Apple Miscalculates WOBBLE Averages in iTunes?

This has been bugging me for a while. Go to Wobble in iTunes, and take a look at the reviews – then add up the total number of stars given by reviewers and then divide by the number of reviews…

In every country our star rating is down by at least one, but for ease of calculation take a look at Canada for Wobble – Wobble has had 119 reviews in the USA (most of them 5 Star) – so that would be a pain to add up… in Canada Wobble has been reviewed just 8 times. (too busy skiing I guess?)

Anyway – Canadians have given us:
6 x 5 Stars
2 x 4 stars

Total 38 – Average (lemme get out my calculator… uh) 4.75 Stars – Wow, thanks Canada, I must visit for a marketing trip next ski season, oops I mean Winter…

EXCEPT Apple show just 3 Stars – Hey! How does that work??Canada 3 star

I know Apple don’t like Wobble, but geez? :-) Or maybe other Apps see the same thing? And before you say it’s a worldwide rating – it isn’t, I checked.

Update – after Grahs comment
(Not sure if this is the answer since there are many ratings with no reviews)

If this average is the result of people leaving ratings after uninstalling the Application then I think the formula to calculate how many people have left a rating when uninstalling is (38+y)/(8+x)=z where 2.75<z<3.25 SO, since that can’t be readily solved this is not a transparent process.

BUT, Let’s assume that X and y are equal – since a 1 star rating is likely, and assume z is exactly 3…. a few lines of scribble shows that 6 ratings of 1 Star is the answer.  In comparison there are just 8 ratings from satisfied customers in Canada. So 6 people who for whatever reason remove the app and give it a bad rating lower the stars from 4.75 to just 3.

If x and y are not equal and the average person removing the app is leaving a 2 star rating, then it means that there must be 12 x 2 star ratings to make an average of 3 stars when combined with the “real” ratings.

This makes the system appear even worse – 12 remove ratings to just 8 ratings from people who like the App enough to keep it AND come back to the App store to review it!!

I suspect that the more installs you get – the more UNinstalls you would probably get as well. This is why free applications – that have no apparent value to the user have lower ratings I guess.

As many people have said before, asking for ratings from people who uninstall the App is asking for skewed results.

average rating is wrong

Exciting Move Up The Charts + Sales Stats

Just woke up… being in Australia I should be excused – I’m not having an afternoon nap!

We’ve moved up the charts quite nicely, and got another Top 5 spot too…

#1 in Japan
#5 in Australia (yay!)
#29 in USA
#30 in UK
#33 in Canada

For the reference of other app developers that equates to total sales worldwide of about 3000 a day.

The sales seem to be pretty much the same between #34 in USA and #50 – which makes sense I guess since people browse that part of the charts evenly.

We’ll hopefully see if getting above #25 causes a jump – I suspect it will.

It sure would be nice to assign different icons for different localisations. Some countries might not like our new icon as much as others. For our next App I think we’ll bring out a different version for each country – not sure of the mechanics for that yet.

Review of the day – Canada

The Wobble that the app provides is as realistic as one could get. It has proven to be the life of many parties. Man no longer has to get slapped in the face for asking girls to jump up and down… Women grab the device out of my hands to try it just as quick as the guys. High entertainment value! FIVE STARS

Most ratings are 5 stars – I wonder why that is? :-)

I like happy customers…

Experiments

Now that we have visibility in the top of the charts, we’re experimenting with a few different names to see if they draw the eye of people browsing the top lists…

top10USA

I’ve seen a few of the aps in the top 10 try different lengthy names, and when you look at the top 10 in the paid list – some names stand out more than others.

The name we tried yesterday had a detrimental impact, so this isn’t an experiment I’m going to try for too long!