How to put keywords in machine-translated content
At locatheart, we prioritise quality, while also recognising our clients’ need for optimisation. This includes cost optimisation, which is what probably crossed your mind, but also search engine optimisation (SEO). Lately, these two strategies have been at odds with each other.
Companies striving to cut costs turned to machine translation, which meant their content was deprived of keywords – the foundation of SEO. Initially, our clients showed little interest in this solution due to the shortcomings of machine translation. But now, as it’s beginning to bring satisfactory results, the interest in using it is growing. And so are concerns connected to optimising machine-translated content by enriching it with keywords.
But is it possible? Yes!
Thanks, Captain Obvious!
Despite being considered expert content, the majority of articles available online offer outdated or ineffective solutions.
The first “pro tip” I found suggested preparing the source text in a way that, during the translation process, the source phrases will turn into keywords.
This approach is based on back-translation and it’s certainly not an efficient way to utilise contemporary technology. Think about the tremendous effort required to prepare multiple language versions. According to this approach, you’d have to create a separate source text for each target language. That’s ridiculous!
The second “golden” piece of advice was to format the content in a Google-friendly way. Thanks, Captain Obvious. This is a best practice recommended for everyone, everywhere. And even though it’s certainly something that should be done, it’s not enough to stand out in the crowded internet space. What’s more, it falls short of being a state-of-the-art method of implementing keywords into machine-translated content.
The third “expert tip” I came across was to make sure that content undergoes thorough human review. This is actually a great way to implement keywords in the text and ensure proper quality. However, it doesn’t align well with the goal of achieving satisfactory cost optimisation.
What can locatheart do to help? First, we have to make sure that businesses see and understand the impact that language services have on their sales. Then, we can support clients in finding optimal solutions – achieving the best possible results within their available budget.
How to optimise for savings and online visibility
We have developed a process that allows our clients to leverage machine translation while simultaneously optimising their translated content. We call this a semi-automated optimisation process.
Why “automated”? Because once we supply the software with all the necessary files and resources, it automatically handles the tasks – similarly to machine translation.
Why “semi-automated”? Because tasks like keyword research and pairing keywords with source phrases to create a glossary require the expertise of specialists.
So, how does this work?
At the machine translation stage, we continue to use our standard computer-assisted translation (CAT) software, Trados. This approach enables us to leverage translation memories, glossaries, and all of our other usual solutions. But now, we can enhance them with professional add-ons from machine translation engines. Notably, one of these engines excels in supporting glossaries, enabling the placement of glossary terms in the translation in their correct grammatical form, even if they are only in their basic form in the glossary. This feature has inspired us to use this tool for automated implementation of keywords.
How to put keywords in machine-translated content
- Keyword research
- Glossary creation
- Technical preparation
- Upload of files (and the magic begins)
- Final files
The process at a glance
The process begins with identifying relevant keywords used in a target market. Depending on the scale and type of your content, these keywords may be compiled into one general list or categorised. For instance, in the case of a large e-commerce platform, we would suggest at least creating separate lists per category/subcategory.
In this step, our linguists create glossaries, selecting phrases from the source text that are to be translated into the keywords identified in Step 1. Multiple source phrases may ultimately be translated into a single keyword.
For this process to work, the glossary has to be saved in a special format. Additionally, the content has to be exported into translatable files. While this step might sound straightforward, it can prove challenging and may require our support or even involvement with the client’s IT teams. This is when clients often learn the hard way how important it is to prepare their websites and stores for localisation.
At this stage, it’s mainly the machine that’s working. Our team oversees the process and uploads subsequent files as required.
The end-result is ready for import to the final destination – the client’s website or store.
For companies dealing with large volumes of content, machine translation has been a game-changer in terms of cost optimisation. With this advancement, they can now venture into international markets with confidence, knowing that their machine-translated content will reach their target customers.
Does it work for all markets?
While there are solutions available for a variety of language pairs, we’ve seen best results with:
Need solutions for a different language? Worry not, technology is advancing day by day.
With our approach, you get the best of both worlds – the efficiency of machine translation combined with the expert guidance of skilled linguists and SEO specialists. This ensures that your content is optimised for maximum impact in international markets.
We recognise the importance of cost optimisation for your business. Our tailored solutions enable you to achieve substantial cost savings without compromising on efficiency – the locatheart team will closely collaborate with you to strike the perfect balance between cost efficiency and content effectiveness.
Leave a Reply