetize billions of users.

A trustless reputation and ranking system is essential to the success and widespread adoption of web3, and when built with the right principles, it empowers developers, encourages people to self-coordinate and achieve consensus, and enables sovereign communities where people find a place of belonging in the vast world of web3.

“That's a very powerful thing,” adds Sahil Dewan, “It's not restricted to one app. That's the beauty.”