Editorial
The Gist
- Compliance is a moving target. Rapidly shifting privacy, accessibility and industry regulations mean outdated documents are now a direct regulatory and legal risk.
- LLMs have already read your content. Generative AI may confidently repeat old or incorrect information from your PDFs as if it were fact, amplifying reputational damage.
- Structure is the safety net. Markup-based, structured documents make it possible to update critical components once and cascade accurate wording everywhere.
- Unstructured documents are the weak link. Static, “print-first” files are harder to update, less accessible, more expensive to maintain and harder for GEO and LLMs to interpret.
- AI finally makes fixing this realistic. Modern tools can convert legacy content, enforce structure going forward and help brands regain control of their online document estate.
The strict regulatory landscape and rise of generative engine optimization (GEO) means brands must ensure their online content is up to date and compliant. But all too often online documents remain out of date and present a risk – structuring documents with the right markup can make all the difference.
The need to ensure your online content is up to date and fully compliant continues to grow more important. Two main reasons: an increasingly complex regulatory landscape and the LLMs that have likely already analyzed your content.
Regulations around areas such as privacy and accessibility as well as specific industry considerations are getting more complex, evolving at different levels across countries, states in the US and regulated sectors. For larger enterprises working across jurisdictions, compliance teams have got to be looking forwards, backwards, up, down and probably sideways too as new legislation drops.
Table of Contents
- FAQ: Structured Documents, Compliance and Generative AI
- Keeping Brands Visible With Right Message in Generative AI Era
- Why Documents Get Overlooked
- The Problem With Unstructured Documents
- What Brands Need to Do to Structure Their Documents Going Forward
- Dealing With the Past
- A New Approach to Managing Structured Documents
FAQ: Structured Documents, Compliance and Generative AI
This FAQ frames the core issues behind structured document management in a GEO and AI-driven world and explains why brands can’t afford to treat PDFs as “done” anymore.
Keeping Brands Visible With Right Message in Generative AI Era
Secondly, generative AI is rapidly changing how we search for and consume information and generative engine optimization (GEO) is rising as a discipline. Marketing teams need to work to minimize risk around compliance and ensure their brand is visible with the right message.
Large Language Models (LLMs) will have already parsed and interpreted your content and may be presenting out-of-date or erroneous details confidently as the established truth.
Dipo Ajose-Coker, solution architect & strategist at RWS Group, and for many years a technical writer in the healthcare sector, commented, “Generative AI has already read your content; the real question is, what story is it telling? The risk isn’t simply that your documents are out of date; it’s that AI is confidently repeating what’s no longer true.” Ajose-Coker added that this means while previously compliance was focused on meeting today’s regulations, it now means anticipating how AI will interpret content tomorrow.
Overall, this presents a reputational and compliance risk and therefore a loss in value, as organizations will need to act and spend money to address inconsistencies and reduce the chance of a regulatory or brand issue. For example, consider a regulated industry such as pharmaceuticals and an old document that refers to a drug being used in a jurisdiction where it has been subsequently withdrawn, a fact that keeps on getting cited by generative AI. This has potential serious regulatory issues.
Structured vs Unstructured Documents
A comparison of risk, effort and long-term value across document types.
| Category | Structured Documents | Unstructured Documents |
|---|---|---|
| Definition | Tagged with markup (XML/HTML-like) that enables updating, reuse, accessibility, and GEO optimization. | Static files, often PDFs, lacking markup and disconnected from source content. |
| Compliance Risk | Low — individual components can be updated globally and remain traceable. | High — outdated details persist, creating regulatory exposure across jurisdictions. |
| Brand Risk | Low — consistent wording and messaging flow across all versions. | High — inconsistencies spread quickly through LLMs and GEO surfaces. |
| Cost to Maintain | Lower long-term — centralized updates cascade automatically. | Higher — manual updates across multiple files and channels. |
| Scalability | High — global variants (e.g., per country) can be auto-generated. | Low — each variation requires independent edits. |
| Accessibility | Strong — markup supports assistive technologies. | Weak — PDFs often fail accessibility standards. |
| GEO/LLM Visibility | High — clear markup improves AI interpretation and brand presence. | Poor — LLMs misread or hallucinate from outdated static content. |
| Ideal Use Cases | Product manuals, service documentation, contracts, global templates. | Legacy files, one-off documents, printable artifacts not optimized for digital use. |
Why Documents Get Overlooked
Managing your content so it is up to date and fully compliant has always been more challenging for documents when compared to web pages.
Why Updating Documents Becomes So Complex
There are multiple reasons for this, often caused by silos that make changing documents a more complex task than it should be, compared to web pages that are easily editable in your Content Management System (CMS):
- Sometimes you have to go back to the original system the document was created in such as Adobe InDesign to make the change.
- Documents tend to have different owners than the web or digital marketing team, so there is a more complicated loop to get the sign-off to action updates.
- Documents can be held on third-party sites with unclear paths to make any necessary changes.
- Documents tend to just get overlooked because they tend to more hidden from view, and sometimes a team may not even be aware of a large number of documents that make up part of a “hidden” digital estate.
- Sometimes documents need to be timestamped for regulatory reasons which makes updating them more complicated.
- The number of documents or equivalent is increasing – video transcripts for example which are rarely checked – and the sheer volume means this is a “wicked problem” (one estimate suggests there are a mind-boggling 15 trillion documents in the world!).
Related Article: Digital Accessibility Drives Customer Loyalty and Inclusion
The Problem With Unstructured Documents
But the biggest problem of all is that so many documents are unstructured.
Why Structure Matters in a GEO and Compliance World
Structured documents can be defined as those that have been appropriately tagged using a mark-up language like XML in the same way a web page uses HTML that can them be optimized for compliance, online distribution and ongoing management.
Using mark-up on particular sections or components of a document means that within the right document management system it is possible to update certain sections of a document that are critical and relate to compliance, such as safety instructions in a product manual. Any sections or particular statements to be updated which can then flow through all the related documents. This is essential to ensure branding and wording is consistent and up to date when there an urgent change is required.
Working with structured documents also has other advantages too, for example where you need to localize particular documents at scale such as adding a different contact name and details for every country-level version of a document.
The right markup within a document can also support accessibility so that a PDF when displayed online can be understood by assistive technologies in the same way as web pages.
Markup as a Bridge Between Compliance and Visibility
Markup schema is also a key part of Generative Engine Optimization (GEO) in order to help the LLM make better sense of the different parts of your document, extract valuable content, and ultimately drive brand visibility.
Unstructured documents, particularly those in PDF format, lack markup language. They are frequently generated using a “printable” mindset that does not take into account them being viewed digitally or needing to be modified. According to Ajose-Coker, they are often regarded as a final output. “Many organizations still treat PDFs as the end of the content journey; but a PDF is only a static snapshot of information that was once verified but is now disconnected from its source.”
However, in the current AI-enabled world this approach represents a risk, particularly in regulated industries. Ajose-Coker said “Traceability and trust demand content that is componentized, where every piece of information can be individually verified, attributed and kept up to date. Every time content is locked in a format that can’t be easily searched, updated or verified, it becomes a liability.”
With a growing collection of unstructured documents, the result is content that is potentially far more expensive to manage and update, less likely to be accessible or optimized for GEO, and presents a greater compliance and reputational risk.
Related Article: Why AI-Optimized Websites Win Higher Conversion Rates
What Brands Need to Do to Structure Their Documents Going Forward
The obvious solution to this issue is to try to apply structure to documents. At first glance this might seem impossible or incredibly expensive to do at scale, but AI is making this more achievable. Here it’s easier to consider the problem in two parts:
- Ensuring documents going forward are structured.
- Converting unstructured documents form the past into structured documents for the future.
Webinar
Nov
20
How to Build a Solid Knowledge Foundation for AI Success
See how leading brands keep their AI honest, compliant and actually helpful.
Register
Webinar
Dec
2
From Manual to Magical: How AI Transforms CX Teams
Learn how to replace manual support processes with automation that actually delivers.
Register
Webinar
Dec
9
Rebrand. Migrate. Optimize. How to Do It All (Without Slowing Down)
Cresta leveled up site speed, design flexibility and marketer sanity (in record time). Find out how.
Register
Webinar
Dec
10
Beyond Composability: How Modern Marketers Build Connected Experiences
Ready to launch campaigns faster, personalize smarter and prove your marketing ROI? Discover the power of a modern DXP.
Register
Webinar
On demand
Fix the Content Bottleneck: Build a Better WebOps Strategy
Content stalled? Dev overloaded? You’re not the only one. Learn how streamlined WebOps bridges the publishing gap.
Watch Now
Webinar
On demand
Beyond Storage: Smarter Content, Bigger Impact with DAM + AI
Discover how the DAM + AI duo makes content smarter, stronger and more accessible.
Watch Now
Building a Framework for Structured Authoring
Let’s consider the future, first. There’s considerable work to set up a system from scratch that truly produces structured documents, but it reduces risk, drives efficiency and presents new possibilities around document management. At a very high level:
- Identify the need and define the scope: Perform an audit and identify document which are most at risk and would be benefit from being structured – these might be your product manuals, service overviews, contracts, or high-profile reports, for example. Analyse where the most sensitive components are that usually need to be updated globally.
- Seek the right solution: Acquire a true technical authoring environment or publishing solution that allows for documents to be structured at scale and has the right levels of granularity for your needs. There are multiple elements to consider in procuring the right solution, including ease of use, scalability and even AI features that can support the authoring process, convert previously unstructured documents and spot potential compliance issues.
- Establish a governance framework: Enable governance with defined roles, responsibilities and review processes to manage global changes when needed and how this flows out to documents available across different channels. Underpin the approach with the right policies and protocols such as audit trails to support compliance and metrics to drive improvement.
- Set up templates: Create the right publishing templates that enable you to recreate your reports as structured documents that will support compliance, but also potentially to optimize for GEO.
- Make it happen: Establish the right training and support to make it all happen and then evolving and tweaking the system using metrics and real-world feedback. Consider starting small and expanding the number of documents covered by the approach as the process stabilizes.
Ajose-Coker pointed out that bringing this approach to the document ecosystem can change the entire mindset around managing documents. “Structured authoring changes how organizations think about documents: you stop editing files and start managing knowledge. Each update becomes a single source of truth that cascades automatically through every version, in every language, across every channel."
Dealing With the Past
How AI Converts Legacy Documents at Scale
But what about the vast number of unstructured documents that already exist? AI is already starting to eliminate this problem with various solutions and agents that can, at scale:
- Convert unstructured documents into structured ones
- Extract data from unstructured that can be integrated into a structured authoring environment
- Be given a URL of an unstructured document and instantly produce a structured document that is also fully accessible and more GEO-friendly
- Crawl your digital estate to find hidden unstructured documents you might not be aware of and give them appropriate structure.
Extracting structured data from unstructured documents is nothing new. However, the speed, volume and levels of accuracy that today’s AI-powered solutions are able to deliver is a game-changer and means getting your online documents back under control is actually achievable.
Ajose-Coker also said that AI is finally reducing the barriers to dealing with the existing unstructured documents:
“For years, legacy documents were seen as too messy or too expensive to fix. The cost of inaction often outweighed the cost of conversion. But AI has changed that equation. It gives organizations a second chance at their content history. Instead of letting legacy documents fade into risk and irrelevance, we can now bring them into a structured ecosystem where they’re searchable, traceable, and ready for the next generation of AI tools.”
A New Approach to Managing Structured Documents
We all need much greater discipline around how we manage online documents — and that means applying structure.
Where AI and Governance Meet
The rise of generative AI will force organizations who have not yet applied a systematic approach to document management to apply far more rigor. Thankfully, AI and the evolution of CMS platforms and structured authoring solutions are lowering the barriers and making this realistic in terms of cost and effort. Organizations can use solutions available today that deal with documents from the past and those that will be created in the future, supporting compliance, reducing risk and supporting GEO.
Learn how you can join our contributor community.
