Nokia does not report all sales and I can prove it

[Utumno]

Users of my app can optionally submit their scores to my server. I can see that about 20% of them do so, and when they do, I record their 'useragent' in my DB. Actually 95% of the time 'useragent' is 'None' ( I don't know why - is it because most operators filter it? ) but sometimes I do get to lie my hands on a full useragent. Let's see how it looks:


mysql> select date,query,useragent from logs where useragent!="None" order by date;
+---------------------+---------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
| date | query | useragent |
+---------------------+---------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
| 2010-01-06 06:24:20 | | Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; InfoPath.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729) |
| 2010-01-06 06:32:59 | | Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; InfoPath.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729) |
| 2010-01-08 05:51:06 | l=0&c=1&t=131&v=&n=Jesus&s=%2b61415011501&h=360&r=0 | Nokia6700c-1/2.0 (07.90) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-08 22:27:15 | l=0&c=1&t=52&v=&n=Annelie&s=%2b27841000000&h=141&r=0 | Nokia6303classic/2.0 (06.40) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-09 20:52:54 | l=0&c=1&t=9&v=&n=T&s=%2b393359609600&h=64&r=0 | UP.Link/6.3.0.0.0 |
| 2010-01-14 08:11:58 | | Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; InfoPath.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729) |
| 2010-01-14 18:00:27 | l=0&c=1&t=46&v=&n=.da.&s=%2b447973100973&h=369&r=1 | Nokia6700c-1/2.0 (07.90) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-15 01:49:17 | | Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; InfoPath.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729) |
| 2010-01-18 21:13:20 | l=0&c=2&t=4&v=TEST&n=Peter&s=%2b886931000099&h=479&r=12 | Nokia3120classic/2.0 (07.16) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-18 21:13:58 | l=1&c=2&t=18&v=Peter&n=Peter&s=%2b886931000099&h=177&r=12 | Nokia3120classic/2.0 (07.16) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-18 21:14:15 | l=2&c=2&t=36&v=Peter&n=Peter&s=%2b886931000099&h=291&r=12 | Nokia3120classic/2.0 (07.16) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-18 21:14:47 | l=3&c=2&t=136&v=Peter&n=Peter&s=%2b886931000099&h=91&r=12 | Nokia3120classic/2.0 (07.16) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-18 21:27:12 | l=4&c=2&t=485&v=Peter&n=Peter&s=%2b886931000099&h=52&r=13 | Nokia3120classic/2.0 (07.16) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-21 07:11:51 | l=1&c=1&t=94&v=&n=Alpocoloco&s=%2b447802000332&h=509&r=0 | Nokia6300/2.0 (06.01) Profile/MIDP-2.0 Configuration/CLDC-1.1 |
| 2010-01-21 07:16:12 | l=3&c=1&t=234&v=Alpocoloco&n=Alpocoloco&s=%2b447802000332&h=258&r=0 | Nokia6300/2.0 (06.01) Profile/MIDP-2.0 Configuration/CLDC-1.1 |
| 2010-01-21 07:33:33 | l=1&c=0&t=6&v=Alpocoloco&n=Alpocoloco&s=%2b447802000332&h=83&r=0 | Nokia6300/2.0 (06.01) Profile/MIDP-2.0 Configuration/CLDC-1.1 |
| 2010-01-22 19:00:50 | l=0&c=1&t=303&v=&n=Bamberlaa&s=%2b447802000332&h=183&r=0 | Nokia6700c-1/2.0 (07.60) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
| 2010-01-24 20:24:58 | | Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; InfoPath.1; .NET CLR 2.0.50727; .NET CLR 1.1.4322; MS-RTC LM 8; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729) |
| 2010-01-31 09:17:00 | l=0&c=1&t=182&v=&n=Sarah&s=%2b491722270333&h=89&r=0 | Nokia6700c-1/2.0 (07.60) Profile/MIDP-2.1 Configuration/CLDC-1.1 |
+---------------------+---------------------------------------------------------------------+----------------------------------------------------------------------------------------------------------------------------------------------------------------+
19 rows in set (0.00 sec)

Sorry, a bit unreadable. Look at the 'useragent' column - we can see several Mozilla/4.0's ( I dont know what model can that be so lets forget about that ) , we can see a Nokia3120classic ( that's me ) and , most importantly, we can see this (cut for brevity):


2010-01-08 05:51:06 | l=0&c=1&t=131&v=&n=Jesus&s=%2b61415011501&h=360&r=0 | Nokia6700c-1/2.0
2010-01-08 22:27:15 | l=0&c=1&t=52&v=&n=Annelie&s=%2b27841000000&h=141&r=0 | Nokia6303classic/2.0
2010-01-14 18:00:27 | l=0&c=1&t=46&v=&n=.da.&s=%2b447973100973&h=369&r=1 | Nokia6700c-1/2.0
2010-01-21 07:11:51 | l=1&c=1&t=94&v=&n=Alpocoloco&s=%2b447802000332&h=509&r=0 | Nokia6300/2.0
2010-01-21 07:16:12 | l=3&c=1&t=234&v=Alpocoloco&n=Alpocoloco&s=%2b447802000332&h=258&r=0 | Nokia6300/2.0
2010-01-21 07:33:33 | l=1&c=0&t=6&v=Alpocoloco&n=Alpocoloco&s=%2b447802000332&h=83&r=0 | Nokia6300/2.0
2010-01-22 19:00:50 | l=0&c=1&t=303&v=&n=Bamberlaa&s=%2b447802000332&h=183&r=0 | Nokia6700c-1/2.0
2010-01-31 09:17:00 | l=0&c=1&t=182&v=&n=Sarah&s=%2b491722270333&h=89&r=0 | Nokia6700c-1/2.0

Before users submit, they first have to choose their username. It's the 'n' parameter in their query, so above we see the following users:

1) Jesus using a 6700c (heh!)
2) Annelie using a 6303classic
3) .da. with 6700c
4) Alpocoloco with a 6300 ( 3 times )
5) Bamberlaa with a 6700c
6) Sarah with a 6700c

Users cannot choose a username that is already taken, so the point is that we have at least four 6700c's, a 6300 and a 6303 here. Now let's look at my sales report 'per device' list:

#1 Nokia E71 22.4 %
#2 Nokia N95 9.5 %
#3 Nokia N95 8GB 7.8 %
#4 Nokia E75 6.9 %
#5 Nokia N96 6.0 %
#6 Nokia N86 8MP 6.0 %
#7 Nokia E66 6.0 %
#8 Nokia E63 6.0 %
#9 Nokia E72 5.2 %
#10 Nokia 6210 Navigator 2.6 %
#11 Nokia N85 2.6 %
#12 Nokia E52 2.6 %
#13 Nokia 6220 classic 1.7 %
#14 Nokia N79 1.7 %
#15 Nokia E71x 1.7 %
#16 Nokia N82 1.7 %
#17 Unresolved 1.7 %
#18 Nokia E90 Communicator 1.7 %
#19 Nokia 5630 XpressMusic 1.7 %
#20 Nokia N78 0.9 %
#21 Nokia 6303 classic 0.9 %
#22 Nokia 5800 XpressMusic 0.9 %
#23 Nokia 6760 Slide 0.9 %
#24 Nokia N81 0.9 %

( yes, yes, I know you can calculate I have sold 116 copies this month

Now, there is one Nokia 6303 here. Where are the 4 6700c's and the 6300 though? The 'unresolved' device stands at 1.7%, which means 2 sales, so that does not explain it.

On top of that, in the sales report we can see a 5800 XpressMusic - a S60 5th edition touchscreen device, with which my app is incompatible and clearly marked as so in Ovi. I don't know what the poor chap with the 5800 did. For sure, on a 5800 with no keyboard he cannot do jack shit with my app.

[shaii]

hmmm have you thought about the option that those highscore entries where from nokia qa test?
the 6700 afaik is a very new handset i think even prototype so that could explaint it...
u shouldnt rely on highscore results
im not saying nokia is perfect (there are the 10% matter of missing sales between the handset table and the country table)

[Utumno]

And who told you about the missing 10%? Also me. How come there are supposedly thousands of developers here but no one has noticed such obvious stuff before?

I am not saying Nokia is hiding part of the sales on purpose - that would be silly indeed. I am saying the sales reports dont hold water ( the missing 10%, the 5800 where it shouldn't be , the missing 6700c and 6300 ) We deserve an explanation for all of this, don't you think?

Besides, 6700c and 6300 are certainly being sold already. Go to www.nokia.com and see it for yourself.

[petrib]

How much of the usage is cracked/warez app users?

[Utumno]

I published the app exclusively to Ovi on Jan 6th. There's no physical way anybody can get it from anyplace else. Are you saying Ovi is leaking? That would be even worse.

[petrib]

It only takes one person to grab the software from/through one phone and then redistribute it using other channels.

Whether that has happened to your software, you can try to find out by trying to locate the various pirated software repositories/forums and search engines on the internet, and see if you can find warez versions of your app.

[tamhanna]

It only takes one person to grab the software from/through one phone and then redistribute it using other channels.

Whether that has happened to your software, you can try to find out by trying to locate the various pirated software repositories/forums and search engines on the internet, and see if you can find warez versions of your app.

I have to admit that I haven't seen this happen to any of my apps so far.

(knocks on wood)