Sample deliverable

Korean product localization QA

This client-ready example shows how Korean language, layout, and localized product issues are documented for handoff. It is a demonstration, not paid client work, and no example below is presented as a real product defect.

Demonstration · not paid client workPublic briefs reviewed · 2026-08-03

Release decision

Pending product access and test evidence

Example scope

Korean product · desktop and mobile

Handoff format

Summary · issue log · evidence · retest

Execution status

Not run · client inputs required

Research basis

Built around real brief requirements

The report fields reflect recurring requirements in the public briefs below. These links document the research basis only and do not imply a relationship with the companies or clients that posted them.

South Korean eCommerce QA

Desktop and mobile review, broken-flow and visual checks, AI-translated Korean review, cultural fit, and actionable improvement notes.

View public brief

Localization QA for web and mobile

In-product validation, truncation and layout checks, terminology and tone consistency, functional localization testing, and Jira/TestRail-ready issues.

View public brief

Korean AI agent localization

Prompt localization, live agent testing, cultural accuracy, conversational tone, and context-handling review.

View public brief

Structured QA handoff

Module, steps to reproduce, expected versus actual behavior, screenshots or recordings, severity, and a post-fix retest.

View public brief

Issue log

Example issue records

Each record separates verified behavior from language preference and pending assumptions. Environment, evidence, ownership, and retest status are completed only after product access is provided.

KO-LQA-001

Major
Module
Product detail / CTA
Environment
Mobile web · Korean locale
Category
Linguistic QA · Terminology
Status
Example finding

Steps to reproduce

  1. Open a Korean product-detail page on the agreed mobile viewport.
  2. Add the item to the cart and compare the CTA with the approved glossary.

Evidence

[Client screenshot or screen-recording link required]

Observed / actual

Example: the same action is translated with two different product terms across the page and confirmation message.

Expected

One approved Korean term is used consistently in the CTA, confirmation, cart, and help copy.

User or business impact

A high-visibility inconsistency makes AI-translated storefront copy feel unreviewed and can reduce purchase confidence.

Recommended action

Confirm the preferred term in the client glossary, update every occurrence, and run a cross-page terminology search before retest.

KO-UI-002

Major
Module
Cart / Checkout
Environment
Mobile web · target devices pending
Category
Visual localization · Functional QA
Status
Example finding

Steps to reproduce

  1. Set the browser to the target mobile width.
  2. Proceed from cart to checkout with the Korean locale enabled.
  3. Inspect primary buttons, validation messages, and order-summary labels.

Evidence

[Viewport, OS/browser, screenshot, and recording required]

Observed / actual

Example: a longer Korean primary CTA wraps or clips, reducing the tappable label to an incomplete phrase.

Expected

The complete Korean action remains legible without overlap at every supported viewport and text-size setting.

User or business impact

The issue appears in a revenue-critical flow and can prevent users from understanding the next action.

Recommended action

Shorten the Korean CTA with product approval, fix the responsive layout, and retest all declared mobile breakpoints.

KO-FUNC-003

Critical
Module
Checkout / Korean address
Environment
Desktop and mobile web
Category
Localized functional flow
Status
Example finding

Steps to reproduce

  1. Enter a client-provided Korean test address and postal code.
  2. Submit the form, return to edit, and complete the sandbox order.

Evidence

[Access and test data required before execution]

Observed / actual

This row stays unscored until the client supplies a staging build, test account, valid test address, and payment sandbox.

Expected

The agreed Korean address formats validate, persist, and reach order confirmation without data loss.

User or business impact

A failure would block checkout; severity must be based on reproduced behavior rather than a fabricated portfolio claim.

Recommended action

Run the test after access is supplied, attach evidence, record the exact expected/actual result, and assign an owner and retest status.

Before testing starts

Inputs needed from the client

Missing information stays marked as pending. These inputs define the test boundary and turn the template into a report the client can act on.

  1. 1

    Staging or live URL, build/version, and any access credentials

  2. 2

    Supported desktop/mobile platforms, browsers, devices, and viewport targets

  3. 3

    Source English plus the current Korean export, or access to the localization/CMS tool

  4. 4

    Approved glossary, brand voice, honorific level, and terms that must not change

  5. 5

    Priority journeys and acceptance criteria, including what counts as release-blocking

  6. 6

    Ground-truth shipping, return, payment, privacy, and support policies

  7. 7

    Test accounts, Korean address data, coupon data, and payment sandbox when functional QA is included

  8. 8

    Preferred handoff tool: Google Sheet, Excel, Jira, TestRail, Notion, or PDF/Doc

  9. 9

    Deadline, reporting language, evidence rules, and number of retest rounds

Working setup

Available environments and terms

  • Platforms: Windows, Linux, macOS, and iPhone environments, major browsers, Korean mobile networks, and a Korean App Store account.
  • Korean voice AI QA: domestic-number calls using a live microphone, with permitted recordings and evidence for pronunciation, honorifics, latency, interruptions, and context retention.
  • Availability: 15–20 hours per week, including weekdays and weekends. A complete 20–30 item batch typically takes 3–5 business days.
  • Communication: Korean live meetings and structured, AI-assisted written communication for cross-language projects. Screen sharing and camera-on Korean meetings are available.
  • Trials and tools: small paid trials and tightly scoped free tests are available. Client-selected tools are confirmed and validated before kickoff.
  • Experience status: no paid client work is claimed. Every public report remains clearly labeled as a demonstration.