UniTranslate

Blog

Website Translation Services Compared: Finding the Perfect Solution for Your Business

1/1/2024

A guide for those who want to have their website translated - comparison of costs and important parameters

In the heart of Europe lies Switzerland, a land where cultures intertwine and languages dance in harmony like the melodies of an alpine symphony. Here, the snow-capped peaks echo with whispers of German, French, Italian, and Romansh, painting a portrait of a small yet vibrant microcosm of our global community. It is within this intricate tapestry that the need for multilingual communication emerges—a necessity as natural as the rivers that flow through its valleys.

The data speaks with undeniable clarity: “76% of consumers strongly prefer purchasing products described in their native language, while 40% will never engage with websites they cannot understand.” These aren’t mere statistics; they represent real people seeking authentic connections with your brand.

This comprehensive guide illuminates the spectrum of translation solutions available today, from artificial intelligence-driven approaches to human-crafted adaptations, each with distinct advantages for different content types and business objectives. Like a skilled conductor selecting the perfect instruments for each passage, we’ll help you determine which translation methodology harmonizes best with your unique brand voice, technical requirements, and international aspirations.

Let us journey together into a realm where your website becomes not merely a digital presence, but a global conversation—speaking directly to the hearts and minds of your customers in every corner of our diverse world.

Understanding your website translation options

1. Machine Translation (MT)

Overview: Machine translation uses artificial intelligence algorithms to automatically translate text from one language into another without human intervention. Popular engines include Google Translate, DeepL, Microsoft Translator and Amazon Translate.

Quality considerations:

  • Neural machine translation (NMT) has significantly improved accuracy compared to older statistical methods
  • Handles simple, straightforward content reasonably well
  • Struggles with industry-specific terminology, cultural nuances, idioms and contextual understanding
  • Can produce awkward phrasing and grammatical inconsistencies
  • Quality varies considerably by language pair (e.g. English→French typically works better than English→Hungarian or English→Japanese)

Cost structure:

  • Most economical solution (0.01-0.05 € per word)
  • Many providers offer tiered pricing based on volume
  • Some services offer limited free usage (usually with API restrictions)
  • Enterprise-level solutions with customisation options range from 500 to over 10,000 € annually

Implementation time: Immediate to a few days for setup

Best suited for:

  • Internal documentation and knowledge bases
  • User-generated content (comments, reviews)
  • High-volume, low-risk content where perfect accuracy is not critical
  • Businesses with extremely constrained budgets
  • First drafts that will undergo further refinement

Practical example: A German mechanical engineering group uses machine translation for its internal knowledge management system with over 50,000 technical documents that are updated regularly and are primarily intended for internal use.

2. Machine Translation + Post-Editing (MTPE)

Overview: This hybrid approach starts with machine translation, followed by professional human editors who optimise the output for accuracy, readability and cultural appropriateness. There are two standard levels:

  • Light post-editing: Focuses on correcting major errors and ensuring the text is comprehensible
  • Full post-editing: Comprehensive revision to ensure the translation reads as if it had originally been written in the target language

Quality considerations:

  • Significantly higher quality than pure machine translation
  • Quality depends on both the quality of the machine translation and the expertise of the post-editor
  • Full post-editing can achieve human translation quality for many content types
  • Ensures consistency through the use of translation memories and terminology databases
  • May still miss subtle cultural references or brand language

Cost structure:

  • Moderate costs (0.06-0.12 € per word for light post-editing, 0.08-0.15 € for full)
  • 30-50% more economical compared to pure human translation
  • Often discounts for repetitive content and use of translation memories
  • Time savings of around 40% compared to complete retranslation

Implementation time: 1-7 days depending on volume and language pairs

Best suited for:

  • Product descriptions and catalogues
  • Corporate blogs and informational content
  • Knowledge base articles and FAQs
  • Marketing materials where high quality is important but not critical
  • Projects with tight deadlines requiring faster turnaround
  • Companies with moderate translation budgets

Practical example: An Austrian e-commerce retailer for sporting goods uses MTPE to quickly translate its weekly updated online catalogue of over 10,000 products into five European languages while ensuring engaging product descriptions.

3. Human Translation

Overview: Professional linguists, usually native speakers with subject-matter expertise, translate content manually while preserving meaning, tone and cultural appropriateness. This process often includes:

  • Translation by a professional translator
  • Editing by a second linguist
  • Proofreading as a final quality check
  • Optional review by experts in the relevant field

Quality considerations:

  • Gold standard for translation quality
  • Accurately captures nuances, idioms and cultural references
  • Preserves brand language and intended messaging
  • Adapts content to resonate with the target audience (transcreation)
  • Ensures regulatory compliance for legal and medical content
  • Takes local market particularities and cultural sensitivities into account

Cost structure:

  • Premium pricing (0.15-0.40+ € per word)
  • Prices vary by language pair, content complexity and specialisation
  • Volume discounts and use of translation memories typically available
  • Additional costs for specialised services such as transcreation or expert review

Implementation time: 7-30+ days depending on volume and complexity

Best suited for:

  • Legal content and contracts
  • High-quality marketing materials and advertising campaigns
  • Brand messaging and slogans
  • Technical content requiring the highest precision
  • Highly regulated industries (healthcare, finance, law)
  • Luxury brands where perception is critical
  • Content addressing markets with significant cultural differences

Practical example: A Swiss luxury watch company has its entire website, including technical specifications, brand history and product descriptions, translated by professional translators into twelve languages to maintain its premium image in all markets.

The critical post-translation process

Receiving translated content is only half the journey. Integrating and managing multilingual content within your digital ecosystem requires careful planning and execution:

CMS integration challenges

Technical requirements:

  • Verify that your CMS supports multilingual functionality (WordPress with WPML or Polylang, Drupal with i18n, Typo3 with its own multilingual features, etc.)
  • Consider URL structures (subdomains vs. subdirectories vs. separate domains)
  • Implement hreflang tags for proper SEO signalling between language versions
  • Ensure correct character encoding (UTF-8) for non-Latin alphabets
  • Configure language-specific templates to accommodate text expansion/contraction
  • Plan for different reading directions (left-to-right vs. right-to-left)

Content management workflows:

  • Implement a central Translation Management System (TMS) that connects to your CMS
  • Establish clear workflows for content updates across languages
  • Consider translation proxies such as Weglot, Smartling or Lokalise for more complex implementations
  • Determine how dynamic content and user-generated content will be handled
  • Plan for continuous maintenance as source content changes
  • Automate the translation process through API integrations where possible

Common pitfalls:

  • Text expansion issues: German and Finnish text typically expands by 20-30% compared to English
  • Problems displaying right-to-left languages (Arabic, Hebrew, Farsi)
  • Broken layouts due to longer words or different character heights
  • Untranslated content in images or embedded applications
  • Missing translations for system messages and notifications
  • Technical limitations in displaying certain fonts or special characters

Quality assurance process

Implementing a robust QA process is essential for multilingual websites:

  1. Linguistic QA:

    • Review translations in context within the actual web page environment
    • Check terminology consistency across pages
    • Verify proper handling of region-specific elements (dates, currencies, units, phone numbers)
    • Ensure all translated content is culturally appropriate
  2. Functional testing:

    • Test all forms and interactive elements in every language
    • Verify that the search function works correctly in all languages
    • Test navigation and user flows in each localised version
    • Check responsive design in all languages (especially important for languages with longer words)
    • Test the website across different browsers and devices in each language
  3. Cultural verification:

    • Review imagery for cultural appropriateness
    • Check colour schemes for cultural considerations (e.g. white carries a different meaning in some Asian cultures than in Europe)
    • Verify that calendar functions handle regional holidays correctly
    • Ensure metaphors and analogies work in the target culture

SEO considerations for multilingual websites

Translated content requires specific SEO optimisation:

  • Keyword localisation: Direct translation of keywords often misses local search behaviour. Research keywords in the target language independently with tools such as Sistrix, Semrush or Ahrefs.
  • Metadata: Ensure all meta titles, descriptions and image alt texts are properly localised
  • Hreflang implementation: Signal correct language/region relationships between pages
  • Local backlink strategy: Develop backlinks from websites in the target language/region
  • Registration with local search engines: Register with local search engines (e.g. Yandex for Russia, Baidu for China, Naver for South Korea)
  • Schema markup: Implement localised schema markup for enhanced SERP features
  • Local social media integration: Integrate local social networks (e.g. Xing in DACH countries, WeChat in China)

Comprehensive solution comparison

Factor Machine Translation MTPE (Light) MTPE (Full) Human Translation
Cost 0.01-0.05 €/word 0.06-0.12 €/word 0.08-0.15 €/word 0.15-0.40+ €/word
Quality Low-Medium Medium Medium-High Highest
Turnaround (10,000 words) 1-2 days 3-5 days 5-8 days 10-20 days
Brand voice preservation Weak Moderate Good Excellent
Cultural adaptation None Limited Good Excellent
Technical accuracy Unreliable Moderate Good Excellent
SEO effectiveness Low Moderate High Highest
Regulatory compliance Not suitable Suitable to a limited extent Suitable for some cases Fully compliant
Ongoing maintenance costs Low Medium Medium-High High
Conversion strength Low Medium Good

Creating a strategic translation roadmap

Rather than using a one-size-fits-all approach, consider a tiered strategy:

  1. Content audit and prioritisation:

    • Map your content by type, volume and business impact
    • Categorise content based on risk level and quality requirements
    • Determine language priorities based on market analysis and business goals
    • Identify content that can remain the same for all markets and content that needs adaptation
  2. Implementing a hybrid approach:

    • Tier 1 (Human Translation): Legal content, primary marketing pages, key product information, terms and conditions
    • Tier 2 (MTPE Full): Secondary marketing content, detailed product descriptions, support documentation
    • Tier 3 (MTPE Light): Blog posts, knowledge base articles, category pages
    • Tier 4 (Machine Translation): User-generated content, internal documentation, customer reviews
  3. Pilot phase and optimisation:

    • Start with one target language/market as a pilot project (ideally one with larger market potential)
    • Collect user feedback and performance metrics (conversion rates, bounce rate, dwell time)
    • Refine the approach before expanding to additional languages
    • Develop a continuous improvement process based on the data collected

UniTranslate’s comprehensive website translation services

At UniTranslate, we offer end-to-end website translation solutions tailored to your specific needs:

  • Expert assessment: Expert analysis of your content and requirements
  • Tailored translation strategy: Customised approaches combining different methods for different content types
  • Technical implementation support: Assistance with CMS integration and workflow setup
  • Continuous quality monitoring: Ongoing reviews and refinement of translations
  • Scalable solutions: The ability to grow with your business as you expand into new markets
  • Integrated SEO optimisation: Language-specific keyword research and SEO adaptation in each target market
  • Cultural consulting: Advice on cultural adaptations and local market particularities

Our team includes specialised translators in over 100 languages with industry-specific expertise in technology, healthcare, law, manufacturing, e-commerce, fashion, tourism and more. We combine linguistic excellence with technical implementation knowledge to ensure that your multilingual website not only says the right things, but works perfectly in every language and achieves your conversion goals.

Making the right choice for your business

The ideal website translation solution depends on several factors specific to your business:

  • Budget realities: Match translation methods to financial constraints
  • Quality requirements: Consider your brand positioning and risk tolerance
  • Content volume: Higher volumes may require a tiered approach
  • Technical resources: Assess your team’s capacity for implementation
  • Market importance: Prioritise human translation for your most valuable markets
  • Timeline constraints: Balance speed requirements with quality needs
  • Industry compliance: Consider legal requirements in your industry

Remember that website translation is not a one-off project but an ongoing commitment. As your business evolves, your multilingual content strategy should adapt accordingly. By choosing the right translation approach from the start, you create a solid foundation for international growth and engagement with global audiences.

Contact UniTranslate today for a personal consultation to determine the optimal website translation strategy for your business needs and secure your success in international markets.