Thought Leadership
Artificial intelligence (AI) opens up new opportunities for companies to save resources. In Germany alone, 27% of the companies surveyed already use AI in their workflows and business processes.
In the previous year, this figure was 13.3%. In addition, around 17.5% of companies are planning to integrate AI into their work processes in the coming months, according to a recent survey by the ifo Institute. One tangible use case: voice AI solutions.
Since then, however, companies have lacked reliable figures to justify the cost-benefit advantage of translation AI in companies and the associated investments. Forrester Consulting, on behalf of DeepL, therefore conducted a study for the first time on the return on investment (ROI) of translation AI for companies and analyzed it.
Study on the savings potential of machine translation providers
Hardly any decision-making body in German companies has not yet dealt intensively with the question of whether and how AI can be used effectively in the company. There is great hope for many use cases: everything should become more efficient, with less human effort and higher quality at the same time. That sounds tempting to decision-makers, of course. Too tempting? Because the promises are often not backed up by reliable figures. At least in the translation sector, these figures are now available – and they speak for themselves.
The Forrester study is one of the first analyses to measure and quantify the return on investment of translation AI. The study found that companies were able to reduce the sometimes extremely high amount of time spent on human translations and possible collaboration with translation agencies for internal documents by 90 percent. Translation AI also reduced the total amount of time spent on translations by 50 percent. This ultimately translates into ROI, as professionals can spend their time on truly value-adding tasks: Forrester estimates the ROI of DeepL at 345 percent over three years. In addition, by using AI, companies reduce the number of external translation jobs, which also leads to efficiency gains and cost savings.
But that’s not all: in addition to the classic savings potential, language AI also offers companies the basis for countering increasing globalization through effective localization and offering high-quality content in new markets.
Localization is a growth driver
Despite global shocks, globalization continues to progress. This was demonstrated by DHL and the New York University Stern School of Business, among others, in their DHL Global Connectedness Report 2024 . A record level of globalization was reached in 2022, and this level remained stable in 2023. Germany is ranked as the 13th most globalized country. It is therefore legitimate that local companies are increasingly grappling with the question of how they can localize their content for new international markets in a high-quality manner.
But why should companies localize content at all? Localization is a process that differs from pure translation. In translation, the content of a text is transferred into another language according to its meaning. In contrast, localization is the additional adaptation of texts to the cultural and linguistic differences of the various markets. In addition, localization must take into account not only cultural and linguistic changes, but also structural changes. Successful localization is characterized by the fact that the end product gives the impression of having been developed for the specific requirements of the target market. Idioms and slang also play a major role in localization, as a literal translation often leads to confusion.
The German Chamber of Industry and Commerce (DIHK) found that 20 percent of companies pursuing international goals perceive local content as a major barrier to trade. A study by CSA Research also found that three quarters of respondents are more likely to buy products if the information is provided in the local language. Good translation AI therefore makes localization easier, better and more efficient. This optimizes customer satisfaction and loyalty and sustainably increases the company’s sales – in the home market and beyond.
What needs to be considered when implementing voice AI?
So there are many arguments in favor of implementing voice AI – especially for large international companies. But what needs to be considered for the successful implementation of voice-based AI in order to achieve its full potential? An overview.
1. safety
AI voice companies do not always focus on the security of the external data they use. Companies should therefore pay particular attention to ensuring that the language AI companies have their own server structure, use end-to-end encryption and do not use the data to train the AI. In addition, the use of non-disclosure agreements (NDAs) is particularly useful when dealing with non-EU countries.
2. user-friendliness of voice AI
For the successful implementation of voice AI, it is also important that the user-friendliness of these applications is guaranteed. This is the only way to make the added value of the technology accessible to a wider audience, i.e. ideally to the entire workforce.
3. return on investment (ROI)
By using language AI to translate documents and content, companies can not only save valuable time, but also improve their ROI. Even if the ROI cannot be determined with 100% accuracy, companies can use it to estimate the effectiveness and success of projects.
4. infrastructure and integration
In order to successfully integrate AI tools into the daily processes of companies, particularly powerful computers and storage are required. This is the only way to ensure seamless integration into existing systems and that day-to-day work can continue uninterrupted. In addition, the systems should be regularly monitored and maintained.
Conclusion
In summary, it can be said that Translation AI not only offers companies potential savings, but also makes localization an important tool for meeting the growing demands of globalization. If important factors such as security, user-friendliness and infrastructure are taken into account, nothing stands in the way of the successful implementation of translation AI in companies.
