Structured source of truth
Keep IDs, source and target text, context, dependencies, state, history, and permissions together so a source change can make affected translations visible.
O05O13O14O15O16U03U07U08Gridly
Keep changing game text, translators, and builds in one traceable localization flow.
Evidence statusPublic-evidence reviewMedium confidenceLast checked Sep 9, 2026By MakeGameWithAI
PRODUCT OVERVIEW
What Gridly is—and how it works.
Gridly is a cloud collaboration platform for structured multilingual content. Game teams organize source text, stable IDs, target languages, context, characters, limits, screenshots, status, and other metadata as records and fields, then use Views, dependencies, history, branches, automations, tasks, and APIs to control how content changes and moves.
The product is sold as three modules that can be purchased separately. CMS is the structured content source and automation layer; TMS manages translation jobs, workflow, progress, and delivery; CAT gives translators an editor with context, translation memory, search, and suggestions. A team may need one module or a combination—three price cards do not mean everyone must buy all three.
Gridly's generative features sit inside that workflow. AI-assisted translation and Pre-translation are currently documented as Beta, with translation memory and machine translation followed by QA and human review. The separately marketed Localization Agent says Launching Fall 2026, so this page does not treat it as generally released. Gridly's durable value is content coordination and delivery, not a promise that AI produces final game localization by itself.
HOW YOU USE IT
A 14-day Team trial is publicly described for all modules, after which help material says the account becomes read-only until a plan is selected. The current public material reviewed here does not confirm whether a payment card is required. API allowance and features depend on the purchased tier and module.
Separate CMS, TMS, and CAT roles
Buy content management, translation management, translator tooling, or a combination according to the actual handoff rather than treating the full feature list as the minimum plan.
O01O03O05U07Files, Unity, Unreal, and API delivery
CSV and JSON routes, official Unity and Unreal packages, and REST record operations create real ways to move approved content into a game. Mapping, stable IDs, secrets, imports, and engine checks remain project work.
O13O14O15O17O18O28O29U02U05U11U12AI and MT as pre-translation, not approval
Beta tools can suggest, rewrite, pre-translate, and run rules. The final language, character voice, cultural fit, variables, and in-game layout still need qualified human review.
O07O08O11O12O21U07INTERFACE & EXAMPLES
See how it works
A frame from Gridly's official walkthrough shows game strings organized into records and language columns.
Official source retrieved September 2026; interface version not specified.Source: Gridly Help Center
REVIEW SCOPE
Continuously changing game text with multiple target languages, collaborators, translators, or repeated builds. This assessment does not promise one-click translation quality, current large-table speed, seamless engine synchronization, or cost effectiveness for a small one-off solo project.
- Beginner friendlinessSome basics needed
- Import guides and a structured interface provide a discoverable start, but users need stable keys, source and target fields, dependencies, Views, permissions, QA, and localization handoff concepts to use the system safely.
- Agent integrationEasy to integrate
- Official REST documentation, permission-scoped API keys, record operations, and official SDK or engine routes provide a practical path for Codex or Claude Code to read and update localization data. Product-internal AI and the future Localization Agent are separate capabilities.
Scope, integration requirements and sourcesChecked
Beginner scope: Structuring and coordinating changing game text for a small multi-language project, not guaranteeing translation quality for a complete release.
Agent scope: Let an external agent read or update scoped Gridly records as part of a localization pipeline, with human approval before production delivery.
Confidence: learning Medium · Agent Medium
- Gridly REST APIOfficial
- Gridly API SDKsOfficial
- Access
- Create a read-only or restricted API key, choose the endpoint scope supported by the purchased tier, keep the key server-side, and expose only the Grid, branch, and operations the task requires.
- Billing
- API requests and automated updates have tier limits; Basic publicly lists 3,000 API requests and 2,500 automated updates per month. Modules, seats, add-ons, upstream AI or MT providers, overages, and the calling agent are separate cost layers.
- Getting results
- Authenticate with a scoped key, fetch records and schema, propose or write bounded field changes, preserve stable IDs and review states, then let a human approve before the content is exported or synchronized into a build.
- Limitations
- Easy means documented composability, not zero setup. The caller still owns schema mapping, secrets, retries, rate limits, deletion safety, conflict handling, language review, provider billing, and game-build validation.
These are independent editorial judgments, not an overall score or a claim of hands-on agent integration testing.
- Gridly localization export and import · Official · Learning evidence
- Gridly JSON workflow · Official · Learning evidence
- Gridly CSV import · Official · Learning evidence
- Working with Views · Official · Learning evidence
- Gridly first TMS experience · Independent · Learning evidence
- Gridly versus Google Sheets migration account · Independent · Learning evidence
- Gridly REST API documentation · Official · Integration evidence
- Create and manage Gridly API keys · Official · Integration evidence
- Gridly Unity Localization package · Official · Integration evidence
- Gridly Unreal Engine 5.8 plugin · Official · Integration evidence
- Gridly pricing and API limits · Official · Integration evidence
- API publishing and engine-integration user records · Independent · Integration evidence
KEY FINDINGS
What the evidence supports—and what it does not.
Editorial inferenceMedium confidence
Coordination is the core product value
Across professional accounts, reviews, and official workflows, the recurring benefit is keeping context, language state, changes, people, and delivery connected—not generating isolated translation lines.
O05O13U01U02U03U04U06U07U08Editorial inferenceMedium confidence
Small projects can fail the value equation
At €50–€140 per month before VAT for one Basic module, the platform needs enough changing content and coordination cost to offset subscription and migration work. A low-key-count selection record illustrates that threshold without proving the product is overpriced for everyone.
O01U05U10EDITORIAL VERDICT
Our take
Medium confidenceGridly is most useful when localization is an ongoing content operation rather than a one-time export. It can replace fragile spreadsheets and file handoffs with stable records, context, workflow, QA, and programmable delivery, but that value grows with source-text change, language count, and team handoffs. A beginner can follow the import and interface path, yet needs some localization and data basics. Its documented REST API and controlled keys make external-Agent integration relatively easy; the Agent must still respect the schema, plan limits, secrets, review states, and human language approval.
Conclusion scopeMedium-confidence public-evidence assessment of localization coordination and integration. It does not generalize vendor case savings, machine-translation quality, large-table performance, or plugin reliability to every project.
BETTER FIT
Worth auditioning when
- A live or repeatedly updated game whose source text changes after translation begins.
- A small studio with several target languages and handoffs among narrative, development, production, and outside translators.
- A localization team that needs translation memory, terminology, jobs, QA, permissions, and history around game content.
- A pipeline that benefits from controlled REST access, automation, or official Unity and Unreal data routes.
POORER FIT
Do not depend on it yet when
- A few hundred stable strings translated once by one person with no ongoing release cycle.
- Expecting AI or MT to deliver culturally correct, production-ready game text without native review.
- A team unwilling to define stable keys, source and target fields, ownership, review state, and export rules.
- Choosing from the lowest headline price without checking modules, seats, API limits, VAT, add-ons, upstream providers, and migration work.
WORKFLOW FIT
Where it fits in a production workflow.
Model the source before translating
Define stable IDs, source language, target fields, context, variables, length limits, ownership, and review states; import a small representative file first.
GuardrailExport a backup and preview mapping; deleting a record in one View deletes the underlying Grid record rather than merely hiding it.
Translate with traceable assistance
Reuse approved TM and terminology, optionally create Beta pre-translations, mark machine output, and assign it to a qualified reviewer with QA rules.
GuardrailDo not route unreviewed AI or MT text directly to the production branch, and confirm every enabled provider's data and billing terms.
Deliver and observe through one controlled route
Export a reviewed file or use one official plugin or REST flow, compare IDs and conflicts, then verify variables, layout, fonts, and fallbacks in the game build.
GuardrailUse least-privilege server-side keys, keep live-preview credentials out of repositories and shipped clients, and provide retry and rollback behavior.
RECOMMENDED WORKFLOW
Use this tool inside a complete production workflow.
Start from a testable brief, then move through prototyping, assets, audio, testing, and release with an explicit handoff and human check at every step.
SHORTEST RESPONSIBLE PATH
Prove the changing-text loop before migrating the whole localization stack.
This is a low-risk starting workflow synthesized from the evidence.
- 01
Choose twenty representative strings with stable IDs, variables, context, limits, and two target languages; use non-sensitive sample content.
- 02
Import them into one Grid and make three source-text changes to see whether affected translations and owners remain visible.
- 03
Run terminology and QA on deliberately planted number, variable, punctuation, and length issues; keep machine suggestions unapproved.
- 04
Export and reimport a small reviewed batch, then choose either Unity, Unreal, or REST for one controlled delivery—not all routes at once.
- 05
Check the game build, export a complete backup, revoke test keys, record the real handoff time, and only then decide which modules and seats are worth migrating.
PRICING & RIGHTS
Pay by module, seats, tier, and optional usage
Pricing and terms last checked: Sep 6, 2026
- Basic TMS is €50 per month for one included paid seat; Basic CMS is €99 for three seats; Basic CAT is €140 for ten seats. VAT is excluded.
- Buying all three Basic modules at their public base prices totals €289 per month. This is an illustration, not an official bundle or required minimum.
- Team CMS, TMS, and CAT list at €249, €125, and €280 per month; all three total €654 before VAT and additions.
- Basic lists 2,500 automated updates and 3,000 API requests per month, 10,000 records per Grid or Job, and 500,000 total records. The page does not establish that buying multiple modules multiplies every limit.
- The public trial is fourteen days with Team features and all modules; the account is described as read-only afterward. Card requirement and reproducible annual prices remain unknown.
COST TO OUTPUT
Character rates measure machine processing—not accepted localization
Public add-on rates checked September 6, 2026. Calculations are linear, before any listed free allowance, base modules, VAT, target-language multiplication, upstream API bills, translation-memory reuse, retries, or human review.
O01O02O03O04O09O10The listed rate is €19.95 per million characters. These are machine-processed characters, not approved strings.
The listed rate is €24.95 per million characters. The 10,000-character allowance is not applied because its reset interval is unclear.
The listed rate is €49 per million characters. Current BYOK documentation leaves possible upstream OpenAI billing unresolved.
Basic TMS plus one nominal million OpenAI characters. It excludes CMS/CAT, seats, human review, and any upstream API charge; it is not a recommended bundle.
Cost per accepted target string equals module and seat fees, Gridly add-ons, upstream providers, implementation, translation, review, and maintenance divided by strings that pass human review and enter the build. Public character rates do not provide that denominator.
Commercial-use condition
Gridly is sold for commercial content operations, and its terms say customers retain ownership of Customer Content while granting service-related rights. The customer must have rights to uploaded source text, translations, screenshots, and other materials. AI output can be inaccurate or non-unique, and third-party provider terms may also apply.
Paid terms, renewal, cancellation, refunds, order forms, logo publicity, provider data terms, data region, and exact checkout totals need account- and contract-level review before migrating confidential production content.
PRODUCTION RISKS
Resolve these before production use.
Workflow lock-in remains after standard exports
CSV and JSON preserve content, but Views, dependencies, branches, tasks, comments, tickets, permissions, automations, TM, and glossary do not automatically migrate as one equivalent system.
O13O14O15O19Keys, deletes, and bulk changes need controls
Restrict and revoke API keys, keep them out of clients and repositories, preview imports, export backups, and separate development live-preview access from production delivery.
O16O17O18O28U11U12Provider and data terms are fragmented
The older provider disclosure and newer integration-key list differ, while AI features can invoke third parties. Confirm the enabled provider, region, retention, training, deletion, and billing path for the exact action.
O09O10O20O21O22Migration and training can dominate a small project
Stable schemas, roles, imports, plugins, and review rules take setup time. For little, stable text and one translator, engine-native files plus version control may remain simpler.
U02U04U05U08U10NOT VERIFIED
Claims this page does not make
- Current Unity or Unreal round-trip success, conflict behavior, large-table performance, and time saved in the target project.
- Whether the fourteen-day trial requires a card and the current reproducible annual prices or discounts.
- Whether every current OpenAI workflow charges both the listed Gridly character rate and the customer's upstream API account.
- Current AI model, region, retention, training, deletion, quality by language pair, and general availability of Localization Agent.
- Human translation, post-editing, cultural review, layout remediation, implementation, migration, and maintenance cost per accepted target string.
RESEARCH METHOD
Research-reviewed from public evidence
Research-reviewed from public evidence. We coded twelve records across G2, MultiLingual, Localization Academy, Reddit, and GitHub, spanning a review platform, named professional accounts, an independent first-use walkthrough, a community selection record, and technical issues. Six G2 records are exactly half of the set and retain invitation or incentive labels. Vendor customer stories, brand logos, and alpha-pilot savings were excluded from independent evidence. The record supports a scoped workflow and integration judgment, not a quality, performance, or savings benchmark.
Research windowReviewed September 6–9, 2026. Twelve independent records cover five platforms and four source types; six G2 records are exactly half of the set and their invitation or incentive status is retained.
OFFICIAL SOURCES29
- O01Gridly pricingChecked Sep 9, 2026
Current CMS, TMS, and CAT module prices, included seats, tier limits, and trial statement.
- O02Gridly add-on pricingChecked Sep 9, 2026
Listed machine-translation, lexiQA, CDN, and other add-on rates; the reset and dual-billing boundaries are not fully explained.
- O03Gridly billing overviewChecked Sep 9, 2026
Company-level billing, advance payment, periods, and seat concepts.
- O04Gridly pricing and payments FAQsChecked Sep 9, 2026
VAT, trial, tier, seat, and purchasing details.
- O05Gridly general FAQsChecked Sep 9, 2026
Product modules, game-localization positioning, and general workflow scope.
- O06Gridly What's newChecked Sep 9, 2026
Current maintenance and feature timeline. Release notes do not establish stability in every account.
- O07Gridly AI-assisted translationChecked Sep 9, 2026
Cell-level translation, rewrite, shortening, variation, correction, sentiment, and custom-prompt features marked Beta.
- O08Gridly pre-translation wizardChecked Sep 9, 2026
Beta flow combining translation memory, machine translation, markings, and QA.
- O09Gridly OpenAI Text TranslationChecked Sep 9, 2026
Current automation documentation requires the customer's own OpenAI API key and says model cost follows that account.
- O10Gridly integration API key managementChecked Sep 9, 2026
Third-party provider key configuration and current provider list.
- O11Gridly AutoQAChecked Sep 9, 2026
Rule-based checks for omissions, numbers, punctuation, and other defects; not a guarantee of linguistic quality.
- O12Set up LQA modelsChecked Sep 9, 2026
Human quality evaluation models, categories, and ticket workflow.
- O13Localization export and importChecked Sep 9, 2026
Localization file export, modification, preview, and reimport workflow.
- O14Working with JSON files in GridlyChecked Sep 9, 2026
JSON import and export behavior for structured content.
- O15Import CSV filesChecked Sep 9, 2026
CSV field mapping and import behavior.
- O16Working with ViewsChecked Sep 9, 2026
View filtering and the important distinction between hiding and deleting underlying records.
- O17Gridly Unity Localization packageChecked Sep 9, 2026
Official open-source Unity data synchronization package and setup requirements.
- O18Gridly Unreal Engine 5.8 pluginChecked Sep 9, 2026
Current UE5.8 plugin, manual setup, language mapping, and import/export route.
- O19Gridly Terms of UseChecked Sep 9, 2026
Customer-content rights, subscription term, renewal, refund, liability, and customer-logo conditions. Not legal advice.
- O20Gridly Privacy PolicyChecked Sep 9, 2026
Data categories, processors, international handling, deletion, and retention context.
- O21Gridly AI TermsChecked Sep 9, 2026
Inputs and outputs as Customer Data, third-party AI use, non-unique or inaccurate outputs, and user review duties.
- O22Gridly third-party AI providersChecked Sep 9, 2026
Provider disclosure last marked September 2024; it does not fully match the newer integration-key list.
- O23Gridly security and complianceChecked Sep 9, 2026
Vendor statements on certification, encryption, hosting, and backups; not an independent audit by this site.
- O24Gridly security helpChecked Sep 9, 2026
Additional account and security documentation.
- O25Gridly statusChecked Sep 9, 2026
Public service-status entry. A current green state cannot prove historical or account-specific availability.
- O26Gridly Localization AgentChecked Sep 9, 2026
Marketing page says Launching Fall 2026 and offers early demos; it is not treated as broadly released.
- O27Getting started with Localization AgentChecked Sep 9, 2026
Early product documentation that does not override the launch status on the product page.
- O28Create and manage Gridly API keysChecked Sep 9, 2026
Read-only and restricted key controls, regeneration, disabling, and revocation.
- O29Gridly REST API documentationChecked Sep 9, 2026
Official REST record operations, API-key authentication, and plan-level access boundaries.
INDEPENDENT EVIDENCE RECORDS12
- U01G2: API publishing workflowChecked Sep 9, 2026
An incentivized small-business review describes publishing game and competition content through the API, with slowdown at tens of thousands of rows.
- U02G2: enterprise localization and Unreal workflowChecked Sep 9, 2026
An incentivized localization-director review values centralization and plugins while reporting Unreal mapping and import conflict friction.
- U03G2: narrative designer daily workflowChecked Sep 9, 2026
A natural-review record says daily changes are visible in the game and Views help, while the interface often feels slow.
- U04G2: game automation and Unreal integrationChecked Sep 9, 2026
A computer-games review, updated in 2025, describes automation benefits alongside loading, group, and permission friction.
- U05G2: small-team Unreal workflow and pricingChecked Sep 9, 2026
An invited small-business programmer finds direct Unreal transfer approachable but wants a free or single-seat option.
- U06G2: localization operations and added costsChecked Sep 9, 2026
An invited localization lead describes routine import, export, versioning, and automation value alongside missing operations and separate ChatGPT cost.
- U07MultiLingual: machine translation for game localizationChecked Sep 9, 2026
A named Social Quantum localization lead describes CMS, API, MT plus human editing, and language-dependent results. One team cannot define a universal saving rate.
- U08MultiLingual: Gridly versus Google SheetsChecked Sep 9, 2026
A three-game localization migration account values TM, glossary, roles, history, and repository connections while documenting training and process change.
- U09Localization Academy: first Gridly TMS experienceChecked Sep 9, 2026
An older shallow first-use walkthrough supports discoverability of basic project, language, and export paths, not current performance.
- U10IndieDev localization-tool selectionChecked Sep 9, 2026
A developer with about 1,200 keys across small projects sees the subscription as expensive. They did not report hands-on Gridly use.
- U11Gridly UE4 plugin issue #6Checked Sep 9, 2026
A historical key-in-build issue supports least-privilege and production-separation safeguards; it is not asserted against current UE5.8.
- U12Gridly UE5.2 plugin issue #3Checked Sep 9, 2026
A historical generated-ID import issue makes stable IDs and previewed imports important test points, not a current universal defect.
DISCLOSURE
How this research was supported
Independent editorial research. MakeGameWithAI has no affiliate link, sponsorship, vendor account, extended trial, free credits, or technical support for this review. Owner Review approved the research and the two scoped decisions on September 9, 2026.
This public research review will be revisited when pricing, terms, product versions, or material new evidence changes.