Open Radar

 
Summary: With the release of Logic Pro X as a new SKU, Apple seems to have made it clear that developers are expected to release new SKUs in order to charge for app updates (http://david-smith.org/blog/2013/07/16/apples-answer-on-upgrade-pricing/). There are many problems with this approach, not the least of which is that the App Store search algorithm seems to favor established apps.

Since the exact algorithm is only known to Apple, it’s tough to demonstrate conclusively what’s happening, but in my experience, an app’s cumulative downloads seem to be very heavily weighted in the search results. This puts new SKUs at a huge disadvantage compared to existing apps that have a long history of downloads. Because of this, releasing a new SKU rather than updating an existing app is much more risky for App Store developers.

Without knowing exactly how the App Store search algorithm works, it’s hard to suggest a specific solution, but I would think that using a rolling 1-3 month average of downloads with the past 24-48 hours more heavily weighted would give new apps a shot at ranking high in search results and maintaining that momentum over time if they remain popular.

Adjusting this and acknowledging the adjustment to developers prior to the release of iOS 7 would go a long way in easing developer’s concerns about going all-in on iOS 7 with a new SKU.

Steps to Reproduce: Search “Twitter” in the US App Store

Expected Results: As a popular Twitter app that is ranked highly in the Social Networking category, Twitterrific should also be high in the search results for “Twitter”

Actual Results: Twitterrific is the 78th result and is very unlikely to be found by users searching the App Store for “Twitter”

Regression: Gedeon Maheux of The Iconfactory tweeted yesterday that Twitterrific 4 used to be the 4th result for “Twitter”, but Twitterrific 5 is now the 68th result. https://twitter.com/gedeon/status/357980681965543425

Notes: I have no doubt that there are many factors that have likely combined to impact Twitterrific’s current low ranking in search, but from what I’ve seen watching search results over the past few years, I would venture to say that Twitterrific 4 would still be ranked very high if The Iconfactory had chosen to update the app directly instead of releasing a new SKU. This is corroborated by the fact that the free version of Tweet Caster is the 12th search result for “Twitter” despite being ranked #159 in the top free chart of the Social Networking category. Twitterrific 5 is the 78th search result and is ranked #22 in the top free chart of the Social Networking category.