★ TouchArcade needs your help. Click here to support us on Patreon.

Cramzy published financial results & analysis

12-21-2009, 10:19 AM
#1
Joined: Apr 2009
Location: London, UK
Posts: 317
Send a message via ICQ to cramzy Send a message via Skype™ to cramzy
Cramzy published financial results & analysis

[original press release, that will be published later today]

Cramzy today published the financial results and analysis for two of their apps, that Apple featured few weeks ago.
You can see the actual boost in sales that appears when Apple features you, but there's more than that. Cramzy found out, that the Top 100 Paid chart is not purely based on number of sales.

On 1-Dec-2009 Apple featured 23,000 GREAT QUOTES in the US AppStore's "What's hot" section. The app was making approximately $120 per day before that, and was #4 in the Reference category in the US AppStore. After being featured, 23,000 GREAT QUOTES achieved #24 in Top Paid overall apps for the US AppStore, #88 in Top Grossing apps, and at its peak made almost $1,600 profit (after the 30% deduction from Apple) for a single day.

On 8-Dec-2009 Apple featured 1,001 MURPHY LAWS in the US and Canada AppStores, again in the "What's hot" section. The app was making about $20 per day before that, and was #85 in the Reference category in the US AppStore. After being featured, it achieved #103 in the Top Paid overall apps for the US AppStore, and at its peak made a little over $500 for a single day.

"After 1,001 MURPHY LAWS started falling in the charts, we observed something really interesting", said Iliya Yordanov, founder of Cramzy. "When this app was ranked #208 in the US AppStore, it made exactly the same amount of sales as another of our apps, 18,000 COOL JOKES which was then ranked #382 in the same chart. The next day, 1,0001 MURPHY LAWS fell by another 10 positions, and 18,000 COOL JOKES rose by 10 positions, and 18,000 COOL JOKES made about 10% more sales than 1,0001 MURPHY LAWS. It turns out that the Top Paid chart is not purely based on number of sales. The question remains- what else does Apple take into account while populating these charts?"

Sales chart


Links
www.cramzy.com
1,001 MURPHY LAWS
23,000 GREAT QUOTES
18,000 COOL JOKES
12-21-2009, 10:26 AM
#2
Joined: Jun 2009
Posts: 10,505
Quote:
Originally Posted by cramzy View Post
The question remains- what else does Apple take into account while populating these charts?"
My guess will be that Apple uses more than one day`s sales for the Top Paid Charts. Something like last 7 days average. Can you see a big difference in the 7 days average sales of both your apps at that time?

12-21-2009, 10:46 AM
#3
The problem is, Apple does not report positions greater than 100 in any category. So these positions of 203 and 382 are just estimates based on factors that are essentially guesses (not sure which site you are getting those positions from, each website appears to have different fomulas). That's why you are seeing odd things when trying to track past position 100.

The top 100 in each category most definitely are sliding window averages. The math's been calculated and posted by a number of independent sources now. Most of which seem to agree

--- ChronoSoft ---
Support your roguelikes! Play Rogue Touch today!
Spirit Hunter Mineko: Demons Reach --- Work in progress! Follow us on Twitter!
12-21-2009, 10:59 AM
#4
Joined: Apr 2009
Location: London, UK
Posts: 317
Send a message via ICQ to cramzy Send a message via Skype™ to cramzy
Quote:
Originally Posted by CommanderData View Post
The problem is, Apple does not report positions greater than 100 in any category.
That's not quite correct. Apple gives these charts from their RSS here:
http://ax.itunes.apple.com/rss

You can manually modify the GET parameter in the URL to adjust it to display up to Top 400 and it works for each category as well as overall.

Cheers,
ILIYA
12-21-2009, 11:01 AM
#5
Joined: Apr 2009
Location: London, UK
Posts: 317
Send a message via ICQ to cramzy Send a message via Skype™ to cramzy
Quote:
Originally Posted by ImNoSuperMan View Post
My guess will be that Apple uses more than one day`s sales for the Top Paid Charts. Something like last 7 days average. Can you see a big difference in the 7 days average sales of both your apps at that time?
Here're the stats:
1,001 MURPHY LAWS for the past 7 days: $1797
18,000 COOL JOKES for the past 7 days: $1480

What you suggest sounds reasonable, though the difference in sales for this period is not that significant in my mind.

Cheers,
ILIYA
12-21-2009, 11:12 AM
#6
Joined: May 2009
Location: UK
Posts: 741
I read a few months back in this forum that someone had worked out that the calculation is based on something like this:

4 x today's sales +
3 x yesterday's +
2 x the day before yesterday +
1 x three days ago +
1 x four days ago +
1 x five days ago +
1 x six days ago +

It is of course impossible to be sure as your graph is relative to the chaos of everyone else's sales but this fitted his case very well.

Those multiples could of course be any fraction and tail off for far further than 7 days. "Today" also probably means a sliding window of the last 24 hours or it could even be a more complex equation starting with 50 x the last few hours. Nobody knows for sure but it's clearly something like this as when my feature dropped off it took several days for this to be fully reflected in the chart position.

-=< Fat Owl With A Jetpack >=-
-=< Topia World Builder >=-
-=< Twitter >=-
-=< Blog >=-
12-21-2009, 11:14 AM
#7
Quote:
Originally Posted by cramzy View Post
That's not quite correct. Apple gives these charts from their RSS here:
http://ax.itunes.apple.com/rss

You can manually modify the GET parameter in the URL to adjust it to display up to Top 400 and it works for each category as well as overall.

Cheers,
ILIYA
That is interesting of course, but it does not appear to have a way tell you how games *subcategories* are doing (such as RPGs, Racing games, Puzzle games), so it's not very useful to me as an iPhone game developer unless I'm somewhere in the top 300 or 400 overall games

--- ChronoSoft ---
Support your roguelikes! Play Rogue Touch today!
Spirit Hunter Mineko: Demons Reach --- Work in progress! Follow us on Twitter!
12-21-2009, 11:53 AM
#8
Joined: Apr 2009
Location: London, UK
Posts: 317
Send a message via ICQ to cramzy Send a message via Skype™ to cramzy
Quote:
Originally Posted by CommanderData View Post
That is interesting of course, but it does not appear to have a way tell you how games *subcategories* are doing (such as RPGs, Racing games, Puzzle games), so it's not very useful to me as an iPhone game developer unless I'm somewhere in the top 300 or 400 overall games
For you and all the game developers it might not be useful, but we do not do games mostly, so I find it quite helpful

Cheers,
ILIYA
12-21-2009, 11:53 AM
#9
Joined: Apr 2009
Location: London, UK
Posts: 317
Send a message via ICQ to cramzy Send a message via Skype™ to cramzy
Quote:
Originally Posted by GlennX View Post
I read a few months back in this forum that someone had worked out that the calculation is based on something like this:

4 x today's sales +
3 x yesterday's +
2 x the day before yesterday +
1 x three days ago +
1 x four days ago +
1 x five days ago +
1 x six days ago +

It is of course impossible to be sure as your graph is relative to the chaos of everyone else's sales but this fitted his case very well.

Those multiples could of course be any fraction and tail off for far further than 7 days. "Today" also probably means a sliding window of the last 24 hours or it could even be a more complex equation starting with 50 x the last few hours. Nobody knows for sure but it's clearly something like this as when my feature dropped off it took several days for this to be fully reflected in the chart position.
This is another sick calculus

Cheers,
ILIYA
12-21-2009, 01:15 PM
#10
Joined: Sep 2009
Location: UK / Toronto
Posts: 602
Quote:
Originally Posted by GlennX View Post
I read a few months back in this forum that someone had worked out that the calculation is based on something like this:

4 x today's sales +
3 x yesterday's +
2 x the day before yesterday +
1 x three days ago +
1 x four days ago +
1 x five days ago +
1 x six days ago +

It is of course impossible to be sure as your graph is relative to the chaos of everyone else's sales but this fitted his case very well.

Those multiples could of course be any fraction and tail off for far further than 7 days. "Today" also probably means a sliding window of the last 24 hours or it could even be a more complex equation starting with 50 x the last few hours. Nobody knows for sure but it's clearly something like this as when my feature dropped off it took several days for this to be fully reflected in the chart position.
That seems to make a lot of sense. If it was based only on daily data then apps would jump in and out of the chart like flies. It seems they've created an algorithm to give the charts realtime movement, while keeping the benefits from weekly sales charts of retail stores.