Lessons learned from Web2 social networking

Original: https://a16zcrypto.com/social-network-status-traps-web2-learnings/

As more and more crypto people explore social, I find myself often talking about what makes social networks work, and sometimes don't. A key conversation is around status. Here are some lessons I've learned from web2.

Social networks tend to promote content that promises to gain attention. Doing so incentivizes a specific type of behavior, providing status to the user performing that behavior. Generally, there is a status indicator that one must strive to accumulate. It comes in various forms - follower/like counts, experience points, certification badges, leaderboards, etc.

Naive implementations of the above methods often lead to a fatal flaw—concentrating status on a small number of "high status" users, while leaving most of the "low status" users, resulting in a poor experience for new users. While doing this maximizes value in the short term, it is a poor strategy in the long run as new users are locked out and overall network quality eventually degrades.

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

In other words: does a small percentage of your users have huge status?

This got me thinking about some personal views on social network design, and how social network builders should think of themselves more as models for economic policy.

1. Most social networks are skewed towards high status inequality (high Gini coefficient) by default.
2. If your social network has high inequality, you will have a hard time retaining newcomers.
3. Fluidity with high status is key to any viable social network, even if your goal is not to grow your overall user base.

What new problems does high inequality create?

The simplest understanding is status = capital. You want capital to flow around looking for healthy behaviors, not locked up, or in conflict with healthy behaviors.

Why?

1. Unhealthy behavior imitated by new users: Your highest status users already know how to play the status game - they know how to get millions of followers/answer the most questions/take actions that give them status. However, you probably don't want your new users to model this behavior, and imitation in social networks will naturally work against you.

Let's take the current twitter as an example: you may have noticed that a lot of tweets are just posts these days (how many times have you seen "a 1/37th post...?) although this might be someone getting the millionth Followers way, but it's definitely not something you want your new users to try.

2. People don't want to play a game they can't win: when a new user shows up in a social network, once they understand the basic mechanics, they accumulate some initial status: their first follower, their first karma, their first points. Then they check the global charts, or see how many followers their favorite celebrity has, or how many followers their peers have. If they see a person with a lot of karma and they have no way of approaching him, they get discouraged and turn to other, easier methods.

Social networks have a sharp version when you have to produce content - nobody wants to post a video/text/photo and it gets no response compared to normal.

It's human nature to discover how to play games/win status games, and if your users think your social network is too difficult to play, or have already been won by someone, they'll move on to other games.

3. Status NIMBYism: When you have a high status group, they usually try to prevent newcomers from gaining status.

You see this often when current users who are familiar with a web "meta" and don't like the change protest. Without high-status flexibility, you often end up in high-status groups who work together to keep newcomers out.

The list goes on and on, as do the various changes in September 1993. Remember when Instagram users protested the app’s release on Android? Or more recently, when Instagram shifted its focus from photos to short videos. These will continue to happen as the means of collecting status change.

How can status concentration be reduced and status mobility encouraged?

1. "Universal Base Status": A common mechanism is to grant newcomers a temporary status boost, often through algorithmic leverage that controls distribution and rewards.

You've probably noticed this if you sign up for a new account on any of the popular social platforms. Your content will get more recommendations, you'll get more boosts in friend recommendations, and that effect will wane over time.

There are various ways to build these mechanisms into the network.

Temporary elevate status: Assign a temporary elevate status at critical moments — eg: when someone new joins a network / when they come back after being away for a while / performs a critical desired action.

This boost is usually through an algorithm, giving the content a better chance of being seen, or getting new people to interact with it (“X just joined, say hello!”). In each case, you're "improving" someone new's chances of having a positive experience (and at a cost, as this improvement has to be at someone else's expense).

"Fair" distribution of status: Through some "fair" algorithm, status signals are distributed to users in the network. For example, build an algorithm that uses humans to decide who should appear on any recommendation page. This is one of the reasons for using a reverse feed—everyone has a fair chance to have their content seen.

Note: Status must have an inherent notion of scarcity to make sense. If you distribute status, you create inflation and can accidentally devalue your status signal, and you can't "replicate" new status without side effects.

2. Make Status Obscure: Another mitigation is to downplay all indicators of status and let people look for it. By blurring status, you are able to give yourself more options and allow players to focus on real game/app mechanics rather than status mechanics.

You can see examples of this in recent years. Instagram tries to hide the number of likes on a post, TikTok downplays the number of followers. All of these ambiguous state changes help to mitigate this effect, as well as other reasons for their existence. The downside of this approach is that if your social network is about status, without indicators people might not know what "game" they are playing.

3. Set up groups of people with similar status levels: If you've played any mainstream competitive game, you'll be familiar with "ranked" games (usually ELO ratings), where the game tries to group you with people of similar skill levels. together, so you're more likely to have a challenging but not impossible experience. Similarly, dating apps often try to use ELO-style mechanisms to divide people into those with similar "attractiveness".

One way for social networks to create great new user experiences is to provide a "ranked" experience where they can reach or interact with a subset of the entire graph. For example, a subreddit, not everyone competes on Reddit.

4. Reset or Decay Status Indicators: A radical measure against status centralization is to have each status indicator decay over time - a deflationary measure of status indicators.

For example, karma decays the longer you stay away from the network, or loses followers over time (especially if you gained a large following due to being on an early suggested user list).

As far as I know, no one has actually tried an extreme version of this logic: periodically setting all status indicators to zero and resetting the network from scratch might be an interesting experiment.

5. Resetting the "meta": One reason Instagram and Youtube's move to short-form videos has been controversial is that they "reset the meta" — a concept familiar to gamers everywhere. Doing this, combined with one of the mechanisms above, shakes up liquidity and changes who gets status in your network.

Accidentally creates status problems

Social networks often unexpectedly run into intractable status problems.

Accidental status hyperinflation: status is closely related to scarcity and/or more "proof of work". A common way to detonate your social network is to take a heretofore scarce or hard-to-obtain status signal and let it spread widely overnight, without regard to downstream effects. In many of these cases, you've either destroyed the network or made people aware of status in other ways that you didn't expect.

What does this have to do with high inequality? All too often you see the web trying to do this to fight inequality (reluctantly speaking up online and getting huge instant attention), only to create an even worse problem by devaluing a key reward mechanism. To quote a line from Superman: "If everyone is a superhero, then no one is a superhero".

Accidental Indicators of High Status: A related problem is accidentally introducing status indicators and creating inequalities when you didn't intend them to.

My favorite example is the "verified" badge on a social network. While the original meaning was "this person is actually who they claim to be Person X," a measure aimed at combating impostors, all networks may initially require "famous people" (read: famous to some extent ) promotion. oops! Thus, leading to him being widely understood as "this guy is famous in the world," something every network has struggled to this day.

Exacerbating status inequalities: A common pitfall of naive enforcement of discovery, ranking, or status is to inadvertently prevent newcomers from "joining in".

Any social experience has an attention or display mechanism that needs to take relative status into account. A naive "suggestion" feed or "top users" might be ranking content based on follower count - ensuring that people with large followings get more views, while new users on the platform never feel discovered. More often than not, such naive realizations exacerbate inequality and make it impossible for newcomers to climb the status ladder.

More often than not, such naive realizations exacerbate inequality and make it impossible for newcomers to climb the status ladder.

As Eugene Wei astutely points out, the biggest mistake you can make is not acknowledging that at the heart of social networking is "social capital". Understanding how capital is created, traded and spent will determine the success or failure of your network. Doing so might mean your role is more of a policy maker/economist than a traditional product maker/engineer.