What makes a good translation?

For businesses, poor translations can lead to lost deals, damaged reputations, and confusion in critical markets. Miscommunication erodes trust, making it harder to connect with customers and establish a globally renowned brand. Without understanding what makes a good translation, even well-intentioned efforts can fall short, leaving audiences disengaged or worse—misinformed.

This guide will explore the elements that define a good translation, highlight high-quality translation characteristics, and show you how they can boost your business’s international success.

What is a good translation?

A good translation is accurate and fits the audience’s cultural and linguistic expectations. Several standard metrics are helpful when determining the quality of your translated content. With these benchmarks, you can calculate the return on investment of your translated content.

1. Accuracy

Accuracy is the foundation of a good translation. Good translations preserve the meaning of the original text in the target language without adding or omitting details. Inaccuracies in translation strip away the richness of the original text, leading to misunderstandings and misinterpretations.

Machine Translation portals like Google Translate are suitable for simple translations, but they can make mistakes with nuanced content. Conversely, professional translators carefully choose words, phrases, and idioms that convey the correct meaning and tone. If using machine translation, some human oversight might be necessary to ensure translation quality. 

2. Quality evaluation

The localization industry often uses quality evaluation methods like Multidimensional Quality Metrics (MQM), Bilingual Evaluation Understudy (BLEU), and Translation Error Rate (TER), to measure translation quality. Here is what each evaluation framework measures:

• MQM provides a framework for measuring translation quality that focuses on identifying and quantifying errors in a text. MQM breaks down translation errors into specific categories, such as mistranslations or omissions, weights the errors according to the framework, and then calculates a score by normalizing the data set against using a standard volume (word count). Working with your language services provider (LSP) to understand and keep track of your MQM scores ensures you meet your translation quality goals.
• BLEU evaluates machine translation output by comparing the translation engine’s output to a human-created translation. A higher score indicates more agreement between the two translations.
• TER also evaluates AI translations. It indicates the minimum number of edits needed to bring the translation to human quality. A lower number means a better-quality translation output.

Businesses need a standardized approach to quality assurance to identify issues in translations from an objective viewpoint. The localization industry relies on Linguistic Quality Assurance (LQA), a process by which human linguists evaluate translations against a schema of objective errors, to evaluate translation quality without bias. 

Smartling’s LQA Suite makes LQA a breeze. Built directly into the Smartling platform, the LQA suite provides a dedicated space for quality assessments separate from your main production environment, helping linguists evaluate quality without interfering with the larger translation process. If linguists notice an error during QA, the round-trip editing feature allows them to send updates straight to the production environment, streamlining the LQA process.

Whether you’re an LQA pro or just diving in, Smartling’s LQA Suite makes quality assessment easier and more manageable than ever.

3. Cultural relevance

Translations should match readers’ cultural norms and expectations. What works in one country might completely miss the mark in another. A good translation considers local customs, idioms, and references to make it more relatable. It shows respect for the values and beliefs of the target culture by being culturally sensitive. The translation should feel natural and create a stronger connection with the audience.

4. Fluency

Fluency requires a deep understanding of both the source and target language. Translators must follow standard spelling and punctuation in the local language to keep your translations fluent. A high-quality translation will feel like it was originally written for the target audience. When done right, the translation will read naturally, free from awkward phrasing.

5. Tone and style

A comprehensive translation strategy captures the tone and style of the original text, making it feel natural to the target audience. For example, a formal business report requires more serious language than a casual marketing campaign for teenagers.

A low-quality translation might miss these nuances, creating confusion or even offending the reader. A professional translation service with first-language experts ensures the final text reflects the same emotional tone as the original.

Transcreation provides an especially effective method for maintaining tone, meaning, and style. It tasks human translators with recreating a concept in the target language through copywriting rather than word-for-word translation.

6. Understanding the target audience

You need to know your audience to find the right tone. Are they young adults? Professionals? Small children? Good translators consider factors like age, culture, and even local slang to ensure the message hits the mark. A casual tone might work for a younger crowd, but that same approach could fall flat in a formal setting.

Jokes or idioms need special attention. What’s funny in one language might not land in another. Without adapting these elements, content can feel out of place or offensive.

7. Attention to detail

Attention to detail separates a good translation from a bad one. Errors like inconsistent vocabulary or a misplaced phrase can alter the meaning of the text and confuse the target audience.

Metrics like TER (Translation Error Rate) measure the effort required to edit a machine translation into a polished, human-quality result. A lower TER score means the AI translation required fewer changes, indicating better accuracy and detail. Businesses should work with a professional language service provider (LSP) to evaluate the quality of their machine-translated content and improve their workflows.

8. Consistency

Consistency keeps your message clear and easy to understand. If a phrase changes halfway through a document, it can confuse readers and make your content look unprofessional. This is especially important for technical documents, such as user manuals, contracts, or marketing materials, where mixed terminology can cause misunderstandings or even legal liability.

With features like bundled content snapshots and round-trip editing, Smarlting's LQA Suite allows reviewers to quickly spot and fix inconsistencies. Incorporate this tool into your workflow to produce translations that feel cohesive and aligned, regardless of the project’s scale or complexity.

How Smartling can help you with business translation

Business translation can be a headache, thanks to tight deadlines, ballooning expenses, and varying quality between translation companies. Fortunately, Smartling’s broad range of solutions and specialized professional translation services can help you maintain accurate translations, no matter your company’s needs.

AI translation, human translation, and everything in between

For high-volume, low-visibility content, like your company blog or help center content, our AI Translation solution instantly delivers cost-effective translations, maximizing your budget and saving you time. To achieve polished translations on high-traffic pages, our AI-powered Human Translation solution brings you the best of both worlds: AI-powered tools for speed and human expertise for precision.

Smartling’s translation services connect you with industry-specific translators who understand the jargon and subtleties necessary for your content to connect with local audiences. Whether you’re working on technical documents or marketing materials, Smartling can help you scale your localization efforts while delivering high-quality messaging that’s aligned with your goals.

Smartling’s best-in-class translation management system (TMS)

Smartling’s cloud-based TMS helps teams collaborate and stay organized. Features like translation memory and glossary management keep terminology consistent across projects, speeding up the translation process and ensuring your brand message is consistent in every language.

The right tools and linguistic support make all the difference when translating your content. That’s why we’ve created a free resource to help you along the way. Our eBook, “What to Look for in a Translation Solution: A Buyer’s Guide,” covers everything from workflow management and automation to integration with your existing tech stack. You’ll learn how to maintain quality control and leverage AI-powered translation.