Opportunities › Localized content management for product teams › Proposed app
Tongue Ledger
Pull-request translation that keeps your glossary, voice and human edits in charge
- Score
- 46
- confidence 42
- Evidence
- 4
- items on the page
- Rivals
- 0
- named in sources
- First version
- 30 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2012
- Mature market · AI estimate
ScreensConcept
Desktop app: Developers review GitHub PR translations, enforce glossary rules, and mix human and AI edits by string—careful approval workflows. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
PR #482 strings
Awaiting approval14 new
About this appAI proposal
Tongue Ledger runs as a GitHub Action. When a pull request adds or changes strings in your .po files, it drafts translations, checks them against your glossary and brand voice notes, and posts the result as a review on the PR. Nothing merges without your approval.
It is built for product teams that keep translations in their repositories, such as Elixir and Phoenix projects using gettext. Human-written translations are locked and never overwritten. AI drafts fill only the gaps, and each one is marked as a draft until a person signs off.
You can also use it only as a checker. Turn off drafting and it flags inconsistent terms, wrong tone and broken placeholders in translations your team or agency wrote by hand.
- Tagline
- Translate on every pull request. Keep every word accountable.
- Voice
- PlainCarefulPractical
- Why this name
- Tongue means language; a ledger records who approved what. The name signals tracked, reviewable translation rather than a black box.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
GitHub Action for .po and gettext
Detects new and changed msgids on each PR, translates only those, and commits or suggests the result on the branch.
No member yet
Locked human translations
Mark any string as human-owned. AI never edits it, and changes to the source text open a review request instead.
No member yet
Glossary enforcement
Define terms per language, including words to never translate. Every draft and manual edit is checked against them.
No member yet
Brand voice profile
Describe tone with a short guide and example pairs. Drafts follow it, and off-tone strings are flagged.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#1F4D3FDeep Pine · primary#F6F3ECPaper · surface#1B1F1DInk · text#D9602BEmber · accent#8FAE8BMoss · data#D8D2C4Stone · border
Typography
Aa
Tongue Ledger
ABCDEFGHIJKLMNOPQRSTUVWXYZ 0123456789
PricingAI proposal
The material gives no competitor prices. These tiers are a starting point to test, scaled by languages, repositories and drafting volume as the opportunity suggests.
Free
Free
- 1 repository
- 2 languages
- Review-only mode
- Glossary up to 20 terms
Team
$49 a month
- 5 repositories
- 10 languages
- AI drafts on PRs
- Locked human translations
- Full glossary
Studio
$149 a month
- Unlimited repositories
- Unlimited languages
- Brand voice profiles
- Consistency view
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$194,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- No loop in the product. It works inside a team's own repository and PRs. Contributors see bot comments, but the output is not published to outsiders, so growth comes mainly from content, community posts and direct outreach.
- Do they become customers?
- Developers reviewing PRs may be teammates who could push adoption, but this is not tested. Test whether a bot comment with a small footer leads other teams or contractors to sign up.
- First customers
- Small Elixir and Phoenix product teams of 3 to 15 developers who keep gettext .po files in GitHub and ship in three or more languages.
- Where to reach them
- Elixir Forum and Slack: Share a free review-only Action and a write-up on glossary-checked gettext workflows; invite teams to try it.
- Hacker News Show HN: Post a demo PR with locked human strings and AI drafts, and ask for feedback from teams shipping multilingual apps.
- Direct outreach on GitHub: Contact maintainers of open, multilingual Phoenix apps and offer to run a free setup on one repository.
- First milestone
- Five paying Phoenix teams on the Team tier within 60 days, each with at least one merged PR using drafts and glossary checks.
Why people already pay for this
- Public claimThe product has paying customers validating demand for automated i18n translation with GitHub integration
- Public claimOne customer is a established product (hiking trail app used for years) indicating market validation in real, mature software products
- Public claimCustomers want manual translation control layered with AI, revealing a key gap: existing solutions lack the ability to mix human and AI translations effectively
- Public claimCustomers request the ability to use the platform primarily for review and consistency validation rather than pure automation, showing demand for control
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · AI translationLarge language model APIDrafts translations using glossary and voice context, with per-string prompts.
- This app · Source integrationGitHub App and ActionReads PR diffs, posts reviews and suggestions, and writes commits to branches.
RisksAI inference
- AIDrafts may be fluent but wrong in domain terms or tone. The glossary and review states reduce this, but quality per language must be tested with native reviewers.
- CommercialTeams may build a simple in-house script with an LLM. The case rests on glossary, locked human strings and review workflow being worth paying for.
- TechnicalRound-tripping .po files must preserve plurals, comments and ordering, or diffs become noisy and trust is lost.
- PlatformDepends on GitHub APIs and Actions permissions. Some companies will not grant repository write access or send source strings to a third-party model.
- DataSource strings may contain unreleased product text. Retention and model-provider data terms must be clear before security-minded teams adopt it.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.