WoW Subscriber Data - Complete Breakdown

[PriestRela]

Hello All..... Someone once said beware me when bored, they were right. So I picked through press release archives and reports from Blizzard Activison to compile a complete breakdown of wow sub numbers over the whole product lifespan. Here goes.....warning graphic heavy.

Overall Product Life

• This chart compiled from offical released numbers. Pre-2008 from Blizzard Press Releases and post from Activison-Blizzard Investor Info
• Not every quarter had an offical report
• This chart is the bible for all calculations after.

Methodology: Calculated Avg Growth/Loss per day in time period assuming 1Q=90 days normal avg with a 2-4% margin of error.
.

• This is extrapolated data. See Above chart for offical release periods as marked.

Growth/Loss
.

.

• This chart shows growth/loss avg'ed for each day in period.
• Only offical release intervals used in this chart.
• Notice the large stagnation period during WOTLK with minor growth.
• Cata Loss avg per day has not be consistent (explored later)

.
.

.

• This chart shows avg growth/loss per day shown with the amount of days that level was maintained
• The red bar shows growth or loss per day.
• The blue bar shows the number of days the growth or loss was sustained

.
Growth/Loss Trends by XPAC
.

.

• Note: Vanilla was started to be released in world markets 2005 Q2
• Most growth in wow was during vanilla, not later on.
• TBC showed steady growth, most of any xpac.

.
.

.

• WotLK was the slowest growth xpac before cata.
• WotLK retained sub numbers but only grew 500k in entire xpac life.
• Cata numbers to date has dropped to 2007 Q4 TBC numbers.

.
Cata Loss Breakdown
.

.

• This Chart shows avg daily losses for Cata by month.
• These are extrapolated numbers.

.
.

.

• IMO: This is the most interesting chart
• This chart shows sub losses in timeline with US/EU patch releases.
• China launch is marked. CHINA LAUNCHED at Patch 4.1 with Troll 5mans and nerfs.
• The numbers are NOT local losses but global reported numbers.
• All numbers on this chart from Blizz offical releases.
• The loss, POST CHINA is not as bad as Blizz said... I'm starting to question that story.
• As China launched with 4.1 and nerfs, the inital diffculty shock the US/EU experenced was lessened. Chart shows a softer blow.
• The biggest Loss period was during Firelands.
• This chart is vaild to the end of 2011.

.
.
Now a Chart Just for Fun
.

.

• This chart figures the value of your (1) account as a percentage of total subs.
• Your basically worthless to Blizzard.. (lol im kidding)
• The number is lower than even spacecraft machined tolerances.
• Hey, on the upside... Cata losses rasied the value of your account to 2007 levels.

.
I'll answer and questions you have, or crunch any data you like to see from these numbers.

[ruga]

Where are the pie charts?!

[PriestRela]

Where are the pie charts?!

Pie charts dont work with this kind of data....

However, did you know that 75% of pie charts resemble pacman?

[Porimlys]

Interesting figures, thanks for the post. Hopefully they'll have better years ahead of them in terms of subscriber numbers

[Cowt]

Well done. Kodos to you.

[Zenko]

Were you one of the blizz 600? You have an insane, yet brilliant, interest in WoW subs =p

[bechill]

I don't think "Cata fixt it."

Nice charts though. I'm no statistician, but I wonder if one would be able to forecast using the data? I'm sure if they could, there would probably be a significant margin of error?

[PriestRela]

Were you one of the blizz 600? You have an insane, yet brilliant, interest in WoW subs =p

I have not ever been affiliated with or employed by Blizzard, activison, or a company related to them. I got putting this together over 3 days, and it was spurred on because of the constant sub arguments. Something always struck me as off about them....

And after cruching some numbers, i now doubt the whole Asian lost subs myth......as the numbers mapped out don't seem to support it. And was shocked to see most losses occurred both before and after firelands.

Some interesting stuff popped up there.

[primusl0l]

How the hell did the game gain 5million+ subs from Q1 2005 to Q4 2005....that's quite an achievement.

Compared to other MMOrpgs out at the time...that growth is fucking ludicrous....can you imagine the people working there, their eyes would light up with $ signs.

[PriestRela]

I don't think "Cata fixt it."

Nice charts though. I'm no statistician, but I wonder if one would be able to forecast using the data? I'm sure if they could, there would probably be a significant margin of error?

Considering my extrapolation of some numbers were already at a 2-4% margin of error, yes it would be 10-20%. Plus the larger Macro-economic theory of bell curving may already be in effect (see 2nd chart....) you'll start to notice a mirroring. Im already working on some Q1 predictions. I'll have a chart soon.... but early numbers:

Lowend: 300k loss, middle: 550k loss, highend: 700k loss.

I'm using a weighted avg based on avg daily losses and patch content release curving. But I may revise those numbers worse.... on the account DS is going down so well. But pure math says: 550k-570k is likely (5% margin of error for projecting 90 days and not counting SWOTR or alike at all)

How the hell did the game gain 5million+ subs from Q1 2005 to Q4 2005....that's quite an achievement.

Compared to other MMOrpgs out at the time...that growth is fucking ludicrous....can you imagine the people working there, their eyes would light up with $ signs.

it was 4.7 mill in that timeframe. It had alot to do with EU and AUS launches. Also big sales in the US. It is an amazing rise if you consider it against rift or EQ or even Ultima back in the day. Even in pure percentage form.

[PriestRela]

Hmm, nice work.
But one problem: With "Growth/Loss Trends by XPAC".
It would be better if the vertical bars were the same for each graph.

Initially it was... However TBC didn;t really show the movement of the data, and wotlk just looked like a flat line and sea of blue. So i move the scales to show detail. The much larger detailed charts over the life spawn show the movements relative to each other, I wanted to show the detailing inside of the xpac's run as a single piece for academic comparison.

[Thetruth1400]

Great charts, just have a few problems with results.

I think your number loss for Q1 of 2012 is a bit on the low side. It's only taking mathmatical trends into consideration, not actual game trends. As a stock holder, it's very interesting, but in terms of business market values, I think you're going to see close to 500K on the low end, 800 the middle ground and most likely, and worst case scenario, 1.5 million subscription loss.

In addition to that, the last chart, basing the worth as a subscriber, from a math stand point is fine, but from a business stand point, is way off. 1) American and EU customers pay more than Chinese, therefore the value of an American or EU subscriber is more valuable. 2) The cost of employing people, maintaining servers, etc doesn't change unless those actually do. If you have 10 employees and 10 servers, the cost to run and pay those is the same if you have 100 customers or 200. Considering most of the work isn't physical that is. While a single customer is a drop in the hat, that $15 a month they pay goes towards paying for employees, servers, etc. $15 not paid comes out of someone's pocket, which is felt from a business perspective.


All in all though, great charts. I hope you make more after this coming year is up.

[PriestRela]

Great charts, just have a few problems with results.

I think your number loss for Q1 of 2012 is a bit on the low side. It's only taking mathmatical trends into consideration, not actual game trends. As a stock holder, it's very interesting, but in terms of business market values, I think you're going to see close to 500K on the low end, 800 the middle ground and most likely, and worst case scenario, 1.5 million subscription loss..

I'll let you into my thinking here, but I'm gonna revise this because of swtor too (they are gonna drop some numbers soon) but I'm also factoring in China too as they are soon to have Fireland's which may rebound or (more likely retain) subs as its new content. HOWEVER, I also must consider that they had Troll 5mans and 353 epics for T11 and US/EU didn't. We had the shock curve and they didn't. So they may be progressing faster than US/EU. Another thing i have to revise it the weighted content times. The MoP release date will drop soon, that's a BIG player here. If its far out I would put a massive drop of 10% off prime median avg in the projection. If its looking fairly close....or at least public beta, my current median may be fine. (Subs will stay up for beta access). I'd like to hear your thoughts as to why.... maybe i'm missing something.

In addition to that, the last chart, basing the worth as a subscriber, from a math stand point is fine, but from a business stand point, is way off. 1) American and EU customers pay more than Chinese, therefore the value of an American or EU subscriber is more valuable. 2) The cost of employing people, maintaining servers, etc doesn't change unless those actually do. If you have 10 employees and 10 servers, the cost to run and pay those is the same if you have 100 customers or 200. Considering most of the work isn't physical that is. While a single customer is a drop in the hat, that $15 a month they pay goes towards paying for employees, servers, etc. $15 not paid comes out of someone's pocket, which is felt from a business perspective.
.

The last chart was strictly for a laugh..... and was just pure math percentage of total subs. Its the ants from an airplane theory... show how small just one account is in the sea. These numbers get so large its easy to lose scope. It wasn't a smack down or anything.

---------- Post added 2012-03-05 at 09:04 PM ----------

2012 Prediciton Data Coming soon.