To treat content as infrastructure is to stop viewing words as a static piece of text and start seeing them as a foundational layer of the product.

By systemizing content as data, you move toward a model where copy is an integrated component of the product’s architecture. You scale your work. You grow your influence with other teams. And you create more reliable, impactful copy.

This shift is happening now because AI has democratized creation. In a world where anyone can—and does—generate text at scale, the value of content design has moved upstream. Our responsibility lies in defining the guardrails, taxonomy, and logic that give a product its meaning.

Words as data: What’s next for content design

To explore this shift, we co-hosted a panel discussion with UX Content Collective on what’s next for content design. We heard from some industry experts:

• Dave Connis, Lead Content Designer, Design Systems, OutSystems
• Chris Greer, Staff Content Designer, Stripe
• Our own Jess Ouyang, Co-founder and CEO, Ditto

The panel discussed how treating content as data fundamentally changes the role of the writer from a manual task manager to a strategic architect.

It brought together perspectives from design systems (Dave), internal systems and AI tooling (Chris), and content-first product tooling (Jess), to provide a blueprint for how to start building content infrastructure. The speakers shared practical steps for content designers to exert more influence, from learning technical foundations to aligning work directly with business objectives.

What does “content as data” even mean?

Design has design systems and engineering has components, but text is the last part of product development to be systemized. Treating content as data means systemizing it so we can use it in the same way we do design or code.

This lets us reuse, govern, iterate on, measure, and improve our content.

In turn, this makes product copy, and content design as a discipline, more strategically valuable, too.

"To treat content as data is to understand the fact that it is representative of something, and not just words on a screen," said Dave. He argued that when you write for a system, your strings represent the objects, the metadata, the entire system underneath.

The secret sauce for building content infrastructure? Context.

Chris pointed out an important caveat: "You can only really call it data if you treat it as such... giving it the proper sort of infrastructure and observability... so we can see where this content lives, what the surrounding context is."

Why content becomes a bottleneck as teams scale

Why should we care about content as data—and why now?

• The cost of manual labor: When we treat content as "words on a screen" rather than part of a system, it leads to manual, brittle processes. Jess noted that if you ever find yourself "writing an error message for the 15th time in the same context," it’s a symptom of relying on manual work.
• The AI shift: As AI democratizes creation, the volume of content is exploding. Without a data-driven approach, writers can’t scale content or ensure quality at the speed of engineering teams, who are already building in hours what used to take weeks.
• Meaning becomes a differentiator: If AI has made it possible for anyone to write product text, "meaning matters more," Dave points out. Next-gen content design is going to own this meaning—orchestrating meaning, instead of trying to keep up with copy generation.

What changes when we treat content as product infrastructure?

By making this shift, content designers can be drivers of the product experience, and make other teams better in the process.

Content designers become more strategic

Jess pointed out that with the right system, content and product designers gain a high-leverage position. They can own the product copy, assess adherence, and make tweaks on a larger scale, rather than manually searching and making changes to text. “Everyone is empowered to move closer to business decisions and direct impact,” said Jess.

Stronger collaboration and relationships

It opens the door to new ways of working. Chris shared an example where, instead of working in Figma, he’s switched to collaborating with his product design partner in GitHub, where he’s using a Claude Code UX writing skill. Ultimately, it’s allowed him to speed up copy work and meet his product and engineering teammates where they are.

Improved governance and quality at scale

When you treat content as data, AI can run in the background to enforce your brand standards and style guides. Chris suggested that it also opens the door to assessing content in a quantifiable way. An example he referenced: you can use AI to run a series of binary tests and score your content.

How to build your influence by shifting to content as data

Growing your influence as a content designer requires architecting the systems that power your words.

• Establish and codify your standards: Influence begins with a solid foundation of great content design. Jess cautions that "weak standards result in weak governance," so revisit and codify your style guides, taxonomies, and terminology (one step at a time).
• Set the context for AI: Think through how you can provide LLMs with your specific product context. Define the concepts and objects within your system to help AI tools generate higher-quality content.
• Learn technical concepts: Dave suggested learning the basics of the product development lifecycle, understanding concepts like API and JSON, so you can talk to engineers on their level.
• Align with business objectives: Chris highlighted the importance of business skills—focus on projects that move the needle for the business and tie your work to metrics.

Our key takeaway: Scale the system you own

While AI lets you generate content at speed and volume like never before, content design needs to prioritize the underlying system by treating content as infrastructure.

From there, you can use tooling to free yourself from manual work, so you have more time for taste and craft—your ultimate differentiators.

‍