Just over 1y ago, Fred Wilson of Union Square Ventures popularized the term ‘App Constellation’, which he defined as ‘…a collection of mobile apps that share a single login and have app to app linking built in.’
At that time, we wanted to dive deeper into the trend, and understand how the strategy was playing out with Facebook, Foursquare, Dropbox, and a few other mobile apps. A collection of Quibb members contributed their perspective on this new, lauded strategy, and made predictions about how the strategy would pan out over time. No one was overly optimistic.
Evolution of mobile app distribution hasn’t become any easier to navigate these past 12months, and competition has only increased. When I asked Toufeeq Hussain (Senior Product Manager at Storm8) if the App Constellation strategy had failed, he noted:
“Maybe failure is too harsh a word as I feel the mobile space is now so well established that the success metric needs to shift from being in the Top-100 to being in the Top-1000.”
Facebook has managed some great results in spinning-out Messenger. Notably, since the beginning of 2015, the iOS version of the app has never slipped below 5th place in the US in either the Social Networking nor Overall category. Impressive.
You could argue that a couple of their other constellation products (Instagram and Whatsapp) that were acquired via acquisitions are also doing very well, while those created internally (i.e. Paper, Slingshot, Rooms, all launched out of Facebook Creative Labs) have more or less failed.
Dropbox’s Carousel - another internally created product - has also continued to struggle over the past year:
It has been hovering around 50th place in the US Photo and Video category of the iOS App Store, and well past top 500th position Overall for most of 2015. Meanwhile, there are rumors of Dropbox continuing to push in this direction with pending Google Docs-style collaborative word processor product.
One year after our initial post, we again chatted to mobile experts from inside the Quibb network to better understand — Does the app constellation strategy still make sense? What have we, as an industry, oberserved and learned about this strategy over these past 12 months?
Casey Winters (Growth Lead at Pinterest)
I don’t think much has changed in the year since we last talked about app constellations. Most companies that have tried them have failed, and the successes seem to crowd around these specific strategies:
Acquire already successful apps and cross-promote
Unbundle a part of existing app that is already successful and go for niche or broad appeal
Cross-promote new apps heavily and recognize you’ll have wins and losses
Facebook and Google are clearly the most mature in their strategy. They make bets on new apps of which they know they’ll have a certain success rate (Rooms, Moments, Keep, Slingshot, Paper, Riff, Field Trip). They also successfully unbundled popular pieces of larger apps (Messenger and Groups from the Facebook app, Sheets and Docs from Google Drive, Hangouts and Photos from Google+, Inbox from Gmail). They do this for two purposes: to make popular features take up another spot on the charts (Messenger, Photos) and to make more performant apps for niche features that are important to certain people (Groups, Pages, Sheets).
Evernote is the only other company of note with a similar success ratio. Almost all other companies have had no success with app constellations (Dropbox and Foursquare, notably). Microsoft, while late to the game, is following Facebook and Google’s strategy of acquiring great apps (Sunrise, Accompli) and launching their own successful desktop apps on all platforms.
Twitter focuses on cross-promoting app launches within their main app. This was successful for Vine and Periscope, and not so successful for Music. Instagram has had one success (Layouts) and one failure (Hyperlapse).
It will be interesting to see what Uber does here. As Uber expands outside of transportation of people to things (via Eats and Rush), they will likely be pressured to unbundle those apps. They are doing the right thing right now which is using the install base of the main app to gain popularity for these features. Then, if those features succeed, they can unbundle them like Messenger.
Andrea Friedenson, Consultant at Halfshark Industries
One year later, app constellations are still an ideal growth strategy for our current silo’d app ecosystem. If you have a big enough audience, a differentiated use case for the non-core apps, and are willing to take the risk of pushing users out of your core app and onto a new property, you can not only acquire a huge audience for little cash outlay, but you can also limit the ability of your competitors to get those organic installs. Facebook has done the best job capitalizing on app constellations. By breaking out Messenger functionality from its core app, and Layout from Instagram, Facebook has claimed 4 of the top 10 spots in the App Store and 3/10 in Google Play. They just dominate app discovery.
But I think App Constellations are a fundamentally short-term play. In a few years, we won’t need as many affordances to make our devices do what we want. Apple and Google have already made strides towards turning apps into service layers for their hardware. Prime real estate in the App Store becomes a lot less valuable in a world where we need fewer apps. I’m sure that Facebook has thought about this. I’m interested in seeing what their plans are for avoiding a declining share of attention in a post-app world.
Alex Schiff, Product Manager at Occipital
Last summer, I made two main claims about the whole unbundling/constellation strategy in the wake of the launches of Swarm, Paper, Carousel, and a few others:
1) Unbundling into “constellations” vs. single apps rarely works. Facebook Messenger is one of the only successful examples.
2) The only time it does work is when two use cases compliment, not cannibalize each other. This is one reason Messenger took off and Paper did not.
One year later, I think those arguments have generally held up. However, I think in many cases, we were thinking of “success” as one-dimensional: Did the company succeed at getting multiple apps on your phone or not? With the benefit of hindsight, you can see that more often than not this wasn’t really the goal.
For example, Paper was not an attempt to get more real estate on your phone — it was a UI experiment.
This card-like UI interaction is what you see when you’re looking at profiles of suggested friends from the friend request page, and it’s straight out of Paper. You’re starting to see more and more Paper-esque UI paradigms permeate the Facebook mobile app, and I think you’ll see that continue.
Somewhat different is the case of Foursquare and Swarm. When we commented on the strategy last summer, Swarm was chalked up as a failed unbundle because Swarm itself didn’t take off. And it still hasn’t:
However, Swarm wasn’t really about Swarm — it was about saving Foursquare. Foursquare needed to escape the legacy of check-ins to go more mainstream, but just axing the feature outright would have caused even more revolt. Swarm allowed them to gracefully remove check-ins, and Foursquare has been steadily rising in the ranks since last fall (when the dust settled from the launch of Swarm).
One interesting new strategy we’re seeing is big companies simply buying constellations — i.e., purchasing ready-made apps (sometimes with existing user bases) that they believe can be successful as part of a product suite. The most notorious example as of late is Microsoft, who has bought their way back into relevance in the consumer productivity space by acquiring Wunderlist, Sunrise and Acompli within 7 months. They’re not the only ones, though — Twitter has been quietly building a constellation via M&A for years (Vine, Periscope), as has Evernote (Skitch, Penultimate).
If anything, I think we’ll start to see more companies entirely outside of tech “buy constellations” as a way of getting in the game. There are plenty of companies with a ton of cash and declining relevance.
Adam Sigel, Product Manager at InsightSquared
A year ago, we talked about how platforms were breaking up into constellations of apps. Each app was targeted at a different use case. Facebook spun off Messenger and launched a slew of standalone apps around different forms of social consumption and creation. Foursquare separated its recommendation engine from its social gamification platform. And so on. People griped at the need for more apps on their homescreens, but optimists saw the promise of a new wave of apps that could be optimized around a more targeted use case.
Today, I see something similar happening again on a smaller scale. Apps are being broken out into separate faces to take advantage of the different form factors that now exist, specifically wearables. Translating an iPhone app to an iPad app by and large gave you the same experience with perhaps more breathing room or fewer options obscured behind a menu. Shrinking that same app down into 390x312 pixels has forced people to think about the core value they provide to users. It’s almost like rediscovering your MVP. Just as related apps must plug back into a single platform, faces must work together within a single app. Apple Maps does this quite well. Once you tell your phone where you’re going, the job of getting you there is handed off to the watch, which nudges you when it’s time to make a turn.
Each app is like its own tiny ecosytem that plugs into a broader multi-app universe. Power users no longer have to choose just one social, payment, or location app. It’s getting easier for them all to play together thanks to Extensions in iOS 8 and the deep linking that’s coming in iOS 9. Conversely, a well-executed ecosystem creates a moat of ever present usability for mainstream users. I love collaborative playlists in Spotify, but I love having offline Apple Music playlists on my watch even more.
I’m excited to see what developers do with watchOS 2 when they start getting access to more of the hardware in Apple Watch. We may have to put up with a few more cycles of cluttered home screens — and perhaps some very ambitious watch complications — but I’m still optimistic that we’ll end up with better experiences as companies learn how to engage users with the right face/app of their platform.
The App Constellation strategy is still going strong — but only really paying off for those with massive, highly engaged audiences and clear differentiation between apps. In those cases, it’s interesting to note that the ‘product acquisition’ approach seems to work best, while in-house spin-offs are less likely to fail outright.
The recognition by these companies that they’ll eventually encounter the Innovators Dilemma is amazingly astute for such young companies. It’s a very powerful statement to their core DNA — they’re willing to keep pushing this strategy forward, even if that means many failures along the way and potentially cannibalizing their core product.