Once you've decided to localize your software or corporate documents, there are many error sources along the path to a translated product. You've devoted time to building an amazing product, and translations need to resonate with your product, company, and, most importantly, your local customers. To help you avoid poor translations, we came up with these 6 common reasons for poor translation quality and how to avoid them in your localization project. Follow this checklist and deliver your product and new features to users worldwide efficiently and with high-quality text.
#1 Complicated and poorly written source content
The first reason for poor translation quality can often be found in the source content itself. Your source text's low quality and mistakes are multiplied through translation. Here are a few simple tips to increase the source content quality and reduce future issues.
Avoid long sentences
Val Swisher, CEO of Content Rules, has been long into the topic. She calls it "the scourge of the long sentence." 95 words is the all-time high Val found in a sentence. Here's why you should avoid long semantic constructions and focus on short ones:
- Long sentences are hard for your audience to understand, not only in English.
- They are also hard to understand and nearly impossible to translate correctly for your translators.
- Focus on short and catchy sentences. They make your content easy to understand and engaging too.
- The users can not keep the writer's thoughts from start to end, especially if unfamiliar with the topic.
Double-check your source text for grammar errors
Whenever you create source content for translations, double-check before passing it on to your translators. Typos like "your" instead of "you're" sometimes occur without noticing them. It's best to ask a second native speaker to check the text for spelling and grammar mistakes. The fresh eye raises the chance of finding minor errors and typos.
Keep a wary eye on the tone of voice
The tone of the content has become casual and friendly. Companies tend to use Internet slang when creating content. This very friendly and plain tone of voice is mostly grammatically incorrect. Sentence fragments and specific punctuation that do not exist in other languages lead to translation issues. Use slang less in your content and focus on grammatically correct texts to increase your translation quality. The best way will be to create a single style guide for all the teams communicating with your customers on the web and live. Make your team speak one tone of voice to build a consistent image of your company.
#2 Manual translation workflow
A common source of poor translation quality is a manual translation workflow. Traditional translation agencies often rely on those manual processes. Companies need a responsible person who identifies and forwards the translatable text to translators. Hence, source content must be copied from the product and pasted into a simple Word or Excel sheet. The same applies to translations, just the other way around. This manual process contains numerous error sources:
- Copy/paste incomplete texts.
- Forget to copy/paste entire phrases.
- Lack of context information
- Traditional translation agencies usually work with offline translation tools to work out translations. The translator copies and pastes the final translation into the document. He might accidentally paste the text in the spreadsheet's wrong cell, resulting in duplicate and inaccurate translations.
- When translators and product managers work on a shared spreadsheet, a text/translation status is not marked. Hence, the product manager might paste a text into the product that still needs a final review.
- Manual processes are time-consuming. A professional translator typically translates 2,000 to 2,500 words a day. All the additional copy/paste work takes time. If deadlines are too close, translators must work faster, and the translation quality might suffer. Plan translations according to release deadlines to give translators enough time to work out high-quality translations.
The best way to avoid such troubles is to entirely avoid Excel file usage from the translation process.
Automate your translation workflow
Avoid manual steps in the process by using translation management systems. Lingohub's most powerful advantage is the automation of the translation workflow and providing continuous localization. That means that only the translation work is carried out by humans while file transferring, quality checking, etc., can be done automatically with the system. The mess, as mentioned above, of documents is eliminated:
- Translators, developers, and product managers work on the same platform facilitating communication.
- Use statuses to organize your workflow for source and target texts and ensure every team member knows what's up. At Lingohub, we provide 5 statuses that help to keep abreast of the project:
- New - the new text segments.
- Preffiled - for all the texts that were translated by prefill function.
- Draft - for machine-translated texts and temporary translations
- Translated - for finished translations
- Approved - for reviewed and proofread texts
#3 Lack of proper translation tools
The most potent weapons in the translation process are versatile translation tools. They increase translation quality and speed and save costs. Traditional translation agencies work with different offline and online tools that are costly and hard to combine. Freelance translators might work with low-budget assets or freeware that can't keep up with professional ones. Translation management tools like Lingohub provides all gadgets needed to work out high-quality translations. Moreover, if you need an extra boost for the existing team - you can order the text translation or proofreading.
A translation memory is a database that learns from every word you translate or can be trained by your already localized files. The translation automatically saves and adds to your translation memory based on the settings. When translators work on similar texts in the future, the translation memory suggests the best options based on similarity. Forget about translating the same text twice.
A term base or glossary is a centralized compilation of terms. It typically comprises industry-specific, company or product-related, and technical terms. A term base is used to share these terms with employees and external translators. Use approved translations in your term base to avoid duplication of mistakes. Lingohub's term base also acts as quality checks, meaning the system automatically checks terms usage.
Machine translation provides suggestions based on computer algorithms. Make sure to use machine translation the right way! It's a great starting point for translators to work out high-quality translations efficiently. However, don't use machine translation suggestions as final translations, as they're often grammatically incorrect. But machine translation is a perfect tool for pre-translating. Using the Figma plugin, you can quickly check how the new language fits the design templates.
Automatic quality checks
Content written for devices with limited screen size will likely have special restrictions such as text length. You might need line breaks to ensure texts fit your layout correctly. Digital products and software contain variable content that is dynamically replaced. It can be represented as placeholders with special syntax in translation management tools. To avoid the issues with the text string Lingohub allows you to predefine quality checks for the following criteria:
- text length
- line breaks
- HTML code
- duplication With quality checks, the most significant part of issues will be avoided without manual rechecking and managers' control.
#4 Lack of context information External translators don't know your product as hard as you do. They also need information about your company's tone of voice, corporate identity, etc. To avoid poor translation quality, you need to provide translators with so-called context information: guidelines the explanation for industry-specific terms screenshots of your product's interfaces etc.
#5 Low-quality translators
Poor translation quality may be your translator's fault. There are fewer trained people in the job field too. You get what you pay for, and costs vary:
- Depending on the skills of a translator and the quality, you can expect translations to make up $ 0.06 to $ 0.25 per word.
- Professional freelancers, language service providers, and translation management tools usually charge $ 0.14 to $ 0.25 per word or more.
- Basic reviews start at $ 0.02 per word, while professional proofreading makes up $ 0.05 to $ 0.08 per word.
Invest time and money to find the best-fit translators for your project. It might be tempting to go for basic translations and cut costs. But as a result, it can lead to even higher translation costs. Your local customers will recognize poor translations immediately and complain.
Hiring a professional translator afterward and doing all the translation work again doubles the costs. Go with certified translators, and don't waste time and money on poor translations. We at Lingohub aim to provide high-quality translations. Thus, we've partnered up with certified translators in their native language. They're experts in over 40 business areas to deliver translations that resonate with your audience.
#6 Translations publishing without in-country review
Another common error source in translations lies in the workflow. Companies tend to think that texts can be published once they're translated. There's one essential step that needs to be added: review! In-country review by a second native translator is critical to high-quality translations. The evaluation by a second set of trained eyes minimizes the risk of incorrect translations and content enormously.
The translation is no small feat. With the right tools and support, you'll permanently eliminate these common error sources in your localization projects. Convince yourself and try Lingohub for free for 14 days!