How to Measure the ROI of Translation and Localisation

This guide is for e-commerce managers, marketing directors and finance teams who need to justify, or review, spending on translation and localisation. It sets out a simple ROI formula, the costs and returns to include, the metrics to track for each language, and how to design a test that gives a credible answer.

Localisation is easy to underrate because its costs are visible on an invoice while its benefits are spread across traffic, conversion, returns and loyalty. A clear measurement plan, agreed before launch, solves most of that problem.

What is localisation ROI and how is it calculated?

Localisation ROI (return on investment) is the profit generated by a language version, relative to what it cost to create and maintain. It is usually expressed as a percentage over a fixed period, such as 12 months.

The basic formula is:

ROI = (incremental gross profit − total localisation cost) ÷ total localisation cost

"Incremental" is the key word. You count only the profit that would not have happened without the language version, which is why a baseline or a control group matters. Use gross profit, not revenue, so that product costs, delivery and payment fees are taken into account.

Which costs should a localisation ROI calculation include?

A credible ROI calculation includes every cost of running the language version, not only the translation invoice. Leaving out ongoing costs makes the first year look better than later years.

• Language work. Translation, transcreation, review and any machine translation post-editing.
• Technical work. Internationalisation of the shop or app, plugins, file handling and testing.
• Local SEO and marketing. Keyword research, local ads and campaign adaptation.
• Legal and compliance. Local review of terms, returns pages and product safety information.
• Customer service. Support in the local language, by email, chat or phone.
• Maintenance. Translating new products, updates and seasonal campaigns.

For a breakdown of how creative work is priced, see how transcreation is priced. For word-based translation, see translation cost per word in the UK.

Which KPIs show whether translation is paying off?

The most useful KPIs compare each language version with its own baseline and with the source-language site. Revenue alone hides the reasons behind a result.

Repeat purchases deserve a place in the calculation. In CSA Research's survey of 8,709 consumers in 29 countries, 75% of respondents said they are more likely to purchase the same brand again if customer care is in their language.

How do you set up tracking for each language version?

Tracking works best when each language version has its own URL structure, so analytics and search tools can report on it separately. This is also what search engines recommend.

Google advises that it "recommends using different URLs for each language version of a page". It also says it does not use "any code-level language information such as lang attributes" to work out a page's language. A structure such as /de/, /fr/ and /it/, or separate country domains, lets you segment traffic, conversions and Search Console data by language. Our guide to website localisation explains the main structural options, and multilingual SEO covers how to make each version rank.

Before launch, record at least three months of baseline data for visitors from the target country: sessions, conversion rate, order value and returns. Without a baseline, any change after launch is guesswork.

How do you run a fair test of localisation?

A fair localisation test isolates the effect of language from other changes, such as new ads, price changes or seasonality. The cleaner the test, the more convincing the result for finance teams.

1. Pick one market first. Choose a market with existing traffic, so you have a baseline.
2. Define the scope. Localise a product group or the whole funnel, and record exactly what was changed.
3. Keep other variables stable. Avoid launching new campaigns or price changes in the same weeks.
4. Use a comparison. Compare with a similar market that remains in English, or with non-localised product groups in the same market.
5. Measure long enough. Organic search effects can take months; allow at least one full sales cycle.
6. Report incremental profit. Apply the formula using gross profit, and show the assumptions.

As a purely hypothetical illustration: if a shop receives 20,000 monthly sessions from Germany and its conversion rate rises from 0.8% to 1.2% after localisation, with an average order value of €60, monthly revenue from those sessions rises from €9,600 to €14,400. The €4,800 difference, converted to gross profit and compared with the full cost, gives the ROI.

What market data helps build the business case?

Market data cannot replace your own measurements, but it helps estimate the opportunity before you invest. Two public sources are particularly useful for e-commerce.

Eurostat reports that in 2025, 78% of EU internet users aged 16 to 74 bought or ordered goods or services online. CSA Research's "Can't Read, Won't Buy" survey found that 40% of consumers will never buy from websites in other languages. Its Chief Research Officer, Dr Donald A. DePalma, put the risk this way:

"Our 2020 findings show that if a company chooses to not localize the buying experience they risk losing 40% or more of the total addressable market." (CSA Research press release)

What are the common mistakes when measuring localisation ROI?

Most localisation ROI calculations go wrong on scope or timing rather than on maths. These are the mistakes we see most often:

• Measuring too early. Organic traffic in a new language builds over months, so a 30-day review undervalues it.
• Counting revenue instead of profit. High delivery or return costs in a market can cancel out the sales uplift.
• Ignoring quality. A cheap, poor translation can depress conversion and inflate returns, so cost per word is not the only variable. The ITI Transcreation Code of Practice warns that if transcreation is rushed, the return from the original campaign idea "can be compromised".
• Blending markets. Reporting all non-English sales together hides a market that is failing.
• Forgetting the support saving. Fewer "what does this mean?" tickets is a real return.

For how campaign tests should be split by language, see localising email and ad copy. For how product content quality affects these KPIs, see product description translation. For the wider picture, read transcreation vs translation.

How TranslationWorks can help with localisation that pays off

TranslationWorks is a UK-based translation agency that translates and localises websites, product catalogues, campaigns and customer service content in more than 150 languages, using native-speaking linguists. Our e-commerce translation services and transcreation services can be phased market by market, which makes it easier to measure the return before expanding. We can also help you decide which content needs creative adaptation and which can be translated, so budget goes where it affects conversion most. TranslationWorks is ISO 17100 certified, ISO 9001 and ISO 27001 registered and has served more than 300 companies. Send your target markets and content scope through our free quote form.