May 202014

My friend Warren Myers posted a question about the viability of a subscription-based social network not too long ago, and the concept kind of stuck with me. I’ve thought about the idea off and on since I’ve read it and initially commented on it, and it’s led to the point where I’m thinking about the concept less as a “Google+/Twitter/Facebook with a monthly fee” but more of a thought experiment of what if I were to design an all-new social network from the ground up, what would it look like? Now, all the social networks we know and use exist, and I’m not going to pretend we’ve never had such things. Also, this is all really based around how I use social networks, which is primarily content consumption, not content creation. That said, here’s where I see social networks going.

Facebook with a fee?

I want to take a second right here and just abandon 1 of the initial assumptions, mainly the idea of people paying a monthly premium for a social network. There’s not going to be a lot of demand for something like that outside of maybe a few bleeding-edge, tries all the newest things, adopters. Existing social networks have enough critical mass and usage that you’d have to offer something mind-blowing to even get them to consider it for just 1 whole second. As a result, while I am going to ditch the traditional ad-based business model, this theoretical social network will not charge all users a subscription.

Why bother?

Well, the incident that prompted Warren’s blog post was Facebook adjusting it’s algorithm for determining what shows up on people’s news feeds. In this instance, businesses that were posting on these networks to try to connect to customers naturally were getting filtered out, essentially forcing them to pay to sponsor their post to get any real exposure to their customers. You see, our Facebook news feeds are so full of crap, Facebook has to limit what displays to what it thinks we’ll be the most interesting. I’m pretty sure Google+ does this as well. Basically, the way we’re doing social networks now generates a very low signal to noise ratio.

So far we’ve been trying to fix this algorithmically, but maybe what’s needed are social networks that do a better job of encouraging more interesting content. Now, we’re still seeing posts from businesses in our Facebook feeds, so clearly this is working right? Well, if you’re ignoring the fact that an ever-increasing percentage of these posts are sponsored posts, sure. That’s the real issue here. People subscribe to companies to see their updates and connect with them outside of ads, and that’s being essentially eliminated. Now it’s a filtered list of stuff we’re not interested in from people we may or may not actually know in real life, a shrinking set of posts from businesses we’re actually interested in hearing from, and businesses whose updates we didn’t actually sign up to see.

All of this leads to a philosophy for essentially posting like-bait. The philosophy of determining what few natural posts from the businesses we follow is based on the number of interaction-based signals, such as likes and comments, on the posts get from the people who do get to see the post. That means everything has to be construed to generate a response from users. How do you think BuzzFeed got to be such a thing, despite being essentially the social network equivalent of eHow? Quick and easy content that generates a lot of likes and shares, which means it gets on people’s feeds. Ultimately, this like-baiting (really, upvote-baiting as it contributes to noise on things like Google+ feeds too, although they haven’t introduced sponsored posts yet).

What would a new kind of social network look like then?

So let’s recap, regular people aren’t going to pay to join a social network, and the current system of encouraging as many posts to a main feed as possible is generating so much noise that a lot of it is getting filtered, which for businesses is compounded by the ability to sponsor content. Here are the key features that I think could make a social network that isn’t ad-supported work.

All RSS, all the time

The main landing page of almost all of these social networks is a feed of some sort, so why not make those feeds RSS? (I’m sure Warren would approve) Remember, this hypothetical social network isn’t going to be making money off of ads, so there’s no impetus to compel people to use our hypothetical app over other clients in order to pay the bills. Your social network could be your RSS reader, it could be the hypothetical app, or some other third-party client. This hypothetical social network may not even need an app for long, if at all. Either way, getting updates is easy and cross-platform without any significant engineering effort. Existing blogs could very easily be migrated over with some plugins to popular open-source blogging engines. An RSS-based social network also allows this social network to be a true network, not merely an app that seeks to mirror our existing networks. Everything that’s compatible with this social network would have to publish RSS, and the world is already full of RSS readers from all sorts of places (except Google).

Dynamic visibility

With privacy being such a big deal these days, social networks have started to get better about what their visibility controls are. Google+ is probably the best in terms of making it easy to adjust what groups you’re sharing with, Facebook deserves credit for giving you the most granular control over who can see what post initially (nobody controls re-shares, and screen grabs are forever). However, it’s time to make determining your initial audience are lot easier.

I’m proposing a new feature called “dynamic visibility”. In short I should be able to define visibility based on the relationships of social groupings without making new groups. Not just “this group but not that group”, but also “people in both these groups” or “people in this group who are within X distance of me”. These are the types of things that make sense for 1 particular post, but aren’t worth dealing with whole groups, especially the location-based option. Dynamically defining your initial audience immediately makes sharing on this new social network more powerful than the existing options out there. By the way, this new social network should absolutely refer to the people you assign visibility to as your “initial audience”. It makes it much clearer that this stuff could be shared beyond whatever group you set, which might encourage less surprise when stuff goes beyond what you had typed into a little dialog box.

Required topics

Another feature of this hypothetical, new-age social network, requiring every post to have some type of topic assigned to it. Ideally this network could also suggest some best-fit tags, sort of like Google+’s automatic hashtagging feature, but the point is everything posted to this network should have something indicating what the post is about. Tagging all posts with a set of topics is probably the trickiest part to this whole endeavor, since you want the tag to be specific enough for granular filtering, but not too specific that it’s impossible to effectively filter. It’s possible, just really hard to do well. This seems like a silly feature to add, but it’s what enables the next thing our hypothetical new social network should be able to do, which is:

Consumer-side content filtering

All lot of time and focus with privacy on social networks goes into controlling who initially sees your updates, but that kind of time and effort should also go into letting users pick and choose what posts they actually want to see. When you pull updates from your friends list, you can actually choose to only receive a subset of those updates. Now you could receive development-related updates from friends A, B, and C, personal life updates from friends B, E, G, and H, political updates (if you dare) from friends A, G, and L. This new type of visibility filtering lets you get only the stuff you’re interested in without any of the other stuff. Want updates on the lives of your family members without all the re-sharing or forwarded stuff? Can do. Want updates from your (current or former) co-workers about stuff related to the type of things you do for a living without their thoughts on contemporary politics? You got it. Don’t want anybody’s updates on the NFL? Just filter it out across the board.

This filtering would be built from the tags we’re going to guarantee appear on every single post. It’s why it’s so important to get that part of things right. But if you do, this social network would give users the most power over the visibility of content out there. Social networks aren’t just about the people posting, they’re also about the people consuming the data. It’s time they got a little control over the stuff coming in on their feed. It’s a lot better than trying to algorithmically guess it or having some sort of hard-to-find volume control that applies to all posts when the reality is you just want to tune out subsets of content.

No upvoting. Period.

Another feature of this proposed social network – no upvoting, of any kind, at all, period. At this point, upvoting has basically given rise to upvote-baiting. This system of liking, +1’ing, etc. is a major driver of the noise we see on social networks. What sort of signal is it sending? You liked something. Great, but what value clicking that little button really adding to anything? Sure it’s a stroke of the original author’s e-peen, but you’re contributing nothing of use or value to your social groups, the Internet at large, or the world. If something someone posted on this network moved you to make some kind of response, formulate a thought and respond. Gather your thoughts, and craft something. Add additional reasons why that post is worth reading, offer another point of view on the topic. Agree but think the original reasons are wrong? Correct them. The point here is that everything being pushed out on the network should offer some type of value to people who see it. “You liked this” isn’t going to cut it.

Now for the making money from subscriptions part

So after making a fuss earlier about how normal people aren’t going to pay a subscription fee for a social network, how do I think this network is going to pay the bills? Simple, make businesses pay a subscription fee. Stay with me, there’s a logic to this. businesses pay a subscription to join the network, and that’s it, they’re on their own for reaching regular members. There will be no paid posts. Let me say that again, it’s an important part – there are no paid posts on this hypothetical social network. Once a business pays the subscription fee, they’re on their own to succeed on their own merits, without upvote-baiting links to force this stuff to get listed in the feeds of people of who didn’t ask for the content, and no paying to guarantee you content gets put on people’s feeds.

In fact, the no paid posts thing is gong to have to be the make-or-break point of this network. If charging a subscription isn’t going to pay the bills, then it’s just a failure. But the good news is, there’s no magical filtering algorithm that’s going to block their posts. Users still have the ability to filter out this stuff, but that’s a problem solved by engaging, relevant, and interesting content.

Still a few problems left to solve

As far as idle ponderings go, this seems like it could be a good idea, but  there’s a few parts that I still haven’t conceptually figured out. At the very least, there needs to be an idea of how these are going to work before a version 1.0 of this social network goes out (even if they aren’t part of the 1.0 release itself).

Sharing of other people’s posts

For starters, I’m not sure how best to handle the idea of re-sharing someone’s post. Ideally, a shared post should cite the original source (it’s common decency after all), and who put it on your feeds.  Shared posts should be clearly delimited from original content. I’m not sure how best this could be implemented in an RSS-based social network, but I think a service like Repost may be the way to go on this one, but re-sharing is a pretty important feature for social networks and something that needs to be figured out. No matter what, sharing can not violate the consumer-side filtering users have in place. Destroying that filtering via re-sharing essentially breaks the social network.

Overly ambitious tagging

Another possible problem is overly ambitious tagging. It’s the kind of thing you see on job board postings and IGN – somebody marks up a posting with just about every relevant, semi-relevant, and somethings not-actually-relevant tag they can think of, and you start seeing stuff that just isn’t that relevant. Ideally, this new social network would have a system that clears the irrelevant tags instead of just blatantly flagging a post as “spam” (although that will probably need to be an option as well, sadly). Odds are, a lot of this content is of interest to some people, so it’s mostly a matter of getting to them and no one else. This is going to a problem that needs to have some sort of process of solving, even if that process still needs work, just to try to keep the network from going under due to spam.

APIs and other social network integration are your friends

The biggest challenge this new social network will have will be getting people to use it instead the existing stuff everyone else is already using. I think the best solution to that problem is an “if you can’t beat them, join them” philosophy. Nobody wants to move from a social network where they’re already set up with all of their existing friends and family to something only a small subset (at best) of their real-life relationships are on. The best way to address that is to integrate this new network with the existing networks out there, at least as best as you can. Some social networks limit read access (like Twitter), so completely baking in other social networks into this may be a tough sell. However, posting to the stuff that’s already out there is a good initial feature that should help lower the barrier to starting out on this new-fangled social network thing.


I think there’s a lot of potential for a non-ad-based social network. Abandoning the idea that you need paid content in user’s feeds and approaching social networks like you were building a fresh one from scratch without trying to match the stuff that’s already out there feature-for-feature leads to some ideas that seem pretty compelling. That said, there’s still a few serious kinks that need to be ironed out before a social network like this can be ready for prime-time. All of these issues seem very solvable, which makes me believe this hypothetical social network has the potential to actually be “the next big social network” for a change.

 Posted by at 12:16 AM