Operations
By Jason Kumpf, International Advisor · November 4, 2026 · 5 min read

Google Translate, DeepL and Claude have made high-quality translation fast and accessible. Companies are using them to reach international customers in their own language sooner.
Launching in a new language used to take months of translation, review and rework. Today, AI translation tools from Google, DeepL and Anthropic let companies produce strong first drafts of websites, product screens and support content in days.
The result is a faster path to international customers. Teams can test a new market with real localized content, learn quickly and invest further where the response is strongest. Here is how the tools have advanced and how companies are putting them to work.
The business case for localization is well established. In a 2020 study of 8,709 consumers across 29 countries, CSA Research found that 76% prefer to buy products with information in their own language. Forty percent said they will never buy from websites in other languages.
The same study found that 75% of consumers are more likely to buy from a brand again if customer care is offered in their language. In Germany, 57% of respondents said they buy only from local-language websites, the highest share of any country surveyed.
In other words, language is a growth lever. Every new language a company supports widens the audience that can discover, evaluate and adopt its product.
Google has expanded Translate in both reach and quality. In June 2024, it added 110 new languages using its PaLM 2 model, bringing the total to more than 240. The additions cover more than 614 million speakers and include Cantonese, and roughly a quarter of the new languages are African.
In August 2025, Google introduced live, back-and-forth conversation translation in the Translate app across more than 70 languages, first for users in the US, India and Mexico. The feature detects pauses and accents and switches languages automatically.
In December 2025, Google brought Gemini's translation capabilities to Search and the Translate app. Google said the upgrade delivers state-of-the-art text quality on the WMT25 benchmark and handles idioms, slang and local expressions with more natural meaning. A beta feature now delivers real-time speech translation through any pair of headphones.
For global teams, these tools make everyday conversations with prospects, partners and colleagues easier in dozens of languages.
DeepL, founded in Germany in 2017, has focused on translation for companies. In June 2025, it reported more than 200,000 business customers across 228 markets and added Vietnamese and Hebrew to DeepL Translator, with early access to Thai through its API.
In November 2025, DeepL announced support for more than 100 languages, including about 70 added in beta for enterprise customers. The release brought all 24 official EU languages to the platform, plus Hindi, Bengali, Tagalog, Swahili and others.
Two other launches matter for localization teams:
Translation is only part of localization. Marketing copy, onboarding emails and help articles also need the right tone, examples and cultural references. Large language models such as Anthropic's Claude are well suited to that adaptation work.
Anthropic's published benchmarks show strong cross-language performance. On a professionally translated version of the MMLU test, Claude Sonnet 4.5 scored between 96.7% and 98.2% of its English results in Spanish, Brazilian Portuguese, French, German, Japanese, Korean and Hindi.
Anthropic's guidance for multilingual work is practical and simple:
Teams often pair a translation engine with a model like Claude: one produces an accurate draft, and the other refines headlines, calls to action and examples for the local audience.
Not every piece of content needs the same treatment. Companies see the fastest results when they start with the assets that customers touch most often:
Starting with these high-visibility assets gives teams a quick read on demand. Longer documents, such as detailed technical guides, can follow once a market shows traction.
Companies moving quickly into new languages tend to follow a repeatable process:
This approach keeps people focused on judgment and brand quality while AI handles volume. It also makes it practical to test several languages at once instead of one at a time.
AI translation has turned localization from a large project into a series of quick, manageable steps. A company can now publish a credible local version of its website, start conversations with international prospects and support new customers in their own language far sooner than before.
A good place to start is your top three international traffic sources. Localize your homepage, main product page and onboarding emails for those languages, ask a native speaker to review them and watch how visitors respond. The data will show you where to grow next.