🚹

The Social Graph

Subtitle
Moving beyond 1:1 social media
Date
May 5, 2021
Type
Essay
Words
1746

Social media has no shortage of scapegoats: The business model, the fame junkies, the addiction fueled habit loops, the human traits of greed and narcissism, Elon Musk. After debating a friend, he landed on the sentiment that "social media is inherently screwed."

I have a more hopeful outlook, I just think we have the wrong culprit. Let's imagine that we put Facebook on a blockchain, stripped the algorithms, stripped the advertising, and removed the Like button: it would still suck.

Since Myspace, our origin, there has been a blindspot in the architecture of social networks. One fundamental shift in the nature of the "Add Friend" button could radically change the entire experience.

Learning from E-mail

I set up ConvertKit recently, and I'm super excited about the ability to use tags. You can't do this on an email platforms like Substack. Typically, a creator has a single list, and if you subscribe - you get everything!

I'm thinking of a tag as a content "Stream." Anyone who wants to hear from me has the ability to control exactly which facets of me they want in their Inbox. This list could get even more granular over time: life updates, music demos, updates on the Metaverse, surreal rambles, or even photos of my Dog's Dookie's.

image

It's a win for the audience - they don't feel spammed.

More importantly, it's a win for me - I don't feel like a spammer. Since I know that everyone in a given stream intentionally signed up, I don't feel the need to self-censor. There's no holding back. As my interests warp over time, and new projects come up, I can simply create a new Stream for them.

The tag/segment feature isn't unique to ConvertKit. MailChimp has it, and I'm sure other email marketing platforms offer it to. What's strange though is how it's completely lacking from ALL of our social networks.

I'd argue that binary 1:1 connections on social media are the main reason why feeds can feel toxic and unnatural.

Subtle Pivots

If you think about social media networks since Facebook, they've each been an iteration at the content level:

Twitter = Facebook with a word count

Instagram = Facebook with a required image

Snapchat = Facebook with Houdini content (it disappears)

While the user base, aesthetic, and culture of each network is substantially different, they all share a similar (sometime identical) underlying architecture:

  • A trophy profile
  • An addicting feed
  • A separate inbox for private DMs
  • A 1:1 social graph

Regardless of shiny new features at the content level (like video editing tools on TikTok), any network built off of this standard bedrock will inherent the same familiar problems.

A change in the nature of the social graph would effect almost every dimension of the social media experience.

Social Graphs

We've been swimming in primitive social graphs for a decade. Every network since MySpace has held the 1:1 connection as a given. The "Add Friend" button is part of our history, and it is borderline sacred.

If I want to connect with < R Y A N >, I'm forced to subscribe to EVERYTHING that he decides to share with his network: his Bitcoin prophecies, his cat photos, his Trump rants, his job updates, his #Ship30for30s, his latest NFT project, and his fortune cookie wisdom.

Maybe I love some of those things, but not all of them.

(( depends on < R Y A N > ))

This idea of subscribing to ALL facets of a person isn't natural. The thought of standing from a stage and sharing your thoughts with everyone you've ever met (plus strangers) is weird. We've basically normalized a Kurt Vonnegut satire. Social media doesn't reflect how real-life works at all.

The reason we have flat social graphs is because they are simple. There's the fear that anything beyond a 1:1 connection will be too complicated for a user. But the problems of this design decision have been manifesting in several ways.

  • Scattered Identities: Many people feel the need to carry several identities (personal, their business, their job, their art), across separate networks (Facebook, LinkedIn, Twitter)
  • Twitter Lists are a response to this problem, but it's a half-effective band-aid. Even if I have a VR list, a family list, and a crypto list - when the White House gets raided, that will spill over into every list.
  • Participation Rates: I'd imagine that most social media networks have terribly low participation rates. Users and monthly sign-ins might be in the billions, but there's a reason we never see stats for unique monthly posters. There are a loud and chaotic few among a mostly silent majority.

Most rage against social media these days is targeted at a broken business model. It's warranted, but misguided. Even if it were fixed, Facebook would still suck. We're suffering from a social graph that can't reflect real-life relationships.

The social graph is the thing ripe for transformation.

Streams

If all new social media networks are just a subtle pivot off of something familiar, what would be the "high concept" feature that sits on top of a reinvented social graph? Streams.

Anything you post is required to go into a "stream"

This one change affects everything downstream - the profile, the graph, the feed, the message.

  1. I am my streams: Instead of a person's profile being a gallery of their content, it could be a list of their streams. Their streams are their sub-identities that all pile together to represent their aggregated and messy self. Similar to the email service Hey, there can be a screener that shows new streams for people in your network.
  2. The sane feed: Instead of connecting to someone in a binary way, you get to follow 4 of < R Y A N ' S > 13 streams. This results in a SANE feed, based not on who you follow, but on the streams you follow. The "hit-rate" of relevant content would be way higher. On top of that, what if we removed vanity metrics, like the "Like." Instead, we give anyone the ability to private reply to a post. Contextual DM.
  3. No self-censorship: From the creator's perspective, each stream is a place where everyone opted in to a specific facet of their identity. There should be no reservations or "stage effect." It should have the feeling of writing a text to someone, but distributing it at scale. 90% of things we could share fall between <no specific context to share via text> and <too personal to broadcast to everyone I've ever met>. This network would capture everything between those two ends of the spectrum.

For the readers: a more sane feed

For the creators: a willingness to share

= L O O P ? !

We would feel a huge difference in a social media network built on a more nuanced social graph.

Money Models

Of course, if there isn't a free base layer, then it couldn't go viral, and it would never take off. No one would have ever heard of Clubhouse if it was $8/month to join.

So how can social media be monetized without selling your thoughts to shady LLCs?

It's interesting to see how services are making money by offering advanced features on top of free social networks. Look at services like Hypefury and Chirr. Some people (like myself) will pay up to $15 a month to optimize their experiences on a public social network.

Free networks can offer paid services on top of a base experience that:

  • Make the consumption experience better
  • Make the creation experience better
  • Allow creators to monetize

Digest Controls for the Reader

In almost all platforms (social & email), the creator controls the frequency of delivery. Why can't the end-readers control the frequency of a Stream?

For example, maybe someone writes a daily newsletter, but I'd prefer to only get a digest, with all 7 batched together on Sunday mornings so I can binge them. It's actually a detriment to the creator to enforce a delivery cadence on an audience. If there is a mismatch, I'm forced to either put up with it, or unsubscribe.

A paid feature could enable a user to create digests, which would include posts from streams, batched together, and delivered at an interval of their choosing.

Twitter Lists 2.0

The general idea of a Twitter list could be a paid feature on top of a free network (I'd pay for it). Lists are required on a Twitter because the main feed so often spirals out of control into an information clusterfuck.

But the base feed with Streams will already be so much more sane, that lists wouldn't be required for a good experience. They could result in a GREAT experience - one that's worth paying for.

The ability to select specific streams and aggregate them into a single list could lead to a new era of curation. Instead of a list being a filtered group of people with wide interests, it could be a hyper-specific Stream around an idea.

Not only would this hold value for the list-creator, but for the market. List creators could monetize access to Streams and get paid for their network and curation abilities. Creators could even earn royalties for being featured on a paid Stream.

Paid Streams

A monthly membership (or lump sum) would also allow creators to monetize their own Streams.

There's risk in having a single paid list (like on Substack), because it limits the number of potential viewers. But if the average person had 10 streams or so, there could be a mix between free streams (50%), $1/month streams, $10/month streams, and $20 life-time access streams.

It could be relatively easy for the average creator to make back the cost of their paid features.

Happy People

A more nuanced social graph would better reflect how connections work in real-life. The distribution potential of the Internet could supercharge (instead of drain) our relationships. This idea above is just a thought experiment, and hopefully proof that social media isn't inherently screwed.

Changing one foundational feature could result in a better UX all around: happy people, more posting, more engagement, a sane feed, and a cleaner monetization model.

🌐

| Twitter - NFTs - michael@michaeldean.site - 2021 |