Go deeper on all five stages, with the full maturity matrix, real enterprise examples, and what unlocks each leap. Everything you need to plan your next move.

The Multilingual Content Maturity Model, Reimagined for the Agentic AI Era

For each, choose the statement that best matches your organization today.

Answer 9 quick questions

Every choice scores 1–5. Your stage is the rounded average of all nine.

We map each answer to a stage

See where you sit on the five-stage model and what to focus on next.

Get your maturity stage + next move

By continuing you agree to be contacted by LILT. We never share your data.

Get your maturity stage plus a tailored set of next steps.

Job title

Global Content Marketing

We put it through a dedicated translation platform (a TMS) that routes and tracks the work.

We send it to an outside translation agency ((a language service provider, or LSP).

We run it through custom AI models tuned to our brand and terms.

Someone grabs a free tool (like Google Translate) or emails a freelancer when the need comes up.

Anyone can ask our company-approved AI assistant to do it, right where they already work.

Our own custom, brand-trained AI models, with a person reviewing the output.

One freelancer, or whoever happens to be free.

An in-house team coordinating vendors and machine translation.

Our company-approved custom AI models and agents do it automatically, bringing in human experts only when needed.

A dedicated translation platform (TMS) using off-the-shelf machine translation.

Free consumer tools (Google Translate, public ChatGPT).

Our company-approved custom AI models, available inside the AI tools and systems we already use (via agents / MCP).

Custom AI models trained on our own content and brand that improve over time.

Agents run quality checks automatically and escalate to human experts only by set rules.

We usually don't, until someone complains.

People on our team spot-check some of it by hand.

AI review agents catch and fix issues first, then a person verifies, and that feedback improves the AI models.

We trust the agency to handle quality; we can't see inside it.

We've moved lower-risk content to AI-only on our own custom models, with AI review as the safety net.

The agency decides what's automated; we don't control it.

We use generic AI tools here and there, but a person still reviews almost everything.

Most of it is AI-only on our company-approved custom models, with agents optimizing for quality, cost, and speed.

Everything needs a person; there's no AI involved.

Anyone can self-serve on-brand translation inside the tools they already use, powered by our company-approved AI models.

Whoever needs something handles it themselves, ad hoc.

A central team owns the process (or a designated owner coordinates it).

It funnels through one department or an outside agency.

Our custom AI models let key content creators do it themselves, on-brand.

It varies every time; there's no shared glossary or memory.

A living context database, enriched with every translation, automatically enforces our voice and terminology across every team and tool.

The agency keeps some glossaries, applied inconsistently.

We keep a shared glossary and a memory of past translations that everyone reuses.

Custom AI models trained on our voice keep it consistent automatically.

Our translation platform reports on projects and volume.

We see agency invoices after the fact, but little else.

We track AI model accuracy, quality scores, and spend.

We have real-time, detailed visibility across AI models, data, and compliance.

It's an invisible layer our tools and agents can tap on demand, anywhere.

Connectors or API move files between our systems and the translation platform.

Our custom AI models plug into a few key workflows.

It's a separate hand-off to an outside agency.

Translation happens on demand with free tools and freelancers. There's no shared memory, no consistency, and quality is invisible until something breaks.

Reactive & Ad-Hoc

You've handed volume to outside agencies. Coverage improves, but cost is high, turnaround is slow, and you don't truly own your data or your translation history.

Managed & Scaled

A central team or platform coordinates the work, but content is still siloed and your AI is generic. People still move files in and out by hand.

Systematized & Tech-Enabled

Custom AI models with human verification keep content on-brand, and AI review agents start capturing real efficiency. Rising quality gives you the confidence to move lower-risk work to AI-only, but access still stops with a small group.

Strategic & Contextualized

Quality has driven most of your translation to AI-only, and agents orchestrate the work, optimizing for performance and cost and calling human experts only when needed. Through MCP, on-brand translation becomes an invisible utility inside the tools every employee uses.

The AI-Native Multilingual Content Maturity Model

Answer 9 quick questions

We map each answer to a stage

Get your maturity stage + next move