Background Social
Theory #42: on apps that let you tell me what you're up to without telling me what you're up to.
Most social products ask us to create original content, i.e. user-generated content, or UGC. On Twitter, you post just for Twitter. On Instagram, you post pics that you likely took with Instagram in mind. This is true in private and semi-private too. You take Snaps just for Snap. You write on Reddit just for the subreddit.
Background social products are different.
Background social apps take content that we already create elsewhere and use it to help connect us with friends and create fun moments along the way. This secondary UGC can be one of two kinds:
Content that we actively create for another purpose or platform, i.e. what we traditionally think of as content — photos, videos, blogposts, messages, links.
Content that we passively generate as we go about our lives, i.e. information or data that’s content-like — location data, music listening history, video watch history, stock portfolio performance, screen time itself.
Background social apps leverage this content and then try to add a social layer and a uniquely fun experience on top. Background social apps are novel wrappers around content that already exists yet isn’t already utilized in a social context.
Background Social Apps
This is best illustrated with an example that most of us understand well.
Find My
FindMy is an example of a the second type of background social apps — an app that utilizes content-like data that we passively generate as we go about our lives.
Everyone with a smart phone is constantly creating location data, and Apple’s native app, FindMy makes it easy to share. This “content” can tell your friends & family where you are, let them know you’re safe, or provide a fun status update.
“Friends now, sometimes unwittingly yet obsessively, check one another’s locations and bypass whole conversations — about where somebody is, what they are doing or how their days are going — when socializing” … “You aren’t actively choosing to do something as you reach a certain location because you’re constantly sharing your location,” As a result, “there’s an intimacy that’s intertwined with that act.”
“My favorite thing to do is to award one of my friends Dot of the Week for being the most interesting dot to watch that week,” she said. One of her friends recently won Dot of the Week two weeks in a row for her trip to Europe. “I was following her little dot in Rome, and I was so happy every day to see that she was somewhere else having fun!”
Snapchat’s Snap Maps and other apps like the family-geared Life360 are similar. And before it was shut down by Snapchat, Zenly was famous for being a fresh location-sharing app, used by as many as 40 million people monthly at its peak.
Airbuds
An app that’s caught the eye of those who follow “consumer social” is Airbuds. The app lets you see what music your friends are listening to on Spotify, and it’s been consistent ranked ~ #10 on the App Store’s Music charts.1
The similarity between Airbuds and apps like FindMy is in how they get their content. Listening to music is largely a passive, background activity. So Airbuds uses your Spotify listening history to let you hear what music is streaming into your friends’ ears throughout the day, learn their tastes, and discover new music. It also has weekly recaps (like a weekly Spotify Wrapped) to keep you checking in.
![](https://substackcdn.com/image/fetch/w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fb89d9660-1c0f-4e47-9a16-f1858b9f87f4_5861x2159.png)
Airbuds uses the traditional mobile app (notifications and all), but it also has a home screen widget — an entry point recently popularized by Apps like Locket. This widget, according to Airbuds founder Gilles Poupardin himself, is key:
"We observe a high rate of interaction with the content that is pushed on the widget, especially the tune twins when you listen to the same song as a friend. The tap rate on the content pushed in the widget is much better than the opening rate of notifications."
The widget no doubt helps keep user engagement high, improving one of the hardest things to crack — user retention (i.e. how many people who sign up today actually keep coming back and using the app). That said, I believe that there’s an equally big advantage that isn’t discussed enough — the background advantage.
Note: We don’t have mature examples of the first kind of background social apps yet either — those that utilize content that we’ve actively created for another purpose or platform — but social apps that help you share photos you’ve already taken are the ones that’ve been tried the most. Google Photos & Facebook resurface old photos for you as new content in the form of “memories,” though these are often single-player experiences. Retro is a newer app that’s focusing more on curating and sharing than creating photos.
The Background Advantage
So how exactly do background social products amplify activation & retention?
I’ve touched on this indirectly, but I’ll make it direct using Airbuds as an example:
They reduce the content cold start problem. As soon as people connect their Spotify accounts, Airbuds can fill their “profile” + “feed” with listening data.
They require no new effort to “create.” You don’t need to be a “creator” on the app. In fact, effectively nobody is a creator, everyone is largely just a consumer (or at most a curator or participant).
They can stay relevant as long as the background activity stays relevant. Listening to streaming music isn’t a fad; it’s here to stay for a long time.2
Retention is king (and we want it flat)
In a recent post on Twitter / X, Nikita Bier pseudo-anonymously touted the Airbuds app’s stickiness, sharing a notable retention chart:3
“If you're active in the social app industry, once per decade you'll stumble across what is known as a Black Swan retention chart. When you see it, don't ask questions and just find a way to be involved in whatever capacity you can. Today was one of those days.”
In case you’re unfamiliar, this chart shows a retention curve, notably a legendary “flattened curve.” You measure retention by looking at how many users in a new cohort return n days later, or n-day retention, then plotting it by time.
D30 (or day 30) retention is king: 20% is ok, 25% is good, and 30% is great.4
A flattened curve just means that retention is starting to stabilize, ideally by d30.
If you look at about day 30 on the Airbuds retention chart, you can see what are considered great metrics (nearly unheard of for social apps):
of the users who have an account, ~30% return on day 30.
of users who have 5+ friends on the app, that jumps to ~50%.
if users have more than 10 friends on the app, it looks closer to 60%.
It’s incredibly hard to get enough things right that people want to use your app, let alone stick around, so needless to say this is impressive. Yet, assuming this is Airbuds’ data, it’s not shocking given the built-in advantages from leveraging both the home screen widget and the mechanics of background social apps.5
Like most new social apps, we don’t know if Airbuds will work long-term, even with these advantages. The two best paths I see for them are to build-in social game mechanics or persist as a lightweight social utility that helps discover your friends’ taste and new music. (If I had to bet, I’d bet it has the lifecycle of a game).
And the perennial question is why don’t the giants themselves (e.g. Spotify) kill these apps and take the opportunity for themselves? The simplest answer: these add-ons don’t offer enough value to them (yet). But for grassroots builders, the dream of creating novel social layers is big enough to try and, hopefully, endure.
Why now & What next
Background social apps are a great fit for the current moment in time because:
People are saturated with “new content” and discovery and they’re tired of making content for new platforms. Instead, people are craving socially-relevant curation of content and semi-private or private social interactions.
People’s willingness to share information about themselves is high, especially among younger generations who seem more willing than ever. In 2022, “69% of Gen Z and 77% of millennials said they activated location-sharing features at least sometimes.” Beyond location sharing, the past 20 years of broadcast social media has made more people comfortable sharing publicly. So in comparison, it feels near harmless to share data privately with friends.
So what type of “content” (secondary / passive UGC) is best for background social?
It’s generated frequently and on a regular basis.
It’s valuable, entertaining, or both (and bonus points if it’s actionable).
There’s potential to learn or discover something new from it.
There’s minimal stigma or fear around sharing it.
It takes very little time to engage with it as “content.”
For “content-like data,” location-sharing has been tried a thousand ways. Personal finance and investing is another appealing category (e.g. sharing credit card purchases, Venmo transactions, seeing your friends’ stock portfolios). There are many applications in health & fitness (e.g. steps, sleep) and work & productivity (e.g. calendars, code commits) too. And these kinds of content aren’t just valuable for traditional “social” apps, but also for social gaming.
The more connected we are to the internet, the more we generate content and data about everything we do, creating more potential for unique single-player and social experiences on top. Background social apps can create experiences that help you make new friends, keep the old, find romantic partners, and more — all by taking what happens in the background and bringing it to the forefront.
The app has charted at least as high as #2 on the Apple App Store Music category chart. The app started with just a Spotify integration and is seeking to expand to more.
Certainly you can point out the disadvantages for these types of apps when they depend on external or third-party app owned data. That said, there’s usually an opportunity to use this dependency as an entry point into a user / consumer’s life and then expand or modify the offerings, usually creating native content formats later on.
As best I can tell, this is a reference to Airbuds (but let me know if you find otherwise!).
Benchmarks courtesy of A16Z, as referenced in the essay.
The content cold start problem is solved in user targeting and activation itself — recruit users who have spotify and listen to music often (a lot of young people) and get them to connect their spotify account and you’ve pre-populated “content.” After that, no one needs to create in-app; all they need to do is keep using Spotify. Now all the team has to do is make the product and social experience compelling enough to use it, stick around, and of course share with friends. (I say “all the team has to do” in jest, because it’s still a very hard thing to do, but made much easier by these advantages.)
The Airbuds example begs the question: where is obviously valuable passive content being generated but ignored by a dominant player?
This kind of social discovery based on listening habits was my favorite feature of Rdio back in the day. I’ve deeply missed it and never understood why Spotify never bothered to build it.
Airbuds nailed the execution, and kudos. It’s a good example of how you don’t need to dig deep to find valuable passive content, but you still need to ace productization.
I didn’t want to include tens of examples in the piece but i’m sure many exist, any that come to mind? I’m going to start adding them here.
Also am mulling on the idea that there are a lot of single-player products that are wrappers around this kind of background content / secondary UGC too.
Example: any summarization / analytics tool that you give access to your data. Mint used to be a popular one for this in fintech, helping categorize and analyze your spending, where the data is created by you using your credit card!