An indie studio needed LootBound localized for Steam Fest on a tight budget. Here's how we narrowed 26 languages to 19 and delivered on time with run.loc.
ArtDock, an indie studio from Moldova, needed to localize LootBound ahead of Steam Fest with limited time and budget. Together we cut the language list from 26 to 19 based on Steam wishlist data, rebuilt the English version, and split the remaining languages into two quality tiers. Allcorrect delivered all 19 localizations on the agreed date, and the game launched at the festival on time.
- Project snapshot
- What was the challenge?
- How do you choose which languages to localize on a limited budget?
- How do you prepare the source text for localization?
- Can you use different quality levels for different languages?
- How do you run translation and client review at the same time?
- What was the result?
Project snapshot
| Client | ArtDock, indie studio from Moldova |
| Game | LootBound, a turn-based tactical PC game with backpack and loot management |
| Volume | About 25,000 words in Russian |
| Languages | English plus 18 more |
| Process | run.loc, our smart localization workflow that combines automated translation and quality control with professional linguists |
What was the challenge?
ArtDock needed the full game localized in time for Steam Fest, with enough time left to test the build and prepare for the festival, all within a limited budget.
We had already worked with the studio on the LootBound demo. This time the scope was the full game. The client first expected the translation in two weeks. Given the volume and the need to prepare an intermediate English version, we proposed a different schedule and agreed on it before work began.
Agreeing on a realistic timeline up front kept the studio’s testing and festival preparation on track.
How do you choose which languages to localize on a limited budget?
Start with data on where your game already has interest, then balance it against budget and the markets that matter to your team.
The original request covered 26 languages. With ArtDock, we narrowed it to 19, based on the geography of Steam wishlists and the commercial potential of each market. That showed where interest in the game already existed and which languages the budget could support.
We also took the studio’s own priorities into account:
- Romanian mattered to ArtDock as a team from Moldova.
- Hungarian was included because of a noticeable number of wishlists.
Demand data decided most of the list, and the studio’s own priorities shaped the rest.
How do you prepare the source text for localization?
Review the base text first. Errors in an intermediate version can carry over into every target language.
The existing English version of LootBound needed rework. Inaccuracies in it could have spread to the other 18 localizations. We kept the established terminology and retranslated the main text from Russian.
Fixing the base English version once was cheaper than fixing the same mistake in 18 languages.
Can you use different quality levels for different languages?
Yes. You can match the process to each language’s priority and budget, as long as the client agrees to the trade-offs in advance.
We split the work into three groups:
| Group | Approach |
|---|---|
| English | New translation from Russian using the full run.loc process |
| 9 priority languages | Full run.loc process |
| 9 additional languages | run.loc with review of Critical and Major issues only |
For the additional languages, linguists handled the system’s flags for potentially critical and serious errors. These include:
- Distorted meaning or terminology
- Errors that make game mechanics hard to understand
- Potentially offensive wording
Minor language and formatting issues, as well as stylistic preferences, were not part of the review. ArtDock agreed to these limits in advance, knowing that small inaccuracies or less polished phrasing could remain.
A tiered approach let the studio cover 19 languages within budget, with clear expectations about quality in each tier.
How do you run translation and client review at the same time?
Deliver the source-language version in parts, so the client reviews one batch while the next one moves forward.
The agreed schedule allowed 6–8 working days for the Russian-to-English translation and about 7 working days for the remaining languages. To run the stages in parallel, we sent the English text to ArtDock in parts of 7,000–8,000 words. While the studio reviewed one part, the previously approved part moved on to localization.
During the project, keys changed, new strings were added, and existing text was updated. We picked up these changes and applied them to every language version.
Batch delivery and change tracking kept all 19 versions in sync without delaying the schedule.
What was the result?
We delivered the English version and 18 more localizations by the agreed date, and ArtDock presented LootBound at Steam Fest on time.
The studio’s CEO singled out the quality of the English text and thanked the team for meeting the deadline and for being flexible with the budget.
In the first three weeks after release, the game collected more than 1,200 Steam reviews.* While monitoring them, the team found no complaints about localization, and players left positive feedback on the Hungarian and Japanese translations. This reflects early audience reaction and does not replace linguistic quality checks.
The approach began with the client’s goals, priority markets, and quality requirements, and the technology followed from them.
* Review analysis is a mandatory step in every game project we work on.
