What is a localization platform?
A localization platform is the software layer that sits between the places your content lives — code repositories, design files, CMSs, help centers, shared drives — and the people and machines that translate it. It pulls content in, splits it into translatable units, translates it with machine translation or AI, routes it to reviewers, checks quality, and pushes finished translations back to where they came from. The goal is a repeatable process instead of a string of one-off translation jobs.
Historically, that job was split across separate tools: a translation management system (TMS) to handle projects, vendors and deadlines; a computer-assisted translation (CAT) tool where linguists actually edit text with translation memory and glossaries; and a collection of scripts, file exports and email threads to move content between them. Modern platforms such as Smartcat combine all three in one browser-based workspace, with AI translation built in and a marketplace of linguists for human review.
Translation vs. localization vs. internationalization
Translation converts text from one language to another. Localization adapts a whole product or piece of content to a market — strings, dates, currencies, images, UI length, tone. Internationalization (often shortened to i18n) is the engineering work that makes localization possible in the first place: externalizing strings into resource files, supporting plurals and right-to-left scripts, and avoiding hard-coded text. A localization platform does its best work when the product it localizes has been internationalized first; Smartcat’s starter guide to internationalization in software design covers the basics.
Who uses a localization platform?
Localization managers use it to run programs across many content types and markets. Product and engineering teams use it to keep app and software strings in sync with every release. Marketing, support and L&D teams use it to publish in multiple languages without waiting on agencies. Language service providers (LSPs) and freelance linguists use it as their production environment. The rest of this guide walks through each layer of the platform and links to the Smartcat pages, articles and help docs that go deeper.
Translation management systems (TMS)
Most companies don’t start with a translation process — they start with translation requests arriving in Slack, email and Jira, files versioned by filename, and a translation memory locked inside one agency’s system. A TMS replaces that with a single queue. Smartcat’s translation management system pulls content from the tools your teams already use, translates it with AI, routes it to whoever needs to review it, and pushes it back — with each job’s cost attributable to the campaign, product or market that requested it.
Connect your content sources
Files are uploaded directly or synced through connectors for code repositories, CMSs, design tools and knowledge bases. Smartcat detects each file’s format and routes it to the right parser — see the full list of supported file formats.
Translate with AI, memory and terminology
Content is segmented, matched against your translation memory, and translated by AI that applies your glossaries. Smartcat’s machine translation layer routes content through multiple MT engines and large language models and applies your terminology to every job.
Review where it matters
Workflows can be AI-only or AI plus human review. Reviewers can be your own team, freelancers or LSPs you add to your account, or linguists hired from the Smartcat Marketplace. Everyone works in the same browser-based editor.
Deliver back to the source
Translations are exported in the original format or pushed back automatically through the connector — to the Git repository, CMS, or design file they came from. Every confirmed segment is saved to translation memory for reuse.
What to look for in a TMS
Useful evaluation criteria include: how content gets in and out (connectors, API, CLI); whether translation memory and glossaries are applied automatically and owned by you rather than a vendor; how workflows, assignments and deadlines are configured; how quality is measured; and how costs are reported. Smartcat’s comparison of localization software platforms for large companies sets out these criteria for enterprise buyers, and the localization project management software page shows how workflows are set up for agile software teams.
Switching platforms is a project in its own right. The webinar on why companies avoid TMS migration projects covers what localization managers worry about, and the Phrase-to-Smartcat migration guide walks through moving memories, glossaries and workflows. For an example of consolidating suppliers on one platform, see how Topcon went from eight translation vendors to one.
CAT tools and translation memory
A CAT tool is the editor where translation actually happens. Smartcat’s CAT tool splits your file into segments — usually one sentence each — with the source on the left and the translation on the right. It pre-fills every segment that matches your translation memory, enforces your glossary as you type, and flags QA issues before you deliver. Formatting tags are preserved, so the exported file rebuilds correctly in its native format. It runs in the browser, so there is nothing to install and no version to keep matched with your reviewer’s.
If you’re new to the category, start with what a CAT tool is and how it works, then read how translation segments work in CAT tools.
How translation memory works
Translation memory (TM) is a database of every segment that has been translated and confirmed, paired with its approved translation. In Smartcat it is shown as Reviewed Translations in the newer interface: once a linguist or reviewer confirms a segment, the pair is saved automatically and becomes available in future projects. Smartcat’s TM matching documentation explains the match bands — fuzzy matches (50–99%), exact matches (100%) and context-verified matches (101–103%) — and how to configure them. The blog explains why translation memory management matters for cost and consistency.
| Capability | What it does | Why it matters |
|---|---|---|
| Segmentation | Splits content into sentence-level units with formatting tags protected | Translators can’t break layout or code |
| Translation memory | Re-uses confirmed translations at exact, fuzzy and context match levels | You don’t pay twice for the same sentence |
| Glossary / termbase | Suggests and enforces approved term translations | Product names and key terms stay consistent |
| Automated QA | Flags spelling, punctuation, terminology, formatting and TM-consistency errors | Issues are caught before delivery, not after |
| Collaboration | Comments, assignments and review stages in one shared editor | No file ping-pong between translators and reviewers |
Working with existing TMs and other CAT tools
Translation memories travel between tools as TMX files. The guide to using TMX files explains the format, and the help center shows how to connect an existing translation memory or glossary to Smartcat projects. Teams that also use memoQ can translate memoQ documents in Smartcat by exporting bilingual XLIFF and reimporting the result. For hands-on editing of memory entries, Smartcat introduced a Translation Memory Editor in 2024.
Glossaries, terminology and style
Translation memory stores whole sentences; a glossary (or termbase) stores individual terms — product names, technical vocabulary, words you never translate — with their approved translations. Glossaries are what keep “workspace” from becoming three different words across your app, help center and marketing site. The guide to creating glossaries for content translation covers what to include and how to maintain one.
In Smartcat, glossaries live under Terminology in Intelligence Fabric — the central hub for reviewed translations, terminology and language guidance that shapes AI output. Intelligence Fabric can also build these assets for you from your existing content; see how to connect an existing TM or glossary if you already have one.
Governing terminology at scale
Terminology only works if someone owns it. Smartcat’s glossaries include role-based permissions and a term approval workflow: administrators and project managers add, edit and approve terms, while linguists can suggest new ones for approval. The glossary setup and management guide covers creating, importing and exporting termbases. If you need outside help building or cleaning up linguistic assets, the marketplace lists specialists for glossary management and translation memory management.
Localization file formats: XLIFF, JSON, PO, XML and more
Software and structured content don’t arrive as Word documents. They arrive as resource files whose keys, tags, placeholders and nesting must survive translation untouched — one broken placeholder can crash a build. Smartcat’s CAT editor handles 80+ file formats, including Word, Excel, PowerPoint, PDF, IDML, XLIFF, JSON, XML, PO and SRT, and each format below has a dedicated page explaining what is preserved.
| Format | Typically used for | What Smartcat preserves |
|---|---|---|
| XLIFF 1.2 / 2.0 | Exchange between CMSs, authoring tools and CAT tools | Inline codes protected; “translate="no"” units locked; exports XLIFF 1.2 |
| JSON | Web and app locale files | Same keys and nesting; placeholder tokens locked via regex rules |
| Gettext PO / POT | Open-source and server-side software (GNU gettext) | Variables such as “%s” and “{name}” become protected tags |
| XML | Configs, content exports, Android resources | Markup untouched; translatable elements chosen by XPath |
| YAML, iOS .strings, Android XML | App and framework string resources | Imported as keys in software localization projects |
XLIFF
XLIFF (XML Localization Interchange File Format) is the standard way to move translatable content between systems. Smartcat’s XLIFF editor imports XLIFF 1.2 and 2.0 — support for 2.0 was announced in 2022 — shows source and target side by side with inline codes protected, and exports XLIFF 1.2, the flavor most tools import cleanly. The complete guide to XLIFF files and XLIFF for software localization go deeper, and the help center explains how to translate XLIFF files exported from authoring tools.
JSON, PO, XML and YAML
The JSON translator translates only string values and returns a file with the same keys and nesting. For Gettext PO files, Smartcat offers three parsers — Default PO, Serge.io PO and Monolingual PO — and with the Serge.io parser, re-uploading a file only reprocesses strings whose source changed; see translating PO files in Smartcat for setup. The XML translator uses a configurable parser where you specify, by XPath, which elements and attributes hold translatable text; the help article on flexible XML parsing shows how to set models, IDs and length limits. YAML files are handled by the YAML translator.
Software, mobile app and game localization
Software localization is different from document translation: content is a set of keys rather than pages, it changes with every sprint, and it is consumed by builds rather than readers. Smartcat’s software localization platform brings developers, project managers, linguists and stakeholders into one workspace, with AI agents that learn from your reviewers’ edits and plugins for GitHub, Android and iOS. The guide on how to translate software lays out an end-to-end process.
How software localization projects work
Alongside regular document projects, Smartcat offers software localization projects that manage keys (strings) directly. Strings are imported and exported in JSON, YAML, Android XML and iOS .strings, and keys are grouped into collections. Smartcat becomes the source of truth for your software copy: UX writers and product managers change the copy themselves, and developers export the subset of keys they need. Plurals are supported natively or through ICU MessageFormat syntax. For product teams localizing every release, the product localization page shows the AI-driven workflow.
Mobile app localization (iOS & Android)
Mobile apps add their own constraints: platform-specific resource formats, tight UI space, and app store listings that also need localizing. Smartcat’s mobile app localization workflow connects your Android, iOS or Java app through Git, translates strings with AI and reviewers, and uploads the translated content back to your app. The integrations automatically pull Android XML string resources or iOS .strings files from the repository and push translations back — either when they are complete or when a chosen workflow stage is reached.
For strategy and step-by-step detail, read mobile app localization: what developers need to know, plus the platform-specific guides to iOS app localization and Android app localization.
Game localization
Games combine every hard localization problem at once: large volumes of UI strings, dialogue and lore that need creative adaptation, character limits, voice-over, and launches across many markets on the same day. The guide to video game localization covers the process end to end, and 10 game localization challenges lists the pitfalls to plan for. Studios can manage game strings in the same software localization workflow described above — see game translation management — and hire specialists through the marketplace’s game localization services, which uses AI matching to find localization specialists for any language pair.
Figma and UI design localization
Localizing at the design stage catches problems — text expansion, truncated buttons, right-to-left layouts — before they reach code. Smartcat’s Figma translation plugin translates designs into 280+ languages while preserving layout, and generates live previews so reviewers can check translations in context in the Smartcat Editor without needing extra Figma licenses.
The help guide to connecting Figma with Smartcat explains how translations are embedded into Figma text objects, so you can modify and test layouts in one file instead of keeping a copy for every language. Two product updates are worth knowing: the developer-friendly Figma integration for software UI translation (2023) and live Figma translation previews (2024). To see the workflow end to end, watch the hands-on Figma and Smartcat AI webinar.
Developer workflows and continuous localization
Continuous localization means translation runs alongside development instead of after it: new or changed strings are detected, translated and returned automatically, so every release ships in every language. Smartcat gives developers four ways in, summarized on the developer portal’s getting started page.
Mobile teams can use the dedicated Git and Android and Git for iOS integrations. Teams connecting their own systems can use the translation API for enterprise workflows, which connects via API and webhooks and applies your translation memory. The help center’s developer portal guide walks through a software localization project set up from the command line.
Making localization agile
The hardest part of continuous localization is usually process, not tooling. The agile localization management page shows how developers, reviewers and stakeholders share one workspace. For real-world examples, read how Xsolla set up continuous localization with Smartcat and Serge, how Infobip made localization and Agile get along, and how Kaspersky localizes big, agile software. The talk on traps of continuous localization covers what goes wrong after launch.
Translation quality assurance and LQA
Quality in a localization platform works on three levels: automated checks while people translate, scoring that tells you which AI output needs a human, and structured linguistic quality assurance (LQA) when you need a formal error count.
Automated QA checks
Smartcat checks translations for spelling, punctuation, terminology, formatting and consistency with the translation memory as linguists work in the editor. Each check can be set to ignore errors, notify, or require correction, and project managers can download a QA report — see performing quality assurance.
Scoring AI translations
Smartcat’s Translation Quality Score rates AI translations down to the sentence and highlights the segments that need human review, so reviewers spend time only where it matters. AI coworkers take this further: the Quality Assurance Coworker checks translations for terminology, formatting and accuracy errors and classifies each issue by type and severity, while the Reviewer Coworker pre-reviews content for fluency, tone, terminology and brand fit before your experts see it.
Linguistic quality assurance (LQA)
LQA is a formal review in which a reviewer labels each error by category and severity to produce a score. Smartcat’s LQA profiles are based on the MQM (Multidimensional Quality Metrics) model, where each error type has a category, a weight and a severity. The article on automating linguistic quality assurance at scale explains the process, and the marketplace offers localization quality assurance services if you need independent reviewers.
Managing translation vendors and the linguist marketplace
Most corporate localization programs mix internal reviewers, freelancers and language service providers. In Smartcat you add vendors to your account and assign them tasks inside projects, alongside your own team. Freelancers and LSPs are both “suppliers”: either can be added, assigned work and paid through Smartcat — the help article on LSPs vs. freelancers explains when to choose each.
When you need extra capacity or a new language pair, the Smartcat Marketplace uses AI to select linguists from more than 500,000 language professionals. You can hire a translation project manager as well as translators and reviewers. On the finance side, linguist payments and procurement consolidate supplier payments into one invoice.
Integrations and connectors
Connectors are what turn a TMS into a localization platform: instead of exporting and importing files by hand, Smartcat connects to your content source, translates the content and returns it — on a schedule or on demand. The help center’s integrations overview and full list explains the two technical systems integrations use, and the integrations directory lists every plugin by category.
Software and product content
Content management and support systems
Translating a website rather than a product? Website and CMS plugins are covered in the website translation plugins category.
Localization platform tools & resources
Quick links to the main Smartcat platform pages, grouped by job to be done.
Translation management & workflow
File format tools
Software & app localization
Marketplace services
Localization platform guides, how-tos & events
In-depth articles and recorded talks on translation management, CAT tools, software localization and running a localization program.
Smartcat product documentation
Step-by-step product guides for localization managers, linguists and developers working in Smartcat.