What a customer story will look like
Each card is a slot for a real, user-provided story — company, quote, author, and outcomes grounded in what we have actually shipped: the editor, AI translation, the glossary and QA workflow, and the developer platform.
Your company here
PlaceholderSample industry · sample team size
“This is where a real customer quote will go — for example, how a team cut translation turnaround after switching from spreadsheets or a legacy TMS.”
Sample Name — Sample Title
—
sample metric (e.g. faster releases)
—
sample metric (e.g. words translated)
Your company here
PlaceholderSample industry · sample team size
“A second placeholder case study slot — real copy will describe a concrete before/after, grounded in shipped features like the editor, AI translation, or the API.”
Sample Name — Sample Title
—
sample metric (e.g. words reused via TM)
—
sample metric (e.g. time to first translation)
Your company here
PlaceholderSample industry · sample team size
“A third placeholder slot for a developer-platform-flavored story — automating translation via the REST API and webhooks in a CI/CD pipeline.”
Sample Name — Sample Title
—
sample metric (e.g. manual steps removed)
Early access
Placeholder quotes — real customer stories coming soon.
Native File Formats
Frameworks Reached
Team Roles
“We moved off spreadsheets and a legacy TMS in an afternoon. The keyboard-first editor alone paid for itself in the first week.”Placeholder Name — Engineering Lead, Placeholder Co. (placeholder)
“AI translation with translation memory means we almost never pay to re-translate the same string twice.”Placeholder Name — Localization Manager, Placeholder Studio (placeholder)
“The API and signed webhooks let us wire translation straight into our release pipeline without any glue scripts.”Placeholder Name — Founding Engineer, Placeholder Labs (placeholder)
“A shared glossary plus automated QA means our product terminology finally ships the same way in every language.”Placeholder Name — Localization Program Manager, Placeholder Group (placeholder)
“In-context editing is the feature translators actually asked for — they ALT+click the string in the running app and fix it in seconds.”Placeholder Name — Frontend Engineer, Placeholder Works (placeholder)
How teams localized before, and how they do it now
Most teams we talk to were coordinating translations through some combination of spreadsheets, email threads, and a lossy CSV round-trip through their translation tool of choice. Here is what that workflow typically looked like, next to how the same work happens in transglot today.
Before
Strings live in a spreadsheet, exported by hand and emailed to translators.
With transglot
Strings live in a virtualized, keyboard-first editor with optimistic saves — no export step, no email thread.
Before
Every round trip through a hand-rolled CSV risks mangling a nested key, a plural form, or a placeholder token.
With transglot
Native import/export for 20 formats (JSON, Android XML, iOS .strings/.stringsdict/.xcstrings, Flutter ARB, Laravel PHP/JSON, XLIFF 1.2 and 2.0, gettext PO, YAML, Java .properties, .NET .resx, Markdown/MDX, Unity Localization CSV, and Unreal Engine PO) — and when you do want a spreadsheet, the CSV/XLSX codecs round-trip plurals and placeholders instead of flattening them.
Before
A translator has no context for what a string means or where it appears in the product.
With transglot
Glossary notes, part-of-speech, and per-locale target terms give translators (and Claude) the context a bare string can't.
Before
Brand names and forbidden terms get machine- or human-mistranslated because nothing enforces the rule.
With transglot
A glossary with do-not-translate, forbidden, and case-sensitive flags is checked deterministically after every translation.
Before
New languages mean manually diffing two files to find what's missing.
With transglot
Adding a locale auto-backfills missing keys against your source language — nothing to hand-copy.
Before
Nobody catches a dropped placeholder or a missing plural category until a user reports a broken screen.
With transglot
A deterministic QA pre-gate checks placeholder integrity, plural completeness, and length before anything ships — plus an AI review pass for what only a fluent reader would catch.
Before
"Is this translated yet?" means opening the spreadsheet and searching by eye.
With transglot
Typo-tolerant fuzzy search (Postgres pg_trgm) and filter-by-state find what's untranslated in seconds.
Before
Shipping a release means someone remembering to export and re-import translations manually.
With transglot
A REST API, a CLI, a GitHub Action, an MCP server, and HMAC-signed webhooks pull, push, and trigger translation as part of your normal CI/CD pipeline.
Real capability numbers, not vanity metrics
We will not invent usage stats we do not have yet. These are the shipped capabilities every customer story on this page will be built on.
20
native file formats
7
frameworks & platforms
2
Claude models — Haiku 4.5 default, Sonnet 5 for Quality Mode
40
glossary terms per chunk, forbidden-first ranked
Questions about switching
Who transglot fits, how to migrate off a spreadsheet or a legacy TMS, when you need the glossary and QA, and what self-hosting means today.
Not yet — transglot is early, and every logo, quote, and metric on this page is a clearly labeled placeholder showing how real stories will appear. We will replace them with real, user-provided customer content as teams ship with transglot. Nothing here is fabricated.
