Comparison of Translation Methods: Machine vs. Human Translation

Do you hesitate when deciding on whether you should get a human translation because machine translation services seem more economical and cost-effective?

BlacKKite compares the performance of machine translation engines with that of professional translators so you can make an informed decision about which translation solution is right for your application.

Machine vs. Human Translation Comparison Based on:

Translation Quality

Translation quality is usually hard to assess. How do you determine whether your translations are best represented by an automated translation service or a human translator?

I could list the benefits of a 100% human translation, but you would much better judge the machine translation of the leading translation service DeepL or Large Language Model ChatGPT yourself in an A-B comparison to the showcase translation by BlacKKite (Dutch language article).

to DeepL comparison

to ChatGPT comparison

Rates & Privacy Concerns

The fees for a machine translation service are charged as a fixed monthly rate (DeepL & ChatGPT) or per translated character (Google & Microsoft). Compared to the translation rates of BlacKKite, machine translation seems to be a more economical choice, but keep in mind that it still has to be proofread by a native Dutch speaker.

Is Machine Translation Free to Use?

Free of charge translation is also available when you use a (limited) web service and are willing to surrender your personal data to help train and improve the translation algorithm. Your privacy is the currency you pay with. In contrast, BlacKKite will always process your personal data in accordance with GDPR guidelines.

Do you opt for commonly used phrasing or prefer a distinctive wording that contributes to a higher SEO score?

Technical translator Erwin van Wouw

Distinctive Capability

The result of a machine translation service trained with a bilingual corpus continues to refer to that very corpus. One could argue that the translation computer gives a cleverly formed stupid answer to a complicated question. For translators, the answer is not necessarily black or white. They know when to avoid or rather emphasize gray areas.

I use terminology very consistently, for example, when machine parts in a maintenance manual refer to a drawing. In contrast, in this blog post about the differences between the translations of translation engines and human translators, I use a variety of synonyms for the term machine translation to keep readers engaged.

Chart with average number of words that deviate or are identical

Legend pie chart

On average, 57.4% of the words in my Dutch showcase translation differ to a greater or lesser extent. This originality rate says little about the quality of the translation, of course, but it does demonstrate that buyers of translation services have something to choose from. Do they opt for commonly used phrasing or prefer a distinctive wording that contributes to a higher SEO score?

Google itself clarified that when it comes to SEO, content run through a translation service such as Google Translate is a no-no.

Gino DiƱo on

Translation Length

Purchasers of technical translations sometimes restrict the number of characters used for the translation relative to the original, e.g. when text becomes invisible if the character count exceeds a certain threshold. As the Dutch language often requires more characters for the same message than the English language, the translator should keep track of text expansion and delete excess characters.

Deleting is one of the translator's standard tools that is missing from the toolbox of the translation engine. Only a machine translation pre-editor or post-editor can truncate a computer translation. The word and character density of my Dutch showcase project are on average just a little below those of Philips' original manual. The automated translation tool DeepL uses on average a good ten characters more than the original manual, and ChatGPT even twenty characters.

Bar chart average character count: Philips 122.9, DeepL 133.9 and Black Kite 122.8 characters.

Translation Turnaround Time

Machine translations excel in speed. The translation of your documentation will appear on the screen in no-time. With an average daily translation capacity of 2,500 words, I cannot match that, not even with the assistance of the most advanced CAT tool. However, keep in mind that a computer translation still needs to be proofread by a native speaker of Dutch before the text is suitable for publication.

Moreover, speed is not necessarily something positive. A translation can benefit from the translator contemplating it for a while to further ponder the matter at hand. A fresh look can prevent mistakes and solve creative blockages!

Machine or Human Translation?

The main differences between human translation and machine translation are the lack of quality assurance and privacy protection in machine translation. For the time being, the human translator remains the leading force in high-quality tailor-made translations with respect for privacy.

Machine Translation Application

When good is just good enough, the text needs to be translated quickly and cheaply, does not contain privacy-sensitive information, and the result will be post-edited by a proofreader, a machine translation could be a sound alternative.

Decide for yourself and order a Dutch translation now or request a free quote.

request a quote

Author bio

English to Dutch translator/editor Erwin van Wouw

Erwin van Wouw (1973) is a native Dutch technician turned translator based in Utrecht, the Netherlands. With over 16 years of experience as a professional Dutch translator and proofreader, he assists businesses in transforming their texts into impeccably written and smoothly flowing Dutch documents.