Our Properties: Gamasutra GameCareerGuide IndieGames Indie Royale GDC IGF Game Developer Magazine GAO
My Message close
Latest News
spacer View All spacer
 
February 10, 2012
 
What drives the developers of Unity?
 
Analyst questions validity of unusual January NPD results [13]
 
Road to the IGF: Lucky Frame's Pugs Luv Beats
spacer
Latest Features
spacer View All spacer
 
February 10, 2012
 
arrow Virtual Goods - An Excerpt from Social Game Design: Monetization Methods and Mechanics
 
arrow Principles of an Indie Game Bottom Feeder [21]
 
arrow Postmortem: CyberConnect 2's Solatorobo: Red the Hunter [1]
spacer
Latest Jobs
spacer View All     Post a Job     RSS spacer
 
February 10, 2012
 
Irrational Games
Systems Designer
 
CCP - North America
Lead Character Artist
 
CCP - North America
Sr VFX Artist
 
CCP - North America
Sr. Tech Artist
 
CCP - North America
Animation Director
 
Toys for Bob / Activision
Senior Programmer
spacer
Blogs

  Fixing the iPhone App Store
by Eric Clark Su on 08/19/09 09:05:00 am   Featured Blogs
11 comments Share on Twitter Share on Facebook RSS
 
 
  Posted 08/19/09 09:05:00 am
 

 Reprinted from the Cavalcade Games' blog.

The runaway success of the iPhone finds Apple in new and unexplored territories. With it comes new and interesting challenges, including a big one concerning its innovative app store. Much has been said regarding the app store and how its utterly broken. Interestingly, its only broken for developers. The other participants on the platform,namely Apple and the device owners are in a state of contentment. But its not rocket science to know that for a platform to have long term sustainability, it has to maintain equilibrium - meaning each of the participants need to be equally benefiting from each other. This equilibrium is threatening to disappear with the 'broken' app store.While the approval fiasco sparked by the Google Voice rejection has garnered more column inches, it's the app store issue that is most crucial and relevant to us developers.

In order to find a real and long-term solution, one has to clearly define the problem and its scope.

It's a common fact that the app store's main problem is visibility. In light of the recent news that most users download apps from the App Store directly from theirphones, we are gonna exclude external visibility points like iPhone websites,iTunes on desktop, etc.

VISIBILITY ON DEVICE

There are several visibility spots on the devices themselves: Featured section, What's new section, Category section, the Top 25 section, search and the updates section. We are going to concentrate on two of the most influential spots (by far) - the Search and the Top 25.

search

Search

Search was recently fixed with the new keyword policy enforced during app submission. Now, shady description field-based SEO tricks are less effective if not totally useless. While a positive step, it was a solution to a problem only few was vocal about.

The thing with Search is it's the last step of a purchasing intentstarted by an external influence (i.e. recommendations from someone ora website coverage). So few improvements can be thought of here. Sorting tools might be a good idea on paper but the I am not sure that it will fit with the small screen on the device.

The Top 25 (aka Top 100)

The Top 25 is based on the number of downloads in a sliding window of time (24 to 48-hours). In other words, its purely volume-based. This decision has its roots in the iTunes mp3 store - the direct precursor to the app store. Hindsight is 20/20, so its easy but not entirely fair to criticize Apple on building their top 100 that way. I would imagine the argument that a great app will have good volume at any price sounding sensible at the time. "Nobody would be desperate enough todrop their price to 99c just to achieve volume of sales. Right?" -that  rebuttal would contain.

This was fine for a time but after several high profile reports of developers hitting it big in the app store, a 'gold rush' happened. Pretty soon the ratio of developers to the app buyers became disproportionate. Too many apps are coming in.  This effectively rendered some visibility spots ineffective in driving sales. The What's New section became too crowded everyday. This was also true for the Updates section.  With that, the Top 100 became the only effective spot to get eyeballs. And the only way to get there is too achieve a certain volume of sales. And like in every market, digital or not, the lower the price of your product, the higher your volume is. So pretty soon, developer after developer broke down and went for the mad dash to the top regardless of the app's cost of development. Who could blame them?  The visibility problem allowed such a tactic to be very effective. With a few holdouts (TouchGrind developer illlusion labs comes to mind) everyone is now doing it including big companies such as Popcap and Chillingo. The app store, crushed by the weight of a thousand developers, morphed into a narrow channel for a small sliver of games. Now the question is- how to fix it?

The Common Fix

Now that we have clearly defined the problem, let me talk about a fix I keep reading about particularly from large(ish) developer / publisher companies. The most common 'fix' I always read is for Apple to change ranking algorithm from volume of sales to revenue. This was echoed recently by Tiger Style which just released their 1st iPhone game, Spider (which is awesome, by the way). By looking at the team's picture, you could feel their pain pricing their game at $2.99.

Tiger Style people

I think that changing to revenue-based is unrealistic for the following reasons:

  1. It will alienate a large segment of Apple customers who are used to 99c apps.
  2. A revenue system will disproportionately benefit license-based games,
  3. and games from big companies or games using existing IPs.
  4. It will be hard to factor in microtransactions.
  5. pop in, pop out games will become invisible (these games are what casual buyers love)
  6. the iPhone average cost (hardware + apps) will rise dramatically.
  7. a certain category of successful apps will instantly fall off the charts.
  8. a rogue app gaming the new algorithm is a more costly scenario.

Furthermore, revenue-based ranking will only benefit the developers.App buyers will not only be confused, they will think all the apps raised their prices and would be annoyed. Some will significantly reduce their app spending and just play the cheap ones. Overall, revenue willdrop and developers will find something else to blame. And we are back to square one except with angry customers and one less selling point. "There is an app for that... but its gonna cost you" doesn't have the same ring as the original.

THE SOLUTION

Customers, customer, customers 

Now that we've shot down that 'fix' out of the way, I think the real and long-term solution for Apple lies with its customers, of which there are two kinds - those who buy the apps and those who make them. First, the buyers.  The fact that they basically rely on the top 100 for their app purchases should point them to the right direction, any attempt to modify this behavior will be impossible. It's the most convenient and natural way of determining what's worth buying. Next are the creators. They have become too many that 100 slots will never be enough. Moreover, the sheer numbers developers create a real need for a store that supports a diverse selection of apps. Diversity is good for Apple and the buyers since it will expand the meaning of the device to in brand new ways.

Expand the charts

Here is a solution that I am laying out that will not only satisfy Apple's two primary customer types but will expand the impact of apps financially and creatively.

-- Expand the top 100 x 3.

In addition to the current top 100 based on volume, add another Top 100 based on Revenue, then add another based on customer Reviews. All unique and not just resorted lists. For a clearer picture, I have provided a mockup below. 

appstorefix
The New App Store?

Here are the advantages of this solution:

  1. Tab #1  will be the 'Top 100 by downloads' which maintains the current chart therefore sustaining the bite-sized or volume dependent ones.
  2. Tab #2  will be a new tab for a unique top 100 but this time by revenue (a better term for non-developers is needed here). This will create 100 more slots for bigger, robust apps. Some kind of algorithm involving no of download x price + microtransactions.
  3. Tab #3 will contain another top 100 based on reviews. No of reviews, average star rating and the country of reviewers = country of the device owner should factor into this.
  4. These 3 charts will be unique and not just the same pool re-sorted.
  5. These will bring up the slots to 300 for the paid charts. Of course, some apps will make a presence on 3 of these charts.
  6. Its relatively simple for Apple to do.
  7. Minimal complexity is introduced for the buyers - just one more tap.
  8. The app store can now sustain three kinds of paid apps - Volume-based, Features-based and Quality-Based.
  9. This will promote market growth as new types of apps will hook new types of buyers.

Is this the ultimate solution to the app store? Are there any disadvantages? Will this be enough to create a truly great app store?

 Cavalcade Games is a interactive software development company focusing on the web and iPhone platform. They just released their 1st iPhone product - Rock, Paper, Laser! - which is now languishing in bottom of the sea of broken dreams known as the app store. 

 
 
Comments

Erik Yuzwa
profile image
I think there's some excellent suggestions there, and it speaks to creating a bit more of a "longevity" model for both the gamer and the developer vs. the current model.

But nothing will be tinkered, until Apple is convinced / arm-twisted into even recognizing an issue. Perhaps they'll leave it for a few years until all the developers producing the content jump to the next one.

Eric Clark Su
profile image
agreed. the main problem here is Apple has a monopoly on this space. these developers wont go away soon because there is no other marketplace as exciting as the iphone. PSP go minis and the rumored DSi changes have a real shot in creating one. MSoft, Android and the pre are all too uncoordinated to capitalize on Apple's mistakes so its an exercise in frustration to wait for them.

Mac Senour
profile image
Although I think "reviews" should be an option, I think we need to keep in mind that there is a smoother path to a negative rating than a positive one. Most people are prompted to give a rating when they are deleting the game, and most people delete a game when they don't like it.

To give a positive review a user must go into iTunes, find the game and write a review. This is a much longer path and just by that fact alone, there will be more negative reviews than positive.

I talk about all this kind of stuff in my blog...

http://aboutmakinggames.blogspot.com/

Mac



Sean Parton
profile image
While the article is interesting and raises some very good points, I find your sources for your data to be terrible. The study you linked to had a sample size of only 190! In addition, the results only came from respondents who responded to a single company's in-app advertising. I find it tough to swallow that those respondents are actually representative of the app-buying public, when, to my knowledge, none of this company's adds are in not only an insignificant amount of well-known apps, but they seem to be only in free apps, as opposed to paid apps.

Your proposed solution of expanding the top 100 charts may not be too far off though. Being able to sort by different methods (such as game genre) may be a better solution, as it allows people who are into a certain type of app to have more visibility to similar apps.

Tom Kammerer
profile image
Although I agree something needs to be done. You are contradicting yourself to appease to the large publishers and developers with deep pockets. You explain why a revenue based algorithm would not work but throw it in your solution? That would un-level the playing field.

Also Customer reviews are very biased, maybe professional reviews?

I feel something needs to be done and I would just forfeit the overall top 100 and just force the users to use the game genres and/or types of apps which already exist. If this were brought to the for front it would give reduce the competition with other types of apps and just the particular type of app and game. It would also force the user to go through and look at even more apps than they would see on just the top 100 overall.


Now see? I just gave the best solution in a few sentences.

Sean Parton
profile image
@Tom Kammerer: I think that me (as both a customer and an iPhone developer) and many others would not like the most convenient way of finding a good app removed from the App Store. I don't feel that forcing the user to do more legwork in a currently poor system will help raise visibility of other apps, which is the issue at hand.

Niles Plante
profile image
If you have to force customers to get them to see your app, you are creating friction; friction that will cause a customer to simply leave the store without buying anything. Most customers will only look through a few pages before leaving. Customers go to a store, even when they are "just browsing", because they want to leave with something. That may be a product or a conclusion as to which product they want.

While more options in rankings are a start, the most successful drivers for digital distribution haven't been noted at all. They need to make the shopping experience contextual to the user. Most popular/purchased categories are a good start if you don't know your customer at all, but after a few purchases, the store should start to push content toward you in the form of "customers who bought this also bought this." Add collaborative filtering via ratings and you start to get a more complete view of a user. To get a complete view of the user, the app store would need to track which apps a user uses the most and which they have downloaded but haven't opened in a month or two.

These are more complicated but have proven successful on amazon, netflix, and most other standard digital storefronts and are well worth the investment if apple is to regain the equilibrium needed to keep developers and customers coming back.

Eric Clark Su
profile image
@mac - yes, i agree that negative reviews have a shorter path but still, based on the current system (really) good apps do get significant number of positive reviews. So in a way it still work, if its really a 5 star then you love it enough to go thru all that steps to rate it.

@sean - yes, the data source is not that great but viewing /buying from the device is the shortest path to purchase. wouldn't common sense dictate that is how the millions of device owner prefer to access the store? if you had a choice of opening the app store and purchasing vs going to your pc/mac, attaching your device, opening itunes, click click click.

@tom - the suggestion i contradicted was to REPLACE the chart with a revenue based one. I suggested to ADD their suggestion and at the same time lay out WHY creating these 3 categories of app models (not to be confused with genres) will be beneficial to the 3 platform participants.

your solution FORCES millions of Customer to change their habit. if you are a wise company like Apple you've learned to follow customers not tamper with their habits. and realistically the solution you offered presents new problems as genres are very subjective and open to dilution. therefore its not the 'best solution in a few sentence.' :D

@Niles - yes, having a contextual solution is a good solution but the article considers the amount of work Apple has to do. building a complex user tailoring system will be too large a task for Apple to do in a realistic timeframe, moreover complexity will be introduced into the UI. the app store needs to be friendly to (moms, grandpas, etc). the solution offered in the article only adds 1 tap to great efffect.

Tom Kammerer
profile image
I love the comments here, I agree with all of your counterclaims as friction would be an issue to my counter solution.

One thing to note is that currently one of the best strategies being employed, is where developers all over are hiring swarms of interns to flood there app with good reviews. It is a branch of Web 2.0 where they are also paying these interns to blog, tweet and facebook their app in order to compete with the 100k apps in the store. This is being done whether the app is great or not and what this is doing is creating a huge wall of fake reviews that the honest and ethical indy companies or sole developers need to overcome.

Not sure if this article is getting any more traffic but Eric if you check back maybe write an article on how to beat this? I am very interested as an iPhone developer myself.

Eric Clark Su
profile image
If you are referring to a certain PR firm's expose... I honestly see that as a non issue. they are paid to do that. Its their job to make their client's product have buzz. That strategy is being done in EVERY big market since the advent of the Internet. Also ethics in the business world is a thing of convenience so don't count on it to persist.

The point is it costs money to do that. so its a not enough of a bulletproof 'cheat' to collapse the whole system. does a 99c price cover cost to hire the PR firm? On top of that, a marketable app is still needed in the core of that strategy. making it a very valid strategy. it still needs a good product to work.

Richard Min
profile image
The appstore discovery, search and basic set is ..well... crap. On that we all basically agree.

But lots of big problems with your suggestions (coming from a developer, marketing and business and yes, user perspective).

The main first issue: "Gaming the system." Yes, that techcrunch expose recently (and I can personally attest), people and, esp big companies, are looking for all angles to "work the system." Think SEO black hat meets the apple AppStore. You brush over the response, but speaks to bigger problem that ethically debates. It speaks to the point of the article. The appstore is wacked.

Case in point: Making a tab ranking based on reviews only will lead to one inevitable result: spam levels of fake reviews. Its already being done (though questionable the effectiveness), but enough for much controversy with the assumption it affects rank. If there was an official exclusive ranking for it. Yikes! Big companies or those with budgets would throw money straight into networks of pre-set reviewers. 100% GUARANTEED. PR firm wouldn't need PR tactics...then just need a ROI conversion rate. THIS biases for big companies 10000x more than a revenue ranking model. Enter PPC for appstore and basically kill the review system people would start to trust reviews even less (from whatever level they trust them now) because it is a straight rank.

Read: they ALREADY trying and doing this NOW (and that's just on assumptions that it helps rank!!!).

* Revenues ranks at least would skew the average back UP instead of this massive race to the bottom and basically giving quality development no viable choices to actually support the effort (see Spider game developers article saying the basically same). There will just be no money in it. And it would be more fair to degree given, well, its not something anyone can directly control. ie: it would balance out.

* Download counts are great for developers, marketers and industry reporters and pundits, but not really the user or the system. Its nice to see but often really has little to do with quality. How often is that refreshed? monthy? weekly? all time? is there an archive to access? Doesn't this bias towards games released last year and discriminate vs new games? "History of Dance" is miserable video imho, but its clear at least part of its status on youtube is simple due to the fact its been there forever as well. every watches the number one all time video at least once...new and old users alike. Plus, biggers companies then could theoretically just "brute force" buy into the top lists then. yes, they absolutely can..which means at least a few absolutely will.

CONCLUSION:
At the end of the day, the algorithm clearly needs to be improved, but it should really be done behind the scenes and should be COMBINATION of factors. Factors that are more than 3 as well. The RESULT should be rational and reasonable, but ultimately, fair and quality determined ("quality" be relative to the kind of app --ie: fun, or useful etc).

Discovery and exposure is now too important in the appstore to be left to three factors, four even 5. They should be thinking google/yahoo scale algorithms.

Hell, Google uses some 200+ different factors to affect search rank... and given the importance of rank in Appstore (and the importance of the appstore itself now as an self-contained economy), three tabs isn't going to cut it. And search is a very relevant benchmark. Apple knows crap about it. They just happened into it this mess.

The appstore needs a reboot or developers may eventually start moving to android (which we can assume will someday have an appstore and have that whole search and exposure thing down pact *ahem* google*)


none
 
Comment:
 




 
UBM Techweb
Game Network
Game Developers Conference | GDC Europe | GDC Online | GDC China | Gamasutra | Game Developer Magazine | Game Advertising Online
Game Career Guide | Independent Games Festival | Indie Royale | IndieGames

Other UBM TechWeb Networks
Business Technology | Business Technology Events | Telecommunications & Communications Providers

Privacy Policy | Terms of Service | Contact Us | Copyright © UBM TechWeb, All Rights Reserved.