Feb 24th 2014, 17:50, by Ouriel Ohayon
Editor’s note: Ouriel Ohayon is co-founder and CEO of Appsfire, a startup focused on mobile native advertising and technologies. Follow him on Twitter @ourielohayon.
Flappy Bird has been a wake-up call to our industry. Not because of its success but because of what it has revealed about the App Store. This little game has unexpectedly reset to zero what works for users. And more.
As the developer of Flappy Bird, at the summit of its success, decided to kill his app before tens of millions of excited users, a tsunami has shaken the industry and the App Store, and has triggered a wave of new greedy clones that have inundated the market.
Suddenly the (well-known) flaws of the App Store have been exposed to the world in an unprecedented order of magnitude, reminding us how broken it is and how little has been done to fix it.
A few days ago about 10 percent of the App Store top 150 free ranks in the U.S. was owned by Flappy Bird clones of some sort, even more if you add other clones (like this clone of Plague). How is that even possible? Some are saying bots. No!
Simple search “spam”: Flappy Bird, once out, has caused an unprecedented PR wave that translated into hundreds of thousands of daily searches for “flappy something” and boom! Developers place the “flappy” keyword somewhere in the title or their keyword or description (now the “Miley” or “Cyrus” or “Wrecking” keyword), and boom! Instant “hit.”
But Apple has decided to arbitrarily reject apps that are “leveraging the success of other apps,” although this rule is not inked in the Terms of Service. And that was not enough, because, as with many rules Apple has, this one is not enforced consistently.
Some developers have even found a way to game the review team by submitting an app with a certain name and then changing it to another down the road based on the hits of the moment. It is even possible to have the exact same game with the same name up in the charts owned by two different developers.
The main reason this is happening is not because some developers have bad intentions, but because Apple doesn’t care enough. At least not yet.
How many billions or trillions of downloads should there be until something happens? How many billions of dollars should Apple distribute to realize that the App Store is the only store hundreds of thousands of developers depend on? There’s no need to wait for the post “pre-page rank” stage that Benedict Evans talks about. Things can be done now.
It doesn’t look as though Apple’s acquisition of Chomp had any significant impact in improving search and discovery at large. It is still difficult to find the apps you want and make a choice out of the endless list of meaningless icons and screenshots. Even to date, well-covered apps like Jelly or the hyped Secret are impossible to find when you search their name. The algorithm seems to ignore the “hotness” factor.
On the other side, developers have found back doors to leverage search keywords to surf the wave on the back of others’ success. Apple can’t always play cat and mouse here.
Search can work only if the right mix of curation, context and experience is brought to the users. For example, instead of an endless lists of cards, Apple could provide users within the search results with a one-tap access to filters, such as “categories” and “recency.”
How about also empowering developers with some knowledge about how their apps are being discovered and searched within the App Store? No one really knows. Everyone is trying to guess with the so-called and unscientific “App Store optimization” magic. But opening the knowledge pipe would actually make search a lot better because there would be more matches between what is searched and what is found.
If discovery for Apple is the App Store and iAD, then Apple needs to fix iAD. Flappy Bird and its clones have been exposed to tens of millions of users in a very short period of time. We all saw at scale what was wrong with mobile advertising: overexposure of the same banners for apps that probably many users already owned, are designed poorly and are intrusive to the user experience.
We can do better than that. Why advertise apps already owned by users? Why not apply a reasonable frequency cap of exposure? Why not offer an opt-out option from the banner itself (not just in the iOS settings) and design a native elegant integration (hint: native ads).
This is not just iAD. Google’s admob is even worse.
Fix The Review Process
I don’t think developers have a problem with guidelines. They have a problem with guidelines being unclear and inconsistently enforced as is obviously the case here. It’s really hard to believe no one at the review team is not jumping off their chairs watching this Flappy show. I’m not sure how their review is organized, but it takes very little effort to actually spot those clones and act quickly.
Apple should also limit the amount of times a developer can change the name of their app and their keywords, at least until they fix search.
Fix Developers’ Communication
Apple can’t communicate with developers just by way of sudden rejection and ad hoc rules. Why do developers have to wait to have their apps rejected in order to find out what the rules are truly about? Apple has the right to do that, because it’s their store. But they can raise the bar.
Why not have a proper developer blog (with open comments) to better explain their intention. Or even a developer forum hosted and managed by Apple execs? Yes, Apple hosts WWDC to talk to developers — those who manage to get tickets — and the company regularly invites a few selected developers to get featured and discuss future projects. But are there enough people dedicated to that?
Ask any developer if they believe Apple does enough there. They don’t and the company could do a lot more.
Fix Top-Ranks Discovery
The top 10 ranks right now are a mess. Apps in the top 10 are mostly clones and certainly not due to the effort of any ad network or even Apple editorial section. I doubt that this is what Apple wants us to see in the App Store. Top ranks right now are not owned by Apple, bots or even ad networks. They are owned by kids who, by way hectic of word of mouth, share the hottest memes at light speed. No one can beat or control that with the system in place.
But that also means that they’re not appealing to most of the users who own an iOS device. In 2014, one monolithic store designed the same for all of us isn’t the right thing to offer.
There are two ways Apple could fix the top-ranks issue:
Design them in a way that will resonate with more people by showcasing those who have more engagement or correspond to an adequate set of tastes. In other words, top ranks that would be more personal and contextual and not based merely on download velocity but also on quality. For example, there could be a separate rank for games and non-games, which are equally accessible and visible (hint: get rid of “apps near me” that don’t deserve this kind of exposure).
Or just get rid of the top-ranks system altogether. Either that or bury them three taps down from the main screen.
If you think about it, top ranks are the only reason we’re seeing abusive practices in the App Store. Developers want to reach out for the top ranks because of the additional organic growth they will get and the promise of more dollars. This is why incentivized downloads have been invented and why bots are still operating at scale. It’s why illegitimate “app discovery” services have been operating with or without Apple’s blessing. And it’s why manipulative and deceptive ads run at scale.
Remove top ranks and instantly remove everything that is wrong about app promotion and marketing in our industry. Many have echoed that idea.
Once a week I receive an email from a service that sells paid reviews in the App Store. For 20 bucks, you can get your dose of undeserved 5-star ratings.
The review system is broken. It is impossible to trust it and easy to game it. Many have pointed to the weird review patterns of Flappy Bird. But the point is not about whether those are gamed, but rather whether the Apple rating system for apps can be trusted. It can’t. During the age of Facebook/Twitter-connected services, the identity of the reviewers is impossible to verify, even at elementary levels. Only Apple can do this.
What could be done to fix reviews? Rotten Tomatoes and Metacritic have done it with TV, movies, games and music. We need a quality index for apps (this is something we started to build at Appsfire with the “App score”).
This quality index could be a composite of power users’ reviews (vetted by Apple) and trusted third-party reviews.
All that users want to know finally is whether an app is worth their attention or not. Not whether Flappy Bird got 75,000 reviews in fewer than 24 hours.
Bottom Line: Fix The Flappstore
The Flappy saga will at some point fade away, and maybe some of the problems that I highlighted here will be less visible. But make no mistake, Flappy Bird or not, the problems will not go away. Something else will come up and will ruin the App Store again and, most importantly, prevent great app developers from succeeding.
Let’s hope by then Apple will do something about it. They have the resources to do it. It won’t be easy, because iTunes is built on outdated legacy software and on a complex infrastructure. Some companies are entirely dedicated to fixing some subsets of those issues.
The App Store is now too big and too important to leave it as is and not take dramatic steps to fix it.