Pros & Cons Of Using Google Translate For Business – Milestone

To use Google Translate for business or not — this is the question.

Many companies rely on machine translation or google translate for business to provide content in their customers’ local languages. And while such tools are usually free and can provide you with the needed translation in literally a second, you should stay away from them.

Why?

In the following blog, you will learn everything about machine translation, why you shouldn’t use Machine Translation tools for your business, and the alternatives out there.

What is machine translation/ google translation?

Machine translation refers to the automated translation process of a piece of text from a source language into a target language. There is no human input involved in the process.

Google Translate, on the other hand, is a translation engine, owned and developed by Google, which is free of charge and works with almost all languages in the world.

Source:

In recent years, the quality and especially the fluency of Machine Translation have improved significantly due to the advent of new technologies. One of them is called a neural network.

What are neural networks and how do they help? 

Neural networks, also known as simulated neural networks (SNNs) or artificial neural networks (ANNs), is a technology related to machine learning and artificial technology (AI).

Neural networks are inspired by the behavior of the human brain. Through neural networks, computers can recognize different patterns and, thus, solve certain problems.

Source:

In the last few years, the technology of neural networks is widely used for the development of translation engines. Google Translate also uses this technology to improve the quality of its translations.

How does Machine Translation work?

Neural Machine Translation engines, such as Google Translate, can translate long texts into various languages. But how?

Big data. Data is a must when it comes to technologies related to the field of AI and machine learning.

Source: 

So, if you want to have a Machine Translation engine, you need billions of parallel segments in at least two languages to train your engine. During the training process, neural networks are created through which the engine can translate content.

Neural Machine Translation relies heavily on data with high quality – if there are mistakes in the data, there will be errors in the output as well. 

What’s more, Neural Machine Translation is believed to deal quite well with unknown contexts. Still, there is much left to be desired.

Indeed, Machine Translation has a hard time recognizing idioms or more colloquial expressions. Have a look at this attempt to translate a Bulgarian idiom with Google Translate:

In this example, Google Translate fails to recognize a rather popular colloquial expression in the Bulgarian language. “I feel so sleepy that I can’t see two” should be translated simply to “I’m exhausted”.

In addition, Machine Translation, despite its neural networks, will never be able to think. While it can recognize patterns much more efficiently than humans, it cannot analyze and reflect. It simply gives you output. You can see this in the following example. 

Below you can see an excerpt from the book “Sie nannten sich der Wiener Kreis (“They Called Themselves the Vienna Circle”) by Karl Sigmund. 

Have a look at the phrase “den Ungeraden” which Google Translate translated as “the odd ones.” Although this translation is not completely wrong, it is not the best choice. This German phrase denotes “politically undesirable people,” so “the odd ones” is a little too vague and does not reflect this political nuance of the phrase.

The Black Box

All in all, neural networks and artificial intelligence are believed to be the future, especially when it comes to Machine Translation. 

However, there is one big problem – experts often refer to it as the “black box.” In fact, no one really knows what is actually happening in the neural networks.

Source:

If a translation engine makes the same mistake repeatedly, there is no way to fix it. Neural Machine Translation cannot and probably will never be able to recognize the different tones and meanings in human language. This, in turn, could reduce the quality of the translation drastically.

Why is Google Translate not a tool for business?

As you already know, Google Translate cannot be a reliable translation tool since it does not recognize and differentiate between contexts. Indeed, there are quite a few other reasons why you should not use this tool, especially when it comes to your business.

In this section, you can find 5 reasons why you shouldn’t use Google Translate or any Machine Translation engine for your business.

• Prone to errors.

  Machine Translation engines often translate word-for-word, which can lead to incorrect and ridiculous translations. Indeed, the last thing you want is to have sloppy translated content on your webpage, which will send the wrong message to your customers.

  Accurate translation is crucial when it comes to attracting new international customers. It can build trust and loyalty since customers feel appreciated and satisfied by providing them with quality content in their native language. However, poorly translated return policies and instructions can cause misunderstandings and dissatisfaction in your customers.

Also read: 10 Marketing Translation Mistakes by Big Brands

• No cultural awareness.

  Informative business materials such as brochures, newsletters, and advertisements need to be transcreated, rather than just translated. This means that the content should be adapted to the cultural norms and beliefs of your target audience. In other words, what works in one region might be offensive and unacceptable in another. For example, ‘bico’ means a small, casual job in Brazilian Portuguese, but in Portugal, the same word means ‘blowjob.’

  Machine Translation will never be able to differentiate between these two words. Culture is a complex system of beliefs and norms which cannot be built into the software. Deciding what is appropriate in a given culture requires thinking and even personal experience – as we know, machines don’t have that and will never be able to.

Source:

Also read: Localizing Colors for Different Cultures

• Privacy issues.

  As a public engine, Google Translate owns any text you enter into the tool. Indeed, this service is only seemingly free of charge — you get to pay with your data. If you want to translate highly sensitive content that contains personal information, you should stay away from Google Translate or any other online tool.

• Lack of specialization.

  Specialized translations, such as legal and marketing documents should be created and translated according to particular certifications, such as import-export licenses, quality certificates, ISO, etc. What’s more, such documents contain subject-specific jargon which Google Translate often cannot recognize and, thus, fail to deliver quality output.

  Having expertise in a given field means not only having the knowledge, but also the experience. Specialization is crucial for high-quality translations since it ensures that the linguist can choose the correct terms and errors are unlikely to occur in the final output.

• Lack of quality control. In order to improve the quality of its tool, Google allows users to correct or approve translations. However, one can never be sure that a user has approved the right translation for a particular term or context. This lack of quality control often can lead to mistakes that can ruin your brand’s image.

Also read: 7 Reasons to Work With a Translation Company on Your Next Project

Examples of companies that failed at translation

Quite a few big companies out there have failed at translation. We have gathered a few examples to show you what you should not do!

McDonald’s

McDonald’s has over 40,000 restaurants worldwide. However, despite its insane popularity, the franchise made a rather silly translation mistake when launching in France. Their signature Big Mac burger was translated to “Gros Mec,” meaning “Big Pimp” in French.

Source:

Amazon

Amazon didn’t have a good start in Sweden either! For instance, products featuring cats often contained the word “pussy” which was then translated as a vulgar Swedish word for ‘vagina.’ What’s more, Nintendo Switch games were described as suitable for the Nintendo Circuit Breaker. Funny enough, a collection of WWII-era Russian infantry figures was described in Swedish as “Russian toddlers.”

Indeed, the Amazon launch in Sweden was a real disaster due to machine translation!

Source:

Mercedes Benz

Even Mercedes Benz had troubles in China! When they introduced their brand on the market, the German automobile brand used the name ‘Bensi,’ which in Mandarin happens to mean ‘rush to die.’

They soon changed it to ‘Benchi,’ meaning ’run quickly as if flying.’ Much better, isn’t it!

Coca-Cola

It’s one thing to translate your slogan incorrectly, but Coca-Cola had a much more serious translation problem to deal with. Translated into Chinese, Coca-Cola turns into ‘Ke-Kou-Ke-La,’ which can mean either “bite the wax tadpole” or “female horse stuffed with wax,” depending on the dialect.

Source: