a16z: What Can Web3 World Learn from Web2 Social Networks

a16zPublicado em 2022-08-15Última atualização em 2022-08-15

Resumo

Social networks tend to elevate content that it expects will get attention. Doing so incentivizes a particular type of behavior that lends status to the users performing said behavior. Typically, there’s a status indicator that people have to try and accumulate. It comes in various forms – karma, follower/like counts, XP, the verified badge, leaderboards, etc.

As more people in crypto explore social, I find myself often talking about what makes social networks work and, sometimes, not work. One key conversation is around status. Here are some lessons I learned from my web2 days.

Social networks tend to elevate content that it expects will get attention. Doing so incentivizes a particular type of behavior that lends status to the users performing said behavior. Typically, there’s a status indicator that people have to try and accumulate. It comes in various forms – karma, follower/like counts, XP, the verified badge, leaderboards, etc.

Naive implementations of the above often result in a fatal flaw – concentrating status in a few “status rich” users leaving the vast “status poor” and ensuring newcomers have a bad experience. While doing so might be value maximizing in the short run, in the long run it’s a bad strategy as new users can’t break in and eventually overall network quality decays.

First, how do you model status in a network? The Gini coefficient is typically a measure of wealth inequality: the more the inequality, the higher the number. For social networks, we can use this as a measure of relative status distribution and use your network’s indicator of status (followers/karma/etc.) as wealth.

In other words: do a small set of your users have outsized status?

This brings me to some personal beliefs on social network design and how social network builders should think of themselves more as modeling economic policy.

Most social networks tend to bend towards high status inequality (a high Gini coefficient) by default.

If your social network has high inequality you are going to struggle to retain newcomers.

Having high status mobility is key to any vibrant social network, even when your goal is not to grow your overall user base.

What newcomer problems stem from high inequality?

The simplest way to think of this is status=capital. You want capital to move around and seek out healthy behavior and not be locked away or used at odds with healthy behavior.

Why?

1. Newcomers mimic unhealthy behavior: Your highest status users have figured out how to play the status game – they know how to get millions of followers/answer questions with the most karma/ do the actions that grant them status. However, that behavior may not be what you want your newest users to model themselves on. The natural mimesis that occurs in social networks will then work against you.

Let’s take a current Twitter example: you might have noticed how a lot of tweets now are just threads (how many times have you seen “a 🧵1/37…?) . While that may be the way for someone to get their millionth follower, it definitely isn’t what you want your new users to try doing.

2. People don’t want to play unwinnable games: When a new user shows up in a social network and once they figure out the basic mechanics, they’re going to accumulate some initial status: their first followers, their first karma, their first points. They will then look up the global leaderboard or see how many followers their favorite celebrities have or worse, their peers have. And if they see someone with 100 gazillion karma and they have no means to get close, they will get disheartened, bounce and use something else easier.

Social networks have an acute version of this when you have to produce content – no one wants to post a video/text/photo and have it publicly get no reactions compared to what the norm is.

It’s human nature to try and figure out how to play/win status games and if your users perceive your social network as too hard to play or already won by a certain set of people, they will move onto another game.

3. Status NIMBYism: When you get a certain group with high status, it is common for them to try and keep out newcomers from attaining status.

You can see this often when there is a protest from current users familiar with a certain network “meta” who don’t like change. Without high status mobility, you’ll often get groups with high status that will work together to keep out newcomers.

There are too many examples to count and all variations of September 1993. Remember when Instagram users protested the app launching on Android? Or more recently, Instagram shifting focus to short form video from photos. These will continue to occur as the means to gather status change.

How can you mitigate status concentration and encourage status mobility?

1. “Universal Basic Status”: A common mechanism is to give temporary status boosts to newcomers. This is typically done through algorithmic levers that control distribution and rewards.

You might notice this if you sign up for a new account on any popular social platform. Your content will get recommended more and you’ll get elevated more in friend suggestions, an effect that will decay over time.

There are multiple ways to build these mechanisms into your network.

Temporary boosts to status: Distribute a temporary boost to status at key moments – example: when someone new joins a network/when they come back after a while away/perform a key desired action.

This boost is typically algorithmic where the content is given more chances to be seen or the newcomer interacted with (“X just joined, say hi!”). In each case, you’re “boosting” someone new’s chances of having a positive experience (and incurring a cost since that boost must come at the expense of someone else)

“Fair” allocation of status: Through some “fair” algorithm, distribute status signals to users of your network. For example, have an algorithm that cycles through people to decide who to show on any recommendation surface. This is one of the arguments for having reverse chronological ranked feeds — everyone has a fair chance of having their content seen.

Note: status has to have inbuilt notions of scarcity for it to matter. If you’re distributing status, you’re causing inflation and might accidentally cause your status signals to be devalued. You can’t “print” new status without side-effects!

2. Make status obscure: Another mitigation is to downplay all indicators of status and make people seek it out. By making status obscure, you give yourself more options to have people focus on the actual game/app mechanic and less on the status mechanic.

You see examples of this direction in recent years. Instagram trying to hide the number of people who liked a post, TikTok downplaying the follower count. All of these status obscuring changes help alleviate this effect among other reasons for them to exist. The downside of this approach is if your network is about status, without indicators people may not know what “game” they’re playing.

3. Set up cohorts of people with similar status levels: If you play any mainstream competitive game, you’ll be familiar with the concept of “ranked” (typically ELO rated) games where the game tries to set you up with people of similar skill levels so you’re more likely to have a challenging but not impossible experience. Similarly, dating apps often try to bucket people of similar “desirability” in an ELO-esque mechanism.

For a network, one way to make for a good newcomer experience is to have a “ranked” experience where they are exposed to or interact with a subset of the entire graph. For example, a sub-reddit instead of everyone competing with everyone else on Reddit.

4. Reset or decay status indicators: One aggressive measure to battle status concentrations is to have every status indicator decay over time – a deflationary measure for your status indicators.

For example, karma that decays the more you stay away from the network or losing followers over time (especially if you have gained a lot of followers from being on an early suggested user list).

To my knowledge, no one has really tried the logical extreme version of this: set all status indicators to zero periodically and reset the network from scratch. Might be an interesting experiment to run!

5. Reset the ‘meta’: One reason the Instagram and Youtube moves to short form video cause controversy is that they “reset the meta” – a concept familiar to gamers anywhere. Doing so in combination with one of the mechanisms above shakes up mobility and changes who can gain status in your network.

Accidentally causing status issues

Social networks often accidentally run into status issues that are hard to unwind.

Accidental status hyperinflation: Status is very connected to scarcity and/or having high “proof-of-work”. One common way to blow up your social network is to take a hitherto scarce or hard to attain status signal and make it widespread overnight without thinking through the downstream implications. In a lot of these cases, you either blow up the network or cause people to figure out status through other means you didn’t intend.

How is this connected to high inequality? You often see networks try and do this to battle inequality and wind up creating worse problems by devaluing a key reward mechanism. To quote The Incredibles “If everyone’s super, no one is”.

Accidental indicators of high status: A related problem is accidentally introducing status indicators and causing inequality when you don’t intend to.

My favorite example is the “verified” badge on social networks. While originally intended to mean “This person is actually X who they claim to be”, a measure intended to battle impersonation, all networks originally rolled it out to only “notable people” (read: famous in some way) who might need it. Oops! Thus leading to its broad understanding as one of “This person is someone notable in the world”, something every network tries to battle to this day.

Compounding status inequality: One very common pitfall for naive implementations of discovery, ranking or status is to inadvertently stop newcomers from “breaking in”.

Any social experience will typically have attention or display mechanisms that need to take relative status into account. A naive implementation of a “suggestions” feed or a “top users” might be to rank content based on follower count – ensuring that people with large follower counts get more views and people new to the platform never feel discovered. Very often, such naive implementations cause compounding inequality and make it impossible for newcomers to climb the status ladder.

Very often, such naive implementations cause compounding inequality and make it impossible for newcomers to climb the status ladder.

The biggest mistake you can make – as astutely pointed out by Eugene Wei – is to not acknowledge how social networks have social capital at their core. Understanding how that capital is created, traded, and consumed is going to make or break your network. Doing so might mean thinking of your role more as a policymaker/economist than a traditional product builder/engineer.

Leituras Relacionadas

A Guide to Grayscale’s ‘Bottom Fishing’: Using Cash Flow to Assess Cryptocurrency Value

**Title:** Grayscale's Guide to Bottom-Fishing: Valuing Cryptoassets Using Cash Flows **Summary:** This report by Grayscale Research presents a fundamental valuation framework for cryptocurrency assets, moving beyond pure speculation to analyze those with underlying cash flows. It distinguishes between "commodity-like" assets (e.g., Bitcoin) and "cash-flow" assets, primarily within DeFi. Using the leading decentralized lending protocol Aave as a case study, the analysis applies traditional financial methodologies like Discounted Cash Flow (DCF) and Price-to-Earnings (P/E) multiples. Key findings indicate that AAVE tokens are currently undervalued. Despite recent challenges, the protocol's strong revenue growth, ~50% net profit margin, and diversified treasury support a fundamental valuation range of $80-$100 per token (compared to a ~$75 market price at the time of writing). In a base-case scenario driven by stablecoin adoption and regulatory clarity, the fair value could rise to around $175 within a year. The report emphasizes that protocol success does not automatically translate to token value. It critically examines the "value capture" mechanisms—such as buybacks, burns, and staking rewards—that channel protocol profits to token holders. Furthermore, it addresses the legal and governance complexities of Decentralized Autonomous Organizations (DAOs), noting their difference from traditional corporate equity but highlighting how robust, transparent governance can align protocol economics with holder interests. The conclusion is that the crypto market is maturing, with capital increasingly flowing towards projects with demonstrable fundamentals, real adoption, and disciplined capital allocation, creating opportunities for value-based investors.

marsbitHá 51m

A Guide to Grayscale’s ‘Bottom Fishing’: Using Cash Flow to Assess Cryptocurrency Value

marsbitHá 51m

After semiconductors lead the gains, are funds buying into AI orders or a macroeconomic rebound?

After US-Iran talks led to a temporary ceasefire and framework for reopening the strategic Strait of Hormuz, U.S. stocks rose on June 18, with the Nasdaq gaining 1.9%. The semiconductor and AI hardware sectors outperformed. This rally stemmed primarily from reduced geopolitical risk, which lowered oil prices and inflation expectations, easing discount rate pressure on high-valuation growth stocks like tech. The key question is not whether tech rebounded, but the nature of the rebound. The market appears to be selectively repricing AI infrastructure plays rather than broadly chasing AI narratives. Gains were concentrated in chips, optical interconnects, memory, and domestic manufacturing—segments tied to tangible data center build-outs and capital expenditure. Intel's ~10% surge, fueled by a Trump statement about potential Apple collaboration, exemplifies this mixed dynamic. It reflects policy catalysts and domestic manufacturing sentiment more than confirmed fundamentals. Meanwhile, strong earnings from companies like Astera Labs (revenue up 93% YoY) provided concrete evidence of AI-driven demand in hardware. In essence, the rally represents a risk-premium recalibration. Lower Middle East tensions opened a valuation repair window, and capital flowed first into AI infrastructure segments with visible near-term revenue streams. The sustainability of this move hinges on upcoming Q2 earnings, specifically continued strength in cloud provider capex, AI server orders, and hardware company guidance. Policy hopes alone are insufficient; the cycle needs validation from orders and financials.

marsbitHá 57m

After semiconductors lead the gains, are funds buying into AI orders or a macroeconomic rebound?

marsbitHá 57m

The Entire Internet Hails Noam's Joining, But OpenAI's Loss Bill Just Got Thicker

While the AI community celebrates Noam Shazeer, co-author of the "Attention Is All You Need" paper, joining OpenAI as Head of Architectural Research, the company's audited financials reveal a starkly different reality. In 2025, OpenAI reported $13.07 billion in revenue but a massive $20.92 billion operating loss. Even excluding a one-time accounting charge, the cash burn is severe, with $3.7 billion consumed in Q1 2026 alone. This high-profile hiring occurs against a backdrop of significant internal research talent drain, with key founders and researchers departing as the company's focus shifts from exploratory research to product iteration. Meanwhile, OpenAI's fundamental business model faces a deep crisis. It paid Microsoft $10.59 billion for compute in 2025, while its vast user base of 9 billion weekly actives includes only 50 million paying customers, making growth a direct driver of escalating costs. The article argues Shazeer's recruitment is less about technical necessity and more about crafting a compelling narrative for OpenAI's upcoming IPO, aiming to justify a rumored $1 trillion valuation to future public market investors. It contrasts OpenAI's strategy with Anthropic's reported path to profitability, which relies on a strong enterprise customer base and cost control, rather than star-powered narratives. Ultimately, the piece concludes that while Shazeer's architectural work may take 1-2 years to materialize, OpenAI's financial clock is ticking much faster, with its massive losses undercutting the celebratory headlines.

marsbitHá 2h

The Entire Internet Hails Noam's Joining, But OpenAI's Loss Bill Just Got Thicker

marsbitHá 2h

Trading

Spot
Futuros
活动图片