How Our Plugin Brings Something New

This week, we continue our exploration from last week, diving deeper into our thought process and strategic choices. If you're interested in the full context, check out Revolutionizing Localization: Our Journey, Challenges, and Ambitions. Now, let's dive into our new product that's officially live and completely free to try!

Why Localization Belongs in Your Design Tool

Let's face it: the traditional localization workflow is broken. Designers create in Figma, developers code in their IDE, and somewhere in between, localization gets treated as an afterthought—leading to misalignment, duplicated work, and endless back-and-forth. Even the current best localization software fail solving this pain.

Our premise is simple but powerful: since developers already go to Figma to see what they need to code and have to create (or reuse) localization keys while developing, why not have these keys automatically managed and displayed right where the design happens?

Why Traditional Figma Localization Plugins Fall Short

Let's be honest: most existing Figma plugins approach localization in a fundamentally flawed way. Tools like Lokalise and Crowdin (which are A players in the localization space) rely exclusively on frame names to automatically generate keys, resulting in nonsensical outputs like Frame15034.Frame182.forgot-password.

Some solutions suggest that designers should meticulously name every frame to improve key generation—but that's like asking painters to label every brushstroke. It's inefficient, unsustainable, and completely misaligned with how design teams actually work.

The real problems with these frame-based approaches are:

1. They waste designer time: Forcing designers to create and maintain naming conventions for localization takes them away from their actual job—designing great experiences.
2. They break easily: Duplicate a frame for a different state or component variation, and suddenly your carefully crafted naming system collapses.
3. They don't scale: As your product grows, maintaining consistent frame names across dozens of files and hundreds of screens becomes virtually impossible.

After countless conversations with frustrated design teams, we've heard the same thing repeatedly: no one has time to micromanage localization keys through frame naming. Frame-based key generation simply doesn't account for how modern design teams operate.

Our Approach: Teaching Machines to Think Like Designers

What if we flipped the script? Instead of forcing humans to think like machines with rigid naming conventions, what if we taught machines to understand design context like humans do?

That's exactly what we've done with Gleef. We've reimagined localization key management from the ground up, using AI to understand your designs and generate keys that actually make sense. Here's how our approach addresses all the problems we identified in the traditional Plugins and how do we tackle this with our new approach.

• Frame-based naming that produces gibberish vs. Context-aware key generation that recognizes what each text element actually is.
• Duplicate frames create duplicate keys vs. Intelligent duplicate prevention that recognizes similar elements across frames
• Keys have no relation to content vs. Keys reflect the actual purpose and meaning of text elements
• Separate workflow from design vs. Seamlessly integrated into your Figma workflow

How Our Plugin Works

Our plugin offers three core capabilities that transform the localization workflow:

1. Automatic key generation based on design context, following your existing naming conventions
2. Intelligent key retrieval that finds existing keys when appropriate, eliminating duplicates
3. Visual key display directly in Figma, creating a seamless bridge between design and development

We’ll deep dive a bit more into how it technically works in future articles, stay tuned!

We won’t stop our plugin with just keys generation. It is the first step to build a localization plugin right in Figma, but we’ll also include the best ever made translation plugin for Figma - integrated in the same plugin.

The Technology Behind the Magic: How We Use AI

Without artificial intelligence, automating localization keys usually relies on two deeply flawed approaches:

1. Using element names or content verbatim (which leads to those horrific "frame1.frame2..." keys)
2. Generating random unique identifiers (which have zero connection to the actual content)

We take a fundamentally different approach using Retrieval-Augmented Generation (RAG), which combines contextual search with intelligent generation. This allows us to:

• Identify existing keys when they're already available in your database, preventing unnecessary duplication
• Create truly contextual keys when new ones are needed, respecting your established naming conventions

Our AI can identify elements as placeholders, buttons, labels, and even recognize screen types. From there, it generates keys that are perfectly adapted to your project's context. The result? A system that's simple, efficient, and—most importantly—well-organized.

The Power of Context

Context sits at the heart of our approach. Just like the word "remote" could mean a TV controller or working from home depending on context, localization keys need contextual awareness to be truly useful.

This is precisely why we chose to work directly within Figma. By staying in this environment, everyone—designers, developers, and translators—shares the same visual context. This allows us to create keys that are genuinely adapted to your project while maintaining consistency.

Getting Started is Simple

We believe that powerful tools don't need complicated onboarding. Our setup process is intentionally streamlined to get you up and running quickly:

1. Define your source language (the language used in Figma to design screens)
2. Import your existing localization keys file in CSV, XML, or JSON format (if you have them—and there's no limit to how many keys our system can handle!)
3. That's it! You're ready to select any frame and generate localization keys

From there, the tool automatically generates contextually appropriate keys, which you can tweak if needed. Finally, you can export everything ready to use in your localization platform or directly in your IDE.

Why Your Existing Keys Matter (But Aren't Required)

Here's where we want to be completely transparent: our plugin delivers the most impressive "wow moment" when it has access to your existing localization keys. Why? Because this allows our system to:

• Follow your established naming conventions and patterns
• Retrieve existing keys rather than creating duplicates

However—and this is crucial—we understand that many teams are just starting their localization journey and might not have an established key database. Others might have legitimate concerns about sharing their data during onboarding.

That's why we've designed our plugin to be valuable even without existing keys. It will create a consistent, logical naming system from scratch if needed. Importing existing keys just enhances the experience by ensuring perfect alignment with your current standards.

I’ve read a super interesting article last week on this onboarding issue: whether or not ask for data in the onboarding. The study showed two distinct patterns: when users uploaded their data during onboarding, they found immense value in the product and rarely churned. However, without data upload, while more users initially tried the product, they often failed to see its full potential and dropped off later. This presents a classic dilemma for data-dependent products: you either face high dropout rates during onboarding by requiring data upfront, or you risk losing users later when they don't experience the product's complete value proposition.

What's Next: Translation That Understands Context

If you've explored our plugin, you might have noticed hints about our next big feature: translation. But we're not building just another machine translation service—we're applying the same contextual intelligence that makes our key generation so powerful.

We're already in beta phase with this feature. If you're excited about localization that truly understands your product's context, join our waiting list to be among the first to try it.

Conclusion

We believe localization should be an asset for product teams, not a hindrance. Our AI-first approach saves time, prevents inconsistencies, and integrates seamlessly into your existing design workflow.

We're continuously refining Gleef based on user feedback, so if you have ideas or specific needs, we're all ears. Together, we're reshaping how localization works in the modern product development process.

Ready to try it yourself? [Get our plugin here] and experience the difference contextual localization makes.

‍