In my dual roles as a Venture Partner at SoftTech VC and Co-Founder of Bionic Panda Games, I’ve met with a bunch of companies and teams who are focusing on trying to solve the problem of mobile application discovery. Simply put, the problem (as articulated by people trying to solve it) is that there are many more interesting applications out there than the typical consumer can find on his or her own. If you look at the number of apps in the Apple or Android markets, that’s probably true – no consumer can be fully aware of everything out there that could be of interest.
By and large, the teams I’ve met are incredibly smart, clever teams of people who I think will impact the world of mobile applications in positive ways. But after having met with a handful of really great folks working in this space, I’m becoming convinced that application discovery is a bigger problem for those applications looking to get discovered than it is for consumers looking for useful or entertaining applications. Three things I’d like to throw out there for discussion:
1. There is a lot of mobile application discovery that happens face-to-face in the real world and will probably never show up in your application analytics system. If you’ve gone out in a major city lately, I’m sure you’ve seen this happen. Someone pulls out a phone and shows off an application. Someone else at the table says, “Wow, that’s a cool application – I’m going to get it right now.” And that person starts downloading the application. That’s the beauty of the the combination of app stores, smartphones, and reasonably good data connections – if you find a new application you want, you can get it nearly instantly. No more waiting until you get home to have a friend invite you or send you an email. Just grab the app in context and grow. I’m not sure that there’s an opportunity for an app to help facilitate that face-to-face communication.
To most analytics systems, these installs will look largely organic (someone just installed your app and it didn’t come from a trackable link of any sort) even though they aren’t. And, as a developer, it’s kind of hard to know how often this is happening for any given application. It should go without saying, but I’ll say it anyway – lots of people will instantly install applications that their friends are using when this discovery happens face-to-face as your friend can often tell you (in his or her own words) why he or she likes the application and how he or she uses it. That counts for a lot for most people.
2. Most consumers are not actively looking for new applications to download, but will download them in order to get something else or based on the recommendation of a friend. I just don’t think most people sit around thinking about applications to discover. Most people are busy doing other things – reading, playing with the applications they already have, chatting with friends, etc. However, as I mentioned above, I do think that there is value in knowing what applications your social graph is using. For example, if you’re trying to make a decision on installing an app with strong network effects (any social networking or communications application, for example), knowing what your friends use has real value. For example, if most of your friends have standardized on using a particular location-sharing, photo-sharing, social networking, or other social application, knowing that can and probably should influence your choice of application to use.
The other use case is the incentivized install use case. If people are using an application and can get a reward in that application for downloading or installing someone else’s application, it shouldn’t be surprising that they are willing to do so – installing an application doesn’t really cost most users anything (other than the space on the device and the time required to install it) and the act of installing an application is a known behavior.
3. Open question – will consumers fire up an application to help them find other applications? One question I have is around how application discovery is delivered as a product. A number of the products in the space today involve opening or installing an application in order to discover additional applications to use. I am not sure this is the model that consumers want, but it’s still too early to tell.
Despite everything I’ve said above, application discovery is a problem for application developers. It’s really just a subset of the marketing challenge we all face in standing out in a sea of tens of thousands of applications that exist in app stores – being discovered is a good way to grow an application and absolutely essential in the growth plans of many companies.
As always, comments are welcome. And if you like this you can always follow me on Twitter.
Good article Charles. Most people discover apps face-to-face from friends, I see it all the time. Apps say a lot about us so we tend to value apps from people we trust, etc. If your a cool Ruby developer, gamer like me, then I’ll probably trust your app selection. I also agree that people are not actively looking for apps, it’s not a big need. Apps are somewhat disposable, how many do you have installed but never use? But the question for developers is how to properly monetize the relationship, how to reach more users without giving away too much? How do developers sell/promote software on the web? SEO positioning, building a kick-ass user experience, word of mouth and PR. Should it be the same for mobile apps?u00a0
I think there is a real challenge for most developers in terms of getting found outside of the context of app store rankings. It is a big opportunity for someone to solve.u00a0
#1: QR-codes or something similar (http://bu.mp/, IM, email) would be feature to add to apps. This is the essense of virality which is no longer such an unknown metric, as you well know.n#2: Yep. The only reason I search is because I’m a pro. Your ‘network effect value’ warrants being part of your bold’ed bulletpoint text – it’s key to many apps’ successu00a0for many reasons (Word/Excel/PowerPoint for file format, Facebook obvious, Loopt/Foursquare/…)n#3: I would, but see #2. If Android is truly open, then a good recommendation system that used something like collaborative filtering could become a new standard. Extracting revenue could be a simple CPI?nnLove the posts, Charles!
All good observations.u00a0 I agree this is problem for the developer and not the end user.u00a0 A regular guy can be very happy installing and living within only the top 25.nnBut as a mobile developer, how do you solve the problem of ‘discoverability’ in Apple App Store? A developer can’t buy their way to the top like you *theoretically* can on FB.u00a0 Must a new developer rely on Tapjoy/Flurry and their kind?n
App Discovery is a business problem for a company. It is one of the many alternatives to the advertising channel for businesses.
Charlesnntake your iphone. Look at it closely. now count the number of apps you have vs the number of apps you are really using. The problem you are observing live illustrates why discovery is both an end user problem and a developer problem.u00a0nnUsers can’t get satisfaction with the app stores because they are built identically for everyone (i see what you see), because they are e-commerce services first (that’s why paid comes before free) and because the amount of metadata available to make a choice is just not enough which creates frustration and pushes out of the app store in search for other discovery solutionsnnFace to Face is only one of them. An important one. But users are clearly indicating more and more they want guides to help them with their choice. The same happened exactly with softwares (download.com, softpedia,..), CDroms, Ringtones, music (last.fm,..) movies (flixster, …) and now appsnnApps the most personal way to customize your mobile device and users are thirsty for relevant personal ways to discover those items. This is why see growth for app related content site, this is why you start to observe more and more paper magazines dedicated to apps, This is why you observe the birth or mobile discovery services.nnThere is a clear cut need from users for better experience in app discovery. Most people don’t realize it at first. They start to download apps and after a while they observe they own more apps than they use. They then realize the discovery process lead them to make wrong decisions oru00a0inaccurateu00a0decisions. And this is precisely when they start to look for new discovery solutions. Solutions like mobile discovery services that talk first to consumers. Some of them are plain copycats and crap guides. But some of them are really good. Appsfire.com, the company i co-created – recently closed a series A- u00a0is doubling audience every quarter (with over 2.6m installs)nnYou are right in another point you’re making: discovery is not a 1 way solution. People discover apps based on different needs. Search is only one of them: not the most critical. And most people don’t want to invest time in discovery. they are too busy for that. This is why passive discovery is also important. Some people want only deals, some people want only apps for their city or their kids. The app store and many app discovery apps fail in providing that. App discovery is a multi modal end user process. Face to Face/Social graph is only the tipping point of the issue (btw try appsfire to discover the app of your friends….)nnnIt is correct this is ALSO a developer problem: Developers needs accountable marketing channels for user acquisition. But you can’t adress it right if you don’t understand first what users experiment during the discovery process. Sure you can build an incentivized system: who does not like incentives. But they are not right because they create a biais and wrong popularity signal and eventually low quality download. this is why Apple has decided to hit them right in the face.nnThe only way you can help developers in getting more discovered is by building a promotion platform that will serve them and that will also serve the users at the same time. And the only way you can serve user’s interest right is by building a consumer service. It is just not enough. But this is the premise. At Appsfire everything that has come for app discovery and monetization and will come in the next semester could not have existed if we started as an platform only company.nnLet me give you an example: App metadata display. People underestimate how important the layout of an app description should be to generate conversion to download. There is no way you can guess how efficient this can be until you have designed it, iterated it on it and measured it directly with consumers. It is impossible to create a good ad format for apps until you have built a relation with consumers.u00a0nnAll the companies that stop at a consumer service only are indeed not going far enough.u00a0nnBut don’tu00a0underestimateu00a0the thirst users have for an actually good discovery service. This was done successfully for music, for movies, for games, for software. And this will be done for apps.u00a0nnPeople want to discover apps they will use and this can only happen if you understand the user. And by doing this right you will help developers getting quality downloads. Rushing for top rankings is good but a blind goal. The market starts to understand that.nnFor the bigger picture: App discovery is only a subset of app marketing. And understanding the users is critical to be able to assist developers along their needs.u00a0nnOuriel Ohayonnappsfire
Wow, that’s a really thorough and complete answer – I think you’ve captured the issue correctly here. Thanks for the comment – it’s a post in and of itself.
Charles, it’s not a post. it s a startup :)rnrnps: say hello to jeff for me. Both Yann, my cofounder and myself know himrnquite well
That’s true – you all are taking a more holistic approach than the app-onlyrnfolks. Marketing and distribution is really important for all of us.
wait for our next product to come out and you ll precisely understand why i meant. thinking discovery as a distributed consumer platform is the only way to do it right.
Very interesting discussion, thanks Charles for your post and Ouriel for your thorough reply. In our case, we have a different scenario… we run a niche site dedicated to the Mayan Calendar (no not 2012) that has a Facebook page of close to 16,000 fans, where we post certain types of content that people love (some literally start their day on our page). Our online community was WAITING for us to build an app for them, asking for it, requesting it, so when we launched it, this core fan base immediately started downloading. (This app has a very specific purpose, which is to give you your daily Mayan energy. It’s called the MCP Mayan Tzolkin, and arguably one of the first apps developed in close counsel with the Maya of Guatemala. It is also the very first app that interprets the daily energies of the Tzolkin calendar properly.).
Despite the fact that people were asking for us to build an app, we did not want to take that for granted and sit on our hands waiting for the whole world to download it. So we did spend a fair amount of time on keywords, the description, all of the meta data that goes into helping users “discover” the app. But because this app sits in such a specific niche, working on app discovery for the mainstream user has been a bit challenging. The Mayan Calendar is simply not on most people’s minds when they think of apps.
Having said that, the Face-to-Face that you discuss above Charles works pretty much exactly as you say: if I show the app to someone, invariably they say “wow” — the animation is beautifully designed — and many do download it right there and then. I also have to agree with Ouriel, there are simply too many apps out there today (and Apple allows a paltry 100 characters for the keywords in your app meta data!) for users to make sense of them all. To return to our particular case, our great challenge is figuring out how to reach all those people who a.) don’t even know the Mayan Calendar exists, or what it is; b.) don’t know why they should care; and c.) are likely to become passionate fans once they do download the app and use it for a few days, which is what happened with the content on our FB page.
I’d welcome any thoughts or suggestions from you guys or any other commenters on this post… and in turn would be happy to share our findings as we continue this journey we just began (we launched only last month).
Birgitte Rasine, Producer, Mayan Calendar Portal (www.maya-portal.net)