Let's find out.
First of all, our total sales to date, across all PC portals for the original Defender's Quest are:
$656,000 USD gross
195,000 units sold
Next, our monthly sales on Steam:
(No y-axis label because Steam doesn't like us being too specific about exact sales on
Wow, another great article about business and promotion, respect!
Iāve done some professional translating in the game business, too. I like games and Iām aware of the language barrier.
Iād like to add some ideas to your post.
As you said developers should finish the translation before releasing their game, Iād add that they should prepare localised press releases, too. If you speak of promotion then localised press releases are very much appreciated by several groups. Even professionals in (online) journalism would better be working with a text in their mother tongue because it allows faster workflow. Then there is something like an independent media in the internet which contains of bloggers, youtubers, and so on. Some of them are limited to their mother tongue, others might try to understand a foreign language but are too young to know all about English or other languages. Some of the most influential media people are teenagers, and we cannot expect them to already speak several languages fluently. And if you can, try to get to know the foreign market, perhaps one of your translators you acquired can tell you a bit about the media landscape in his or her country.
What you should avoid is Google translate or similar services, as the results are not worth talking and you get mocked as a developer for your ācheapā try of mimicking someones mother tongue.
On the other hand, you at least should interrogate your professional translator on the topic of gaming before he or she starts working. If you think he or she does not have an affinity for games then itās best to not hire him/her. It means, do not act like Microsoft and hire people for simultaneous interpreting their E3 live stream, who do not know what a video game is. This results in negative feedback by your potential customers. This is because there are numerous translations of a single word or sentence and if someone is not aware of the context, they will fail in translating it. It would be best if your translator can also playtest the game during his or her work and give you feedback. It even might resemble some obstacles within the original version.
45.000 words is a lot of text. ^^
Iām glad you think that translating your game was worth it, overall.
Thereās one other big thing you didnāt mention (or didnāt know) - Steam counts gift purchases as sales in the region it was purchased from, so the big spike in russian sales on 90% off is probably other region players asking their russian friends to buy and āgiftā the game to them. I know several friends who do this regularly for 0/minimal personal overhead. Or thereās resellers who buy in bulk from the cheap russian region and sell them after the sale ended.
This practice is mostly prohibited on big AAA releases by regionlocking russian store games but it still works on smaller games/devs who do not have a big publisher. Example: http://f6.s.qip.ru/2Q3rLe5K.png
Thatās an interesting point, Iāll have to look at the data to see. Steam breaks things down by purchases and by key redemptions (we used only actual purchases in our data set), are you saying that gift purchases from Russia might still show up as actual purchases and not key redemptions?
I donāt know for certain but it would make sense that gifts system is not a key redemption thing. Every gift you buy is initially stored in your inventory as an item even if you choose āsend directly to that personā on checkout. You never see a key. I think key redemption only counts for keys the developer generated manually themselves.
Another point of information, although it might be completely irrelevant - up until recently redeeming a key on your account did not remove the game from your wishlist, while activating the game from the gift (accepting a gift or unpacking it in your inventory) did. Now I know it can be just an oversight in the wishlist system but itās likely that there are two different systems of redeeming games (gift/key) and they just didnāt do the check on one of them.
And another point of info - redeeming a key creates a different entry in your steam account log, example: http://f6.s.qip.ru/2Q3rLe5P.png
āRetailā entries are me activating games from bundles (as keys and as their openID key activation system)
āSteam Storeā entries are me buying the games directly from the store
āGift/Guest passā entry is me getting a gift from a friend.
The wording on that last type, āGift/Guest Passā makes me think that the system was initially designed for handling guest passes and expanded later for full games.
I know it doesnāt prove anything but itās something to think about.
Definitely some good information. My stats explicitly break things down by āsteam unitsā (ie sales) and āretail unitsā (key redemptions).
The question is, if Alice is Russian and Bob is American, and Alice buys a game for Bob, thatās one purchase ā how does that get logged? Is it tracked as a āsteam unitā when Alice purchases it? I assume it would be but Iād have to ask Steam to know for sure.
Bottom line: this would be a lot easier to draw conclusions from with some cleaner stats from Steam.
Well, the other tab on the account page (STORE TRANSACTIONS) has two different tables, one called āRecent Steam Store Transactions for Your Accountā and other called āRecent Steam Store Gift Transactionsā
Hereās how all the gift events look: f6.s.qip.ru/2Q3rLe5R.png (doesnāt let me post a link to āthis hostā anymore)
And hereās how āfor your accountā events look: f6.s.qip.ru/2Q3rLe5S.png
In both cases the event is called āPurchaseā when I spend money in Steam Store. But yeah, youāll have to check with Valve for definite information, Iām just a user who draws conclusions from what few data I have access to. And I have to tell you, a lot of steam game collectors have a russian friend just to buy them games for cheaper.
Yeah, thanks for that. I think what we really need from Steam is this information:
I suppose I could have put in some tracking information myself to see if people were using the localizations. Maybe next time.
So um, hereās my tl;dr version:
Localizing is always good (at least not bad), specially when itās a clean fan translation. The fan makes it, you add it to the game, the fan gets credited, and everyoneās happy.
Defenderās Quest got many sale-boosts and promotions with the new localizing, as seen through the charts in the original post. It is also noticeable that a few months after a new localization was added, the sales were still considerably higher than no localized at all.
So sale-wise, a new language is always good, it helps alot people that donāt know English very well, and those people start buying the game (assuming they played the demo and didnāt understand any of the languages). And the earlier a localization is added, the more long-term sales are gotten (from the corresponding country).
And with this you learn that adding multiple language support to your games upon launch will boost their sales by a very high amount (may not be noticeably big in a single month, but will make a difference over time).
The End.
So, an even shorter form of that:
Localizing is never bad, it may not be good but it is never bad, so itās always worth a shot.
Localizing your game boosts its sales.
The earlier you localize, the higher the sales/promotions will be.
EDIT: Oh yeah and make sure localizing doesnāt cost too much, fan translation is always good too.
Pretty much - assuming of course:
The cost of setting up a localizable game engine is not enormously large.
Oh yeah, I forgot to add that. I edited my post regarding that.
Iām korean Translator, Team.SM. And Korean is still need some fix 
Sorry about that!
If you email me a list of problems and or changes, Iāll gladly include them in the next update (Thanks for all your hard work by the way!)
Ah, what need fix things are just translation error. not your error
i think iāll send you fixed file up to september. we are working now!
Thanks 
Pity Iām from a small country, so my localication mostly gave you an extra work with the graphs above :).
Thank you, Lars, and congrats on your game.
Your post is most interesting. I wish more companies would offer similar research and results on localisation 
All the very best
Miguel
http://www.roehampton.ac.uk/staff/Miguel-Angel-Bernal-Merino/
A very nice post, lots of good information in there!
You should look into the CrowdIn portal, to better support your fan based localization, weāre using it here and have had a great experience with it so far - http://translate.z2.com/
Thanks for the post!
Andy J
Thanks so much for sharing with us,
Itās very interesting to see how some markets will go that extra mile to have even DIY translations the minute you guys enabled character settings. Where others, like Spanish (my native language), will still be waiting for the translation to even start engaging. We had a customer who decided to publish their expected language release dates (as they couldnāt afford to launch simultaneously) and they got some interesting mixed results depending on the country.
Loved to also hear your experience with both Volunteer (crowd-sourcing) and traditional language professionals.
I am currently evaluating a workflow that will allow for automated delivery of the translator kit to volunteers and later add professional QA stages. This will help improving time and quality to market, still engaging with the volunteer community.
Congrats on the article!!
It might also be worth looking in the education of that land when it comes to foreign languages.
For example, in Germany (where I am from) nearly everyone learns english as their second language and as far as I can tell, weāre pretty good at it. So Localization is not as much an issue here because most people already understand english very well and some people (like me) donāt even want to read or listen to any localization because they prefer the original (in this case, english).
So a german localization in comparison wonāt attract that much additional people, because a lot of them are already playing the game in english and might even prefer the english version.
And you can even see that with your data and you said it yourself that Germany was your second-highest earning country, most likely because of the fact that a lot of us here (at least the target audience) is pretty fluent in english already, which would result in a localization not giving as much of a bosst as in other countrys (with an x1.17 increase germany had the lowest).
And thatās just my 2 cents for that. (And I donāt mean to say that localization for germany is a bad thing, far from it. Your data already shows that this is not the case. Just wanted to add my thoughts on that ^^).
