Designing Social Products

Why Snapchat is winning.


Over the Christmas holiday I read multiple reports of Snapchat’s recent financing and when I headed on over to a my usual internet watering holes to learn what everyone thought I was tickled to read a many comments to the tune of:

Good on them, but I still don’t get Snapchat

As someone who [thinks he] gets Snapchat, I though it would be an opportune time to write down my personal thoughts on the subject and what makes social products successful in general. All opinions — and theories—are my own.

What is “social”?

Let’s lay some groundwork so we’re all talking about the same thing:

A product is social if it’s primary use is in communicating with other humans.

Uber is not social, Airbnb is not social, your local newspaper is not social. Facebook is social, MySpace is (was?) social, AOL instant messenger was social, Snapchat is social.

Now that this is out of the way let me present

A Grand Unified Theory of Social Apps

aka GUToSA

All successful social products supersede their predecessors through two fundamental design principles:

  1. Minimizing social liability
  2. Enhancing expressive power

As a disclaimer, designing products that fit this mold is a significant challenge. And even if you designed such a product, distribution is still very hard. Getting traction tends to be more about who starts to use your product first (in addition to having a good product)—but that’s a topic for another time. If I were an active investor, however, I would be very much looking for products that fit this mold.

With GUToSA in mind, let’s back up a little bit and take a tour of some of the internet’s social products of yore:

Image for post
Image for post

If you weren’t part of the geo and age demographic that used AIM (maybe you used ICQ or MSN), the appeal here versus what came before it was simple: free, instant, point to point messaging with you and your friends.

Looking back, it’s pretty amazing all the things they got right:

  • Presence
  • Group chats
  • Self-moderation
  • Profiles
  • Status messages
  • Attachments
  • Delivery status
  • Emoji

Clearly there were many different ways to expressive yourself on AIM and with this in mind I’d like to introduce the concept of Expressive Power:

Image for post
Image for post

Because AIM was approximately instant, it enabled the round trip time for communication to be much shorter than ever before, resulting in a higher expressive power than e-mail or other less synchronous forms of communication. “Round-Trip Communication Time” includes the time required for a pair of humans to use the associated product (read: it’s not just how fast your internet connection is). “SNR” stands for signal to noise ratio such that sending 2 hours of TV static is not considered “expressive” since there isn’t any signal. You can approximate the numerator here by how much information the receiving end actually consumes. Good social products will both increase the numerator and decrease the denominator.

One additional thing is that if you factor in onboarding and discovery into round-trip communication time, this concept also accounts for nascent networks that have no traction — getting all the people you want online and connected to you is a slow process. To make an analogy, a language that only you speak isn’t going to be very useful in letting you express yourself.

Beyond AIM’s expressiveness, during its peak AIM was a tour de force of social proof. If you weren’t on AIM you missed out on a lot of gossip, flirting, and other important teenage moments in high school. In other words, not being on AIM was a liability.

1. the state of being responsible for something
2. a person or thing whose presence or behavior is likely to cause embarrassment or put one at a disadvantage.

Many would call this as a “Network Effect,” but I think that’s a layer too high and actually obscures a more primal instinct. People join networks because they don’t want to be out of the loop and at a disadvantage.

MySpace is a decidedly different product than AIM but a social one nevertheless. An identity-backed, friend-linked Geocities if you will, MySpace enabled it’s users to communicate through wall posts (they weren’t called that back then) and share their photos in their own unique way: dressed up in abominable GIFs and CSS.

The key differentiator of the network approach (and I know MySpace wasn’t first here) is that with a more persistent expressive canvas, you can share your selfies with many many more people without talking to them individually — it cuts the communication round trip approximately in half and enables many more bits with many more people.

I don’t remember much about the privacy controls in MySpace, other than it was mostly public, but some people hid their profile by default. Why did they hide them? Likely because it was a liability to have them in a place where some creepy stalker could find them and do who knows what with them, which leads us to:

When I started college in 2006 I used some of the early iterations of Facebook. To recap: you had to actually have a real university e-mail address to join and everything was shared (as far as I remember) within a network specific to your school. This was great: it helped you find other people you want to talk to. If you factor in “finding another human being” into the communication round trip time, Facebook provides pretty significant savings. Additionally, the closed network nature prevented unflattering frat party photos from escaping into the wrong hands. It minimized the social liability of a future recruiting coordinator deciding you weren’t fit for a position because of some poor choices you made in college.

As Facebook grew as a company, it wanted to encourage sharing and suddenly defaulted a bunch of information to being public. This, in my opinion, was the original sin of Facebook for which is now pays an immeasurable price for, Facebook’s apple in the Garden of Eden, if you will.

Why was this so damning? If there is any possibility over-sharing, usage of a product suddenly becomes a huge liability. Anthony Weiner learned this the hard way. Email, for comparison, doesn’t have this liability: there simply is no way to cc: the entire world.

Mark Zuckerberg has often touted the mantra of making the world “more open and connected,” and I sympathize with the intent, but a few things stand in the way, practically speaking:

  1. You can’t anticipate what will be embarrassing or inappropriate years down the road. Have you ever looked at an old diary (old Xanga in my case) and just cringed endlessly? I have. For teens, it’s more urgent: by posting this am I going to look like a loser amongst my friends tomorrow?
  2. Harassment is real. It may be hard for us straight, white males to relate to this from our place of privilege, but the internet can be a horrible place and over-sharing can act as bait for the scum of our society to say and do despicable things.
  3. When you share content online, you are implicitly starting one half of a conversation. For some things, sharing to a wide audience isn’t worth the implicit obligation to keep up with the unwelcome conversations that come of sharing.

I want to call out the last point here and bring back the definition of liability, but with a different emphasis:

1. the state of being responsible for something
2. a person or thing whose presence or behavior is likely to cause embarrassment or put one at a disadvantage.

Having a presence on a product means you are responsible for it and all of it’s consequences, as a product designer you want to minimize the bad ones.

It’s very hard to put the Genie back in the bottle once your users are subconsciously aware of the possibility that something they post may accidentally end up public or even just in front of your crazy relative who is bound to embarrass you by bringing up his crazy political beliefs in your Facebook comments.

Facebook is trying to obtain other bottled genies (to grossly extend a metaphor) through Creative Labs, but there’s quite a bit they could do in their main product: privacy settings are the last thing you choose before you post—don’t get too eager and press post too soon or you might end up making a big mistake.

Image for post
Image for post
(It was pointed out to me that Facebook’s mobile flow is slightly improved from this layout)

Snapchat is a product that puzzles grown adults across the world, but when seen through GUToSA suddenly becomes much clearer.

First, we have a generational shift. Kids hate to share interests, or really anything, with their parents and thus if you are a teenager and you use Facebook, not only is there the risk you actually overshare to your parents but there is the liability that your use of Facebook makes you uncool amongst your friends.

Second, we have a mode of communication which is primarily photo based, but can be drawn over or annotated: more bits of information than a normal photo, compressed heavily so that it sends ridiculously fast.

Finally, the ephemeral element: it minimizes the risk that your embarrassing selfie ends up in the wrong hands, and there are no 10-year-later cringey moments at your younger self. Of course someone could take a photo of their phone, but this is pretty hard to work around — a fair assumption is if someone can see it with their eyes, someone can probably “see” it with a camera too.

Snapchat is experimenting with stories which are wider distribution but a) aren’t fully public and b) moderated to be appropriate if the distribution might be beyond your friends. They’re pushing the edges and if I were them I would tread very carefully.

But there’s more!

I could expand upon this post endlessly, but here are a few other notable examples:

  • Instagram: Don’t embarrass yourself by posting ugly photos, with filters everything looks wonderful!
  • Twitter: One-to-millions communication enables a lot of expressive power and with everything defaulted public, expectations are set appropriately.
  • Secret: Your statements are anonymous, it’s much harder to embarrass yourself (at the cost of expressive power, because efficient communication is enabled by identity-specific context).
  • Google+: You can pretty much do all the same things on Facebook, but now we’ve made it so hard to figure out who you’re sharing to that the overhead of sharing drastically reduces the expressive power.

I would be remiss to point out that both AIM and MySpace are effectively dead. AIM was left to starve as AOL withered to a ghost of it’s former self and MySpace died because it was simply superseded by Facebook in expressive power and minimized social liability.


If there’s anything to learn from this hopefully it is this: while you push so hard to encourage frictionless sharing of rich content, always remember to place your mind in the user’s head and ask “am I going to regret using this?”

Minimize social liability, enhance expressive power.

Go forth and build great products.

Written by

code + design, entrepreneur in hiding. cofounded bubbli (acquired by Dropbox). previously made yelp monocle (for iPhone and Android). stanford alu

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store