How many languages should you translate your videos into?
Three working strategies: follow your own analytics, follow the money, or cover everything. What each one costs in points and when to move to the next.
Part of the guide: YouTube localization: the complete guide
The right answer depends on what you want to learn, not on your budget. Ten languages test a hypothesis, twenty cover the major markets, one hundred and seventy remove the question entirely. Here are the three strategies, what each costs, and the signals that it is time to move on to the next one.
Why «translate into everything» is not always the best start
Technically, translating a video into every language YouTube accepts is no harder than translating into five: the run is the same, only the points spend differs. But blanket coverage has a hidden cost: you will not learn which markets responded. The lift spreads across dozens of countries, and analytics will say «better» without telling you where.
So for a channel with live statistics we suggest starting narrow and expanding on data. For a back catalogue that has sat untranslated for years, the logic flips: blanket coverage makes sense immediately, because those videos cost nothing more to produce.
Strategy 1: follow your analytics, 5–10 languages
- Open YouTube Studio, Analytics, viewership by country for the last 90 days.
- Write down the countries already watching you without any translations. That is proven demand, not a projection.
- Map each country to one language. Skip regional variants for now.
- Add two or three heavyweights that are missing from your list: Hindi, Spanish, Portuguese.
That usually lands on 5–10 languages. Two months later the analytics show which ones fired, and the second wave is chosen with open eyes.
Strategy 2: follow the money, 6–8 languages
If the goal is revenue rather than reach, the list changes: German, Japanese, French and Korean sit in expensive advertising markets with moderate audiences. Hindi, Indonesian and Portuguese are cheap per view but enormous. The mix most channels land on is half and half: three high-value languages plus three high-volume ones, then compare revenue, not views, after two months.
The full breakdown of what each language brings is in the top 20 languages ranking, and why published CPM numbers disagree is covered in CPM by language.
Strategy 3: blanket coverage
All languages at once make sense in two cases. First: content that works without speech, where a translated title is the only barrier and the cost of a wrong guess is zero. Second: a one-off pass over the archive, where the goal is not an experiment but «make everything findable».
What it costs, in points
One language of title and description costs 1 point. Subtitles in the same language add 1 more. Spend depends on the number of languages, not the length of the video.
| Volume | Titles only | With subtitles | Basic pack covers |
|---|---|---|---|
| Video in 10 languages | 10 points | 20 points | 150–300 videos |
| Video in 20 languages | 20 points | 40 points | 75–150 videos |
| Video in 100 languages | 100 points | 200 points | 15–30 videos |
Packs are one-off, no subscription: €15 for 3,000 points, €40 for 10,000, €80 for 22,000, prices before VAT.
Signals that it is time to expand
- Countries appear in your analytics that were not there before the translation: the mechanism works, add neighbouring languages of the region.
- One language produced a disproportionate lift: cover its market fully, subtitles included.
- The lift is uniformly small everywhere: the topic, not the language list, is the constraint. Expanding the list will not fix that.
One limit stated plainly: translated titles and descriptions work in search and recommendations inside YouTube. The video page still hands Google the original title, so measure the effect in Studio analytics by viewer geography, not in web search rankings.