Google Translate vs. DeepL comparison

In the application Kudy z nudy, Tips for the weekend, we are now using automated translations for dynamic content. We compared the two most used translation tools.

Tens of thousands of descriptions of places for events to translate

Since 2020, Mapotic has been powered by the Kudy z nudy – Tips for trips application, which displays selected content from the popular tourist server
In addition to the static content, which describes tourist destinations, the application recommends hundreds of new tips for events and trips throughout the Czech Republic every week. To provide an English version for foreign tourists, it was a logical choice to involve an automated translation tool.

Mockup table of translations

Using data connectors, data is synchronized several times a day between the backend of the application and the portal, We have inserted a process into the data bridge, where the text content also flows through the automated translation tool, after which, the newly translated texts are included in the appropriate language branch.

Quality of machine translations = very decent

In the current pilot version, the application uses Google ML (machine learning), however, the recently launched startup DeepL is worth considering.
The operator of the application and we were very curious about the quality of translations – in the tourist industry it is not about highly technical or specialized texts, but there are many topographical and geographical names in the content. Google translate handles it well, but DeepL goes a step further.

Comparison of the quality of translations on a sample of data

Below we performed a smaller test on a sample of existing data from The translations from both tools reach a quality that was hard to imagine a few years ago, however, it seems that DeepL, a tool of German origin, reaches an even higher level in some texts when translating Czech to English than the Google ML, which has been trained for years. After all, you can compare translations yourself in the table below.
Link to the document here.
If you are interested in the article or would like to know the details, do not hesitate to contact us.

Printscreen of translations

Related Posts


Work with maps like never before. Build your structure, visualise data and let people interact. Use maps to get you mission fullfiled.


European structural and investment funds Operation programme Prague – Growth Pole Czech Republic.

In 2020, the company Mapotic, s.r.o received support from the European Structural and Investment Funds, implemented through the Operational Program Prague – Growth Pole of the Czech Republic and drawn through II. call of specialized vouchers (reg. no. CZ.07.1.02 / 0.0 / 0.0. / 16_027 / 0000607). The support was aimed at the participation of the Mapotic s.r.o. at the Prague implementation site in order to increase the competitiveness. The expected output is innovation and launch of other services and products of the company.

© Mapotic 2020 All rights reserved.