Your translation project is complete and the work is published or in the queue to go live any day now. What’s the point of paying a vendor to look over the translated content at this point in the process?
- You’re getting bad feedback from the market: too many support calls, negative feedback or online reviews suggesting the information is confusing or incorrect.
- You’re seeing poor in-country adoption, which may signify a quality problem. Quality problems can turn a user off very quickly, especially in how-to or troubleshooting content.
- You need metrics tying spend to quality levels in order to benchmark a vendor.
- You have no true sense of quality within a specific market or a specific product line.
- You’ve got a chance to do updates. Maybe a rush in the product release cycle forced you to skip some QA process, including in-country QA by your partners, or in-context linguistic testing, and now you can go back and fix it.
If these scenarios sound familiar, you may need a diagnostic review that investigates the symptoms to uncover the underlying problem. This process sheds light on patterns or issues so you can fix the root cause and thereby improve future projects.
Typical root causes of quality issues
- Lack of consistent quality standards. If you find that quality varies a lot among deliverables, languages or product lines, it may be that no one — translators, reviewers, program managers — clearly understands what constitutes quality for your organization, and so they are not able to drive towards it.
- You might not have appropriately qualified linguists doing the work. Linguists must be trained, bilingual, in-country professionals. If you just use native-speaking partners to do translations, they might not be using TMs or glossaries effectively, they might not understand the content, or worse, they might not have sufficient training in best practices to stay faithful to the source.
- Your quality reviewers may not be qualified. Are your resources in-country speakers of the language with day jobs or are they linguists who know what to look for? Have they been trained?
- You might be rushing through QA. Tight schedules might force QA steps to be skipped.
- You might need quality automations — think “SpellCheck” — which would catch errors such as untranslated text, identical translations for different source, different translations for same source, punctuation issues and case mismatches, to name a few. As an added benefit, automations can speed up your quality process and save you money.
Get retroactive to be proactive
A diagnostic review may be the only way to gain insights that elevate your quality to a program-level, such as implementing processes and tools that mitigate quality issues down the road.
The first step of a diagnostic review is to get very clear on what “quality” means to your organization, so your vendor will evaluate your current definition and/or help you define it in terms that everyone across your localization program can understand. You will then be able to evaluate a translation against key performance metrics. Once quality is clearly defined, your vendor will review a specific set of translated content produced within a date range, for a specific language, product, etc., to look for patterns of issues.
The output from this exercise should include:
- a diagnostic report (or scorecard) with specific errors and their severity
- a professional evaluation of the patterns, including the identification of possible root causes
- suggestions on how to improve the process to mitigate future issues
If you’re tempted to cut corners, this is not the place to do it. Professional in-country bilingual reviewers can point out errors, but you really need an LQA specialist who knows localization inside and out to review data across translations or languages, to notice the patterns and draw the meaningful conclusions. Also, that LQA specialist will provide the concrete recommendations. Bilingual reviewers can identify problems with the language, but they probably don’t have the experience or insight required to identify program-level problems that lead to actionable suggestions for process improvements.
What are the benefits?
The benefits of linguistic quality, whether in a current project or within your ongoing program include better adoption in-market, more efficient spend, or spend focused in the “right” places, and the ability to understand where process improvements must occur. Linguistic quality diagnostics enable enterprises to make intelligent decisions around translation quality programs, resources, process and spend.
Have you done a diagnostic QA? What were the concrete results?