SteamSpy - Approximate LTD sales for every game on Steam (Updated Daily)

Hey Galyonkin, any chance you have the sales data archived for Windborne?

The developers have just abandoned development and pulled it from steam claiming they've invested 10x more then they made in sales. I'm just curious to see how poorly it actually did.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

Ì assume you are happy with the sales then? :D Congratulations man, it's great when developers of good games achieve success.
 
Like on Reddit? It still wouldn't be accurate, I'm afraid.
You could always try that Wilson score algorithm that steamDB uses. It's essentially a confidence interval for the "true" percent of favorable reviews on a binomial distribution (there are only two options, recommended or not recommended).
 
neat
AcaICjN.png

The sales for XIII -> XIII-2 are half. If they would fix the frame pacing on both I wouldn't have felt burned by XIII and picked up XIII-2.
 
You could always try that Wilson score algorithm that steamDB uses. It's essentially a confidence interval for the "true" percent of favorable reviews on a binomial distribution (there are only two options, recommended or not recommended).

Yeah, I thought about that. But my sorting now relies on DataTables.js and a lot of stuff there is hardcoded. I can't change sorting algorithm without changing the source :(
 
Hey Galyonkin, any chance you have the sales data archived for Windborne?

The developers have just abandoned development and pulled it from steam claiming they've invested 10x more then they made in sales. I'm just curious to see how poorly it actually did.

It's all there :) http://steamspy.com/app/221560

When Steam removes a game I remove it from the search, but all collected stats are still there.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

Dood, you made "Cook, Serve, Delicious!"? Awesome! :D
 
Interesting to see that the owners of Massive Chalice have jumped with almost 50%, despite the game not being a daily or flash sale yet.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

Congrats on the sales, I was one of the people who bought the game but I haven't gotten a chance to try it yet.
 
There seems to be a trend of more ambitious sequels to popular indie games releasing to an increased price point and seeing poor sales as a result. Look at the drop for Defense Grid 2, Audiosurf 2, Legend of Grimlock, etc. Even if you discount bundling the drops are large.

Indie devs are counting on a brand loyalty that doesn't seem to exist.

So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

I bought Cook, Serve, Delicious to support you, even though my backlog is immense.
 
Absolutely insane how much Steam sales take in. If NPD tracked digital quite a few of the sales biggest sellers would crack the top 10 quite regularly.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

I always wonder how they determine the games that are featured in the monstergame.

Was that an opt-in thing? is it random where and how you are placed? There was allready 2 o3 games we didnt unlock, the dev of those games must be pissed.

(If you can talk about that)
 
Maybe it's just me, but I feel like the summer sale is only a fraction of the size of the Winter sale as well.

Yep, I think the winter sale is the "big one".

But what would be really interesting is to know what % of sales are done in these special days vs the rest of the year. Are Steam sales too concentrated in the special sales? The impression is that a fair number of people wait until the sales to buy lots of games.
 
Yep, I think the winter sale is the "big one".

But what would be really interesting is to know what % of sales are done in these special days vs the rest of the year. Are Steam sales too concentrated in the special sales? The impression is that a fair number of people wait until the sales to buy lots of games.

True, but the tail coming off sales I would guess makes up for that.
 
More than 30 million copies of games sold so far. Jesus Christ.

Even more.

NOTE: Only games with significant changes in their owners were included in this table.
Actual number of total copies sold during the sale is bigger than shown below, because many games sold less than 10,000 copies and weren't accounted for!
 
The great thing is that those sales are spread out over a huge variety of games. Steam Spy shows more than 700 games that achieved 10,000 sales or more.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D
I finally bought it. The price was too good man.
 
It would be cool if there was an info about every game if it steamworks enabled or if other versions exist, that are sold in retail or in other stores whithout mandatory steam registration.
And a minor thing: Textboxes should have a small border, it's really hard to find the search for example.
 
I'd be interesting to know if coming out of Early Access does anything for the sales of a game. I feel like a lot of EA games botch their actual launch.

For example I bought Vertiginous Golf a couple of months ago and thought it was pretty sweet, heard a couple of people talking about it etc. Fast forward to today, I completely forgot about the game until today, learned it came out of early access over a month ago and haven't heard anybody talking about it.
Might be cause it's probably a fairly unknown game though.
 
So my game went on sale for 24 hours two days ago as part of the Monster Sale unlocks (from 75% to 90%), and the stats on the SteamSpy Sale Page site are about ~15,000 lower than what it actually sold, which is within the ~24,000 margin of error. Anyways thought that was interesting. I love Steam. :D

I bought it!

Will be playing this week ;)
 
I'd be interesting to know if coming out of Early Access does anything for the sales of a game. I feel like a lot of EA games botch their actual launch.

For example I bought Vertiginous Golf a couple of months ago and thought it was pretty sweet, heard a couple of people talking about it etc. Fast forward to today, I completely forgot about the game until today, learned it came out of early access over a month ago and haven't heard anybody talking about it.
Might be cause it's probably a fairly unknown game though.

Galyonkin wrote an article that touches a bit of it. Seems like the games sells more during the first period of EA than when they release the complete version. https://medium.com/@galyonkin/some-things-you-should-know-about-steam-5eaffcf33218
 
It would be cool if there was an info about every game if it steamworks enabled or if other versions exist, that are sold in retail or in other stores whithout mandatory steam registration.
And a minor thing: Textboxes should have a small border, it's really hard to find the search for example.

That first request doesn't sound like it can be automated so I really doubt it'll happen.

And search does have a small border.
 
Thanks everyone!

I always wonder how they determine the games that are featured in the monstergame.

Was that an opt-in thing? is it random where and how you are placed? There was allready 2 o3 games we didnt unlock, the dev of those games must be pissed.

(If you can talk about that)

It was an opt-in thing, in that I opted into the 90% discount. However I didn't know if I was selected ahead of time, nor did I know which category I'd be in if I did. So it's always a fun/stressful time for me during the sales, hah.
 
Galyonkin wrote an article that touches a bit of it. Seems like the games sells more during the first period of EA than when they release the complete version. https://medium.com/@galyonkin/some-things-you-should-know-about-steam-5eaffcf33218

Speaking of the article and the somewhat low number of average total sales per game...
How does the situation change if you disregard the 2015 (and last few months of 2014) games? It's not like some of them had a lot of time to catch up to Half-Life 2/etc sales...
 
Galyonkin, any idea why some games have had reasonably drastic drop in sales/owners on the Summer Sale page? For example, Project CARS used to show a bit over 191K yesterday and is now showing 175K, cutting the sales amount in over half. It's also still showing 185K on the normal steamspy page with the multiple day sample. I don't have the numbers anymore, but I noticed the same thing on some games by my friends.
 
Galyonkin, any idea why some games have had reasonably drastic drop in sales/owners on the Summer Sale page? For example, Project CARS used to show a bit over 191K yesterday and is now showing 175K, cutting the sales amount in over half. It's also still showing 185K on the normal steamspy page with the multiple day sample. I don't have the numbers anymore, but I noticed the same thing on some games by my friends.
The sales page uses 1 day samples, so a higher degree of variance day-to-day should be expected.

The sales amount is a first order differentiation of a value with higher variance, so it's even less reliable :P
 
The sales page uses 1 day samples, so a higher degree of variance day-to-day should be expected.

The sales amount is a first order differentiation of a value with higher variance, so it's even less reliable :P
Yeah naturally, though it still seemed unusually large of a deviation to me. I should probably ask around to see how accurate the data might've been and just generally see how well people have done this Summer Sale.
 
Over 40 Million games were sold. Not bad.

Galyonkin is there an information available how much new owners (or copies sold) Steam had from new year till the summer sale? So that we know how much percent of games sold were in this Sale in reagrds to YTD
 
Over 40 Million games were sold. Not bad.

Galyonkin is there an information available how much new owners (or copies sold) Steam had from new year till the summer sale? So that we know how much percent of games sold were in this Sale in reagrds to YTD

I'm afraid, not - I've only started to gather information in the late March this year;
 
only if there is a huge drop of ownership. And it must be a really HUGE drop so that it statistically matters and doesnt get nullified by new buyers.
 
only if there is a huge drop of ownership. And it must be a really HUGE drop so that it statistically matters and doesnt get nullified by new buyers.

Another note on that, since it's the first day of data, you will only see starting ownership; the result of which refunds have already begun. It'll be tough to extrapolate unless there is a delay in the refunds processing.
 
Top Bottom