On this page
Four criteria for every content typeScore your contentFour quality levels, not oneA worked exampleTactics that stretch the budgetA company decides to go into Korea and asks for a quote to translate "the website, the manuals and the sales deck". The quote arrives, and it is three times the budget. The usual reaction is to cut quality across the board. The better reaction, and the one that keeps translation for business affordable, is to cut scope with a method: translate the right things first, at the right level, and leave the rest for later or for lighter treatment.
This is the prioritisation matrix we use with clients. It is our own working tool, not a published standard, and it fits on one page.
Four criteria for every content type
List your content by type, not by file: product pages, onboarding emails, user manual, safety labels, contracts, support macros, blog, investor deck. Then score each type from 1 to 5 on four questions:
| Criterion | Question | Scores high when |
|---|---|---|
| Reach | How many target-market readers see it? | Every customer sees it (product pages, app UI) |
| Revenue | How directly does it affect buying or renewing? | It sits on the path to purchase (pricing, checkout, sales deck) |
| Risk | What happens if it is wrong? | Legal, safety, financial or regulatory consequences (contracts, warnings, terms) |
| Shelf life | How long will the text be in use? | It stays live for years (manuals, UI strings, templates) |
Score your content
Enter scores for one content type. The tool suggests an order and a quality level. Risk is treated separately: a high risk score sets the quality level regardless of the total.
Illustrative thresholds. Adjust to your budget and market.
Four quality levels, not one
The matrix only helps if you buy translation at more than one level. A workable set:
Specialist, revised, reviewed
Subject specialist translator, second-linguist revision (the step ISO 17100 requires), plus a review by your own Korean reader. For contracts, safety text, regulated content.
Translated and revised
Professional translation with independent revision. The default for customer-facing content.
Light treatment
Human translation with a self-check, or carefully post-edited machine output. For internal or short-lived material. Below that: machine output for understanding only, never published.
A worked example
An illustrative software company entering Korea scored its content like this:
| Content type | Reach | Revenue | Risk | Shelf life | Total | Decision |
|---|---|---|---|---|---|---|
| App interface | 5 | 5 | 3 | 5 | 18 | Now, level 2, with in-context review |
| Terms of service and privacy notice | 3 | 2 | 5 | 4 | 14 | Now (risk), level 1 |
| Pricing and checkout pages | 4 | 5 | 3 | 3 | 15 | Now, level 2 |
| Help center articles | 3 | 2 | 2 | 3 | 10 | Next: top 30 articles only, level 2 |
| Blog archive | 2 | 1 | 1 | 2 | 6 | Later, or not at all |
| Internal release notes | 1 | 1 | 1 | 1 | 4 | Machine output for understanding |
The result was a first phase at roughly a third of the original word count, with the money concentrated where readers and risk are. The help center was translated in stages, starting with the articles that generated the most support tickets.
Tactics that stretch the budget
- Fix the source first. Delete outdated pages and merge duplicates before counting words.
- Own your translation memory. Repeated and similar text gets cheaper every round; see translation memory as a strategic asset.
- Build the termbase before volume. A short glossary of product terms prevents expensive rework later.
- Use data for phase two. Translate the help articles and pages that Korean users actually visit or ask about.
- Rescore yearly. Priorities shift as the market grows.
For the budget and calendar around this, see planning a yearly translation budget; for the strategic frame, localization ROI drivers and a localization maturity model for Korea. If the question is whether you need Korean at all yet, start with triggers that tell a business it needs professional Korean translation.
Have a list of content and a fixed budget? Send us the list and we will score it with you and propose a phased Korean plan. Contact us or see our services.