Brands with strong traditional SEO rankings are getting skipped by ChatGPT, Perplexity, and Google’s AI Overviews. Not because their content is bad, but because it’s structured for a retrieval system that AI pipelines don’t use. Four days at SEO Week NYC 2026 laid out exactly what’s broken and what fixes it.

You’ve likely watched this play out already: rankings hold, but traffic from AI-generated answers goes to competitors. Or a prospect mentions they “looked it up” before calling, more recently meaning they asked an AI. The problem isn’t your SEO. It’s that the signals AI systems use to decide what to cite are often different than the signals that determine traditional rankings, and most brand sites haven’t been built for them yet.

AI Systems Filter Out Most Content Before They Ever Read It

AI retrieval pipelines evaluate content through a series of eligibility checks before a model ever sees it. Content that fails those checks can’t be cited, regardless of its quality or ranking. Krishna Madhavan, Principal Product Manager at Microsoft AI, opened the conference by describing what he called the “invisible, converged web”: a layer of grounding confidence scores, safety filters, publisher controls, and attribution signals that sits between your content and the AI systems your audience is using. If your content doesn’t carry the right signals, it gets filtered out at that layer. It never reaches the model.

This is the structural gap most brands don’t know they have. A page can rank on the first page of Google and be completely absent from AI-generated answers on the same topic, because ranking signals and retrieval eligibility signals are different. Google evaluates pages. AI pipelines evaluate whether individual content blocks are structured, sourced, and verified enough to ground a response. Madhavan’s framing was precise: modern SEO and GEO now feel less like ranking tactics and more like a distributed systems challenge, where the goal is coordinating the signals that let AI pipelines trust and reuse your content.

In every GEO audit we run at Oomph, most brand sites are missing at least two of those signals. The most common gaps are invisible in standard SEO tooling, which is exactly why brands with strong traditional performance are still surprised when their AI citation share is near zero.

Traffic Metrics Will Lie to You About Your AI Search Performance

AI Overviews, ChatGPT, and Perplexity are answering questions in your category and not sending traffic to your site, but your analytics won’t show you that as a problem, because there’s no traffic to track. Jori Ford, Chief Marketing and Product Officer at FoodBoss, introduced the HEO (Hybrid Engine Optimization) framework at SEO Week specifically to address this measurement gap. Her Hybrid Engine Score is a weekly composite that tracks both traditional ranking performance and AI citation performance in a single number. Measuring them separately, or measuring only one, gives you an incomplete and often misleading picture of your actual search visibility.

Dale Bertrand of Fire&Spark extended this into the CFO conversation that most marketing leaders are currently losing. If your traffic is down but AI-influenced conversions are up, you’re actually winning. GA4 misses most AI-driven attribution, so you look like you’re failing. Bertrand’s work with global brands showed that revenue-focused GEO consistently produces stronger business outcomes than traffic-focused SEO when you measure far enough downstream. The brands building that measurement and implementation frameworks now are the ones who’ll be able to defend AI search investment in 12 months, when leadership starts asking why organic traffic hasn’t recovered.

“Revenue-focused GEO consistently produces stronger business outcomes than traffic-focused SEO.”

A Weak Paragraph Loses to a Strong One Every Time

AI systems retrieve at the paragraph level, not the page level, which means every paragraph on your site now competes independently to be cited. Mike King’s session was the most direct of the conference on this point. His framing: Google has been operating semantically for over a decade, most SEO tooling still does keyword math, and the gap between what tools measure and what AI systems actually evaluate has become the opportunity for brands willing to close it. A well-structured, well-evidenced paragraph on a thinner site gets cited ahead of a buried, unfocused paragraph on a high-authority domain.

The practical consequence for your content team is specific. Paragraphs need to open with their conclusion, not build toward it. Each paragraph should address one provable idea with enough context that it can be extracted and stand alone. Sourcing needs to be explicit and named. An unnamed statistic is an assertion an AI system can’t ground.

The brands we work with who’ve rebuilt their content architecture around these requirements are seeing measurable improvement in AI citation rates within 60–90 days. That improvement shows up in AI Overview appearances and third-party platform citations before it shows up in traffic numbers.

Four Technical Requirements Now Sit Between Your Content and AI Citation

Structured, sourced, crawlable, and machine-readable content gets cited by AI systems. Content missing any one of those properties gets filtered before retrieval. Andrea Volpini, CEO of WordLift, described the staged retrieval process AI systems use: models don’t consume your site whole, they pull from a pre-filtered subset of content that met a minimum bar for structure and verifiability. Content that isn’t structured, connected, and verifiable gets excluded from that subset, regardless of how good it is as writing.

The four technical requirements that determine whether your content clears that bar are: AI crawler access confirmed in your robots.txt; schema markup that is complete, accurate, and present on key pages; an llms.txt file that correctly tells AI agents what they can use from your site; and content blocks written so individual paragraphs can stand alone as cited answers. In every GEO audit we run at Oomph, most brand sites are missing at least two. None of these are advanced configurations. They’re the new baseline for being retrievable. Platforms like Scrunch can show you exactly where you stand across ChatGPT, Perplexity, Gemini, and Google AI Overviews: which prompts surface your brand, which source pages are driving citations, and where competitors are getting cited in your place.

The Brands That Close This Gap First Will Be Significantly Harder to Displace

AI citation compounds the same way that traditional search authority compounds. Brands that establish consistent citation history build a signal advantage that takes competitors real time to close. The difference from traditional SEO is that the window to build that advantage early is shorter, because the field is moving fast and the gap between brands actively building for AI retrieval and brands waiting to see how it develops is widening every month.

The sequencing for closing that gap is straightforward. Technical access for AI crawlers comes first, because you can’t be cited if you can’t be read. Schema markup and structured data come next, because they’re the verification signals AI systems use to trust your content.

Passage-level content architecture follows. Reformatting existing strong content for standalone paragraph retrieval is often faster than creating new content. Third-party brand presence on the platforms AI models train on comes last, because it’s the authority signal that determines whether an AI system treats your content as a credible source or skips it in favor of one it recognizes.

At Oomph, we run GEO audits that score your site across all four of these dimensions and return a prioritized 30-day action plan. If you’re not sure where your brand stands on AI visibility, the audit tells you exactly which gaps are costing you citations right now. Talk to us about a GEO audit.

This week Salesforce signed a deal to acquire Contentful, and most of the coverage is filing it under CMS M&A. I think that misses what’s actually being bought. The interesting part is what an AI agent does with a content platform.

An agent doesn’t take your content and show it to someone. It reads structured, governed source material – your context – and writes the content itself: the answer, the reply, the assembled page, generated on the fly, per customer, per channel. So the most valuable thing you can hand an agent is context it can trust.

That distinction matters more than it sounds. Agents are moving into production quickly, and they will answer your customers and your own staff straight from that material. Point one at clean, governed context and it’s useful. Point it at stale, contradictory, ungoverned content and it will hand back a fluent, confident, wrong answer. Garbage in, confident-sounding garbage out. Either way, it’s speaking for you.

I’ve argued before that content won the web, but context wins the agents, that the CMS was quietly becoming the layer agents reason over. This deal is the biggest CRM on the planet putting real money behind exactly that. Oomph is a gold-certified Contentful partner. We build on the platform daily, and we’ve integrated it with Salesforce and the systems around it. So I’ve been watching this one closely.

So look at what Salesforce already owned. Customer data, and plenty of it – they paid $8 billion for Informatica last year to pull it all together. CRM, marketing, commerce, service. Agentforce on top, ready to act. The one thing they couldn’t pull out of a customer record was the context an agent needs to say something trustworthy: approved product facts, current pricing, the disclosure that has to run in a regulated market, the brand voice, and the right localized version for each audience. Contentful holds all of that. It fits a pattern, too – Informatica, then Momentum, Qualified, Cimulate, and now Contentful. That’s a company building an agent stack on purpose, one acquisition at a time. They decided the context layer was worth buying rather than building.

A customer record can’t tell an agent what it’s allowed to say

Your CRM knows who the customer is. It has no idea what you’re allowed to say to them.

That gap is the whole game. Ask an agent “what’s the current return policy in Germany for this product line” and it will answer either way. Whether the answer is right comes down to three things. Your data tells the agent who is asking. Your context, the structured, governed, current truth behind everything, is what it reasons from. And the content is what the agent generates out of that context in the moment. Data and content are the parts most companies already plan for. Context is the part that gets treated as an afterthought, and it’s the part an agent leans on most.

Jujhar Singh, who runs the applications group at Salesforce, framed the deal around three things working together: the data, the AI-driven content, and the experience. Context is the thread running through all of it. The agent generates the content live, the experience is only as good as what it generates, and both stand on whether the context underneath is trustworthy. That’s what Salesforce bought: the governed context everything else gets built from.

The old acronym survives. The job is changing.

For twenty-five years a CMS has been a Content Management System: somewhere to keep content and push it onto a page for a person to read. An agent uses the same system for a different job. It reasons over that governed context and generates whatever needs to be said from there. Same three letters, new system. The CMS is turning into a Context Management System, and Salesforce just paid to own one.

The headless platforms saw this coming. They had already solved the hard parts an agent needs – structured models, versioning, permissions, and audit trails – and were adding the connective tissue to talk to agents directly. When the biggest player in the category backs that direction with a checkbook, the rest of the market should take the hint.

The old job doesn’t go anywhere, to be clear. Plenty of organizations are running Contentful today purely to manage website content, several of them clients of ours, and that work is the foundation the rest of this gets built on. The structured models, the editorial workflows, the governance, the API-first delivery you set up to run a modern website are the exact things an agent reasons over. You built it to publish pages. You were also building the context layer, whether you planned to or not.

And context reaches well beyond the public website. A lot of it lives in intranets and internal systems: the policies, the procedures, the product knowledge, and the operational detail that actually run the business. That changes who owns the platform. For most of its life a CMS was a marketing purchase, a tool for the brand and the website. Once agents need that governed internal context too, it outgrows marketing, and ownership moves to IT, data, and the enterprise. That shift is already underway, and if you sit in one of those teams, it’s landing on your desk whether you asked for it or not.

The part most of the coverage will skip

The blast radius isn’t the same for everyone. For a retailer, a wrong answer is a bad refund. For a hospital, a bank, a university, or a regulator, it’s a misinformed patient, an exposed internal document, or a benefit you never actually offered. Same technology, very different exposure, which is why governed context is a bigger deal for the organizations we work with most – in healthcare, higher education, government, and associations – than the headlines suggest.

There’s a sovereignty question here too, and it lands hardest on the public sector. Contentful is a German company, and bringing it under a US owner pulls it under US law, including the CLOUD Act. With the EU moving to restrict sensitive public-sector data on US clouds, a government or higher-ed buyer should put that on the question list, not in a footnote.

Where this leaves you

The instinct in most organizations is to make more content. The work that matters now is getting your context in order – structured, current, owned, and governed – because your agents will generate from it whether it’s clean or not. The content platform becomes a system of record: where brand, policy, product knowledge, and compliance live, and where your agents go to find the truth.

If you already run Contentful, the practical news is reassuring. Salesforce has said it keeps operating on the same platform, APIs, and support model, with tighter Agentforce integration as the roadmap rather than a forced migration. Nothing about your current build breaks. The move now is to point the foundation you’ve built at what’s coming next, and that’s a conversation we’re already having with clients.

The organizations that get their context right now are the ones whose agents will be worth trusting later. The big platforms have started spending real money to get there. The window to do the unglamorous work before your agents go live is open right now, and I don’t think it stays open long.

Oomph is a digital experience consultancy serving regulated industries and mission-driven organizations, including healthcare, higher education, government, and associations, where compliance, accessibility, and trust are non-negotiable.

Structured content distribution is the decoupling of content from presentation through a headless CMS and Content as a Service (CaaS) architecture. It is a sound strategy for organizations managing complex content distribution networks across multiple channels. 

To be the most successful, this digital transformation requires organizations to change both their publishing workflows and their content ownership structures. Governance complexity affects 41% of CaaS adopters (PDF), workflow mismatches impact a third, and training requirements average 14 to 18 weeks. 

We have implemented these systems for clients in healthcare, financial services, and higher education, and the pattern is consistent: the three failures that kill structured content initiatives are the preview gap, the ownership vacuum, and the training deficit. Here is what we have learned about each one — and what actually works.

The Promise 

The pitch for structured content distribution is compelling: create content once, store it as modular data in a headless CMS, deliver it via API to any channel (web, mobile, kiosks, AI agents) without reformatting. The CaaS market is projected to reach $2.8 billion by 2035, and over 65% of enterprises have adopted headless CMS architectures.

What they do not tell you is that integration challenges affect 46% of adopters using legacy CMS platforms, and that 31% of enterprises encounter deployment delays exceeding six months. The technology works, but the governance requires just as much attention and is often overlooked. We have seen this avoidable pattern repeat across many structured content implementations.

Why Do Structured Content Migrations Stall?

In short, because organizations implement the technology without redesigning how their teams create, review, approve, and own content. That’s the governance problem.

A headless CMS decouples content from presentation. But most editorial teams have spent years, sometimes decades, working in systems where creating content and seeing how it looks are the same activity. WordPress, Drupal, and even SharePoint have a visual editing experience: build a page, see the page, publish the page.

Structured content does not work this way. Authors fill in fields like title, body, metadata, and related entries to publish content objects, not pages. As one analysis of Contentful’s editorial interface notes, “content editors work in structured content entry forms without seeing how content will render in production.” The front-end determines how those objects appear to users.

That architectural distinction is the correct one for consistent omnichannel delivery. It is also the one most likely to break editorial workflow expectations when teams do not deliberately plan for this big shift. In our experience, three governance failures account for the vast majority of structured content stalls.

What Is the Preview Gap, and Why Does It Derail Teams?

The preview gap is the loss of visual context that editorial teams experience when moving from a WYSIWYG (what you see is what you get) environment to a structured content interface, and it is the most immediate friction point in any headless CMS migration.

Authors who previously built pages visually are now filling in form fields and trusting that a front-end will render them correctly. The shift from “building a page” to “managing a content object” takes adjustment, and “once teams adapt, the structured approach tends to produce more consistent, reusable content.” The problem is what happens before they adapt.

What happens is that authors create workarounds. They paste formatted content into rich text fields, breaking the structured model. They submit tickets to developers asking “what will this look like?” multiple times per week. They maintain shadow documents in Google Docs so they can see their work in context. Every workaround is a governance failure — content that exists outside the system, formatting that undermines the content model, and developer time consumed by preview requests instead of feature development.

The planning that pays off includes building live preview environments for as many content sources as possible. This development work typically gets deprioritized because it is not user-facing, but it determines the success of the new system. As one migration guide puts it, headless platforms deliver excellent editorial experiences “when configured correctly — visual editing, live preview, flexible page-building, role-based permissions. But that configuration is work, it doesn’t happen by default.” Budget for it, build it first, and do not launch editorial access without it.

What Is the Ownership Vacuum?

The ownership vacuum is what happens when structured content crosses departmental boundaries without clear governance over who maintains the content model, who approves changes to shared components, and who is accountable when content is reused in a context the original author never intended.

In a traditional CMS, the marketing team owns the marketing pages, the product team owns product pages, etc. Structured content breaks this model deliberately — a product description created once might appear on the website, in a mobile app, in an email campaign, and through a chatbot simultaneously. But governance complexity affects 41% of CaaS adopters, and multi-team collaboration across 6 to 10 departments increases governance overhead by 27%.

Questions seldom asked include: 

We have seen organizations discover these questions six months post-launch, usually during a content audit that reveals inconsistencies no one can trace. In regulated industries — healthcare, financial services, higher education — those inconsistencies are compliance risks.

Knowing these pitfalls ahead of time can lead to the establishment of a content model governance board before migration begins. A small, cross-functional group (typically 3 to 5 people spanning content strategy, development, and compliance) owns the content model as a shared organizational asset. They approve changes to content types, evaluate reuse implications, and maintain a living inventory of where shared content objects appear. This role does not exist in traditional CMS organizations because it’s not needed. But in structured content environments, it is absolutely necessary.

Why Does the Training Deficit Compound Everything?

Because organizations allocate 90% of their transformation budgets to technology and implementation, and only 10% to change management — the part that determines whether anyone actually uses the system they built.

Training requirements for CaaS implementations average 14 to 18 weeks, the elapsed time from initial exposure to genuine editorial fluency. This training creates the confidence for authors to create, structure, and publish content without reverting to old habits or filing developer tickets. Most implementation budgets account for a one-day training session and a knowledge base article. The gap between that and actual fluency is where adoption dies.

The compounding effect of the training deficit makes this particularly damaging. Undertrained authors hit the preview gap and panic. Without clear governance ownership, there is no one to answer their questions authoritatively. They build workarounds. Those workarounds corrupt the content model. The corrupted content model undermines the case for structured content. Stakeholders lose confidence. The transformation stalls.

BCG’s study of 850+ companies found that only 35% of digital transformations meet their value targets globally. The failure rate is a change management problem that looks like a core problem with the technology itself.

To avoid this failure spiral, structure editorial onboarding as a phased engagement, not a one-and-done event. In our implementations, we start with a pilot group of 3 to 5 authors working with the system while the front-end is still being built. They surface friction points the development addresses in real-time. When the broader editorial team is onboarded, the common pain points have been resolved, and the pilot group serves as advocates who can answer questions and support their peers. This approach adds little cost and dramatically improves adoption velocity.

What Should Organizations Do Before Starting a Structured Content Migration?

Treat governance design as a foundation to build a successful digital transformation:

  1. Audit your editorial workflows as they actually operate. Map who creates content, who reviews it, who approves it, and where informal workarounds exist. As one migration planning guide advises, most publishing workflows “are often based on legacy systems, informal approvals, or staff availability. The result? Delays, missed steps, and content that never quite gets finished.” Your structured content governance must account for the real workflow, not the theoretical one.
  2. Define content model ownership before selecting a platform. Determine who will own the content model as an organizational asset, who can request changes, and what the approval process looks like. This governance structure should be platform-agnostic — it is an organizational decision, not a technical one. We have helped clients build this through our roadmapping and strategy engagements, and it consistently reduces mid-project governance confusion.
  3. Budget for editorial experience parity. If your authors currently have WYSIWYG editing, live preview, and visual page building, do not assume they will accept a simpler and more limiting form-based interface. Calculate the development effort required to provide contextual preview in your new architecture and include it in the implementation scope, not as a phase-two enhancement. Phase two rarely arrives before editorial frustration does.

Wrap Up

The CaaS pitch is not wrong. Structured content distribution is the right architecture for organizations publishing across multiple channels, and it is increasingly the right architecture for AI readiness — structured data is what AI systems consume most effectively. But the promise underestimates the organizational effort to make it successful.

Technology is the easy part. Governance, training, and editorial adoption are harder, and that is where implementations succeed or fail.

We have built these systems on Contentful, Drupal, and composable architectures for organizations in regulated industries where getting content wrong has real consequences. The lesson we keep relearning is the same one: start with the team, not the platform.

Summary

On April 20, 2026, the Department of Justice extended ADA Title II web accessibility compliance deadlines by one to two years for state and local government entities. The extension does not pause underlying accessibility obligations, and it does not extend the separate HHS Section 504 deadline that may apply to hospitals and nonprofits receiving federal funding. Organizations tracking a single calendar are exposed to what we call the two-deadline trap. The right response is to use the extension to systematize accessibility, not to defer it.

This article is not legal advice. Confirm which rules apply to your organization with qualified counsel.

Four days before the original compliance date, the DOJ reset the clock.

Per a summary from Jackson Lewis, state and local governments with populations over 50,000 now have until April 26, 2027 to comply with WCAG 2.1 Level AA under Title II. Smaller entities and special districts have until April 26, 2028.

If you run digital for a public entity, exhale. If the extension made you slow down, recalibrate. The deadlines moved, but the risk did not.

What actually changed, and what did not?

The DOJ pushed back the date that specific technical requirements become enforceable. 

What did not change: the underlying ADA obligation to provide accessible programs and services. Title III public-accommodation risk for hospitals, providers, and nonprofits is unaffected. Demand letters and accessibility-related litigation continued straight through the extension announcement; they did not pause for it.

The compliance date is a deadline, not a start date. Organizations that wait will spend the extension period accumulating debt in templates, content, and vendor contracts, then attempt to remediate it in a sprint. That sprint is where the avoidable risk lives.

Why doesn’t the extension help hospitals and nonprofits?

The DOJ rule covers state and local government entities. It does not cover hospitals and nonprofits whose accessibility obligations come from a different source: federal financial assistance under Section 504 of the Rehabilitation Act.

Jackson Lewis notes that HHS has a separate Section 504 web accessibility compliance date, and as of this writing it has not been extended. Until HHS acts, plan as if it holds.

If your organization receives HHS funding, operates patient portals, runs scheduling or billing flows, accepts donations online, or hosts learning and event platforms, your timeline is likely shorter than the DOJ headline suggests. Title III public-accommodation exposure runs alongside it.

What is the two-deadline trap?

The two-deadline trap is the assumption that a single, well-publicized accessibility deadline is the only one that applies to your organization.

It happens when leadership tracks the DOJ Title II extension and treats it as the program’s primary clock, while a separate Section 504 or Title III obligation governs the actual exposure. The result is a roadmap pegged to the wrong date and a remediation budget that arrives late.

Avoiding it requires confirming, in writing and with counsel, which rules apply, which deadlines govern, and which user-facing services fall inside each scope.

What does day-one compliance actually look like?

Day-one compliance is the day your organization can demonstrate that new content is published accessibly, high-impact user flows work with assistive technology, vendors are managed as part of your posture, and governance is in place.

In our experience working with regulated organizations, the failure mode is rarely the homepage. It is the publishing system that keeps creating new accessibility debt — new pages, new PDFs, new embedded forms, new third-party widgets — faster than remediation can clear it. A defensible program stops the inflow before it works down the backlog.

That means accessibility moves upstream into design system components, CMS templates, content briefs, QA gates, and vendor intake. “Archived content” stops being a folder name and becomes a governance decision with rules. Procurement language changes so the next contract renewal does not lock in another year of vendor risk.

Will an accessibility overlay protect you?

No. Overlays can adjust some visual and interaction settings for some users, but they do not remediate the underlying barriers in your templates, components, content, or third-party tools. The Overlay Fact Sheet, signed by hundreds of accessibility practitioners and organizations, documents the consensus position.

If a widget is your strategy, assume you still need code-level fixes in templates, manual testing with assistive technology, content authoring training, and a third-party tool plan. The widget is not a substitute for any of those, and a number of overlay vendors have themselves been named in accessibility lawsuits.

What should accessibility leaders do this week?

Five actions, in order.

  1. Confirm which rules apply, and which deadline governs. Title II, Section 504, Title III, or more than one. If there is uncertainty, this is a counsel question, not an internal one.
  2. Name a single accessibility owner. Not a committee, but one person responsible for coordinating across IT, content, legal, and procurement. Accountability is the program.
  3. Test your top five user-critical flows manually. Forms, authentication, scheduling, payments, donations, patient portal — whatever blocks access to your primary services. Manual keyboard-only and screen reader software spot checks find what automated scanners miss.
  4. Inventory third-party tools and audit their contracts. Where contracts are silent on accessibility, flag them as priority renewals. Your compliance posture runs through every embedded vendor whether the contract says so or not.
  5. Write a 90-day plan and share it with leadership. Specific, resourced, and tracked beats comprehensive and aspirational every time.

The extension is not a year off. It is a year to put a defensible program in place before the rules apply more explicitly than they already do. Use it wisely.

Bill Gates wrote “Content is King” back in 1996. He was right for about thirty years. On the open web, the winners were the ones who could produce, distribute, and monetize content at scale. That era shaped how we built digital products, how we organized marketing teams, and how we thought about content platforms.

That era is getting a new chapter.

When content becomes context

In the age of agents, content is context. It’s the raw material an AI uses to answer a customer’s question, draft a proposal, summarize a policy, or make a decision on behalf of your business.

If your context is a mess, your agent is a mess. Garbage in, confident-sounding garbage out.

For organizations in healthcare, higher education, and associations (industries where we work every day) that governance layer isn’t a nice-to-have. A health system deploying an agent to answer patient questions needs to know which clinical protocol is current, who approved it, and what the agent is and isn’t allowed to cite. An association managing member benefits can’t afford an agent that surfaces a two-year-old policy document as current guidance. And it’s not just the regulated organizations themselves. The enterprise technology companies that serve these industries, the SaaS platforms, the data providers, the system integrators, face the same challenge: if the content powering their products isn’t structured and governed, the agents built on top of it will inherit every gap. The stakes in regulated industries make the content-as-context problem concrete and urgent, but the same dynamics show up everywhere brand, voice, and accuracy matter: retail pricing, financial disclosures, B2B product specifications, public sector policy. Different risk profiles, same fundamental problem.

This isn’t theoretical. Gartner predicts that 40% of enterprise applications will include task-specific AI agents by the end of 2026, up from less than 5% in 2025. The shift is already moving from prediction to product.

The platforms we work with every day show the movement clearly. The Drupal AI Initiative launched last June and hit $1 million in funding within five months, with the Drupal AI and AI Agents modules reaching production-ready status in October 2025. Acquia built on that foundation with Acquia Source, shipping three AI agents for its Drupal-powered SaaS CMS in December. Contentful open-sourced its MCP server and has been publishing active guidance on agentic content operations. These aren’t experiments. They’re shipping.

Across the category, the pattern is broad. Contentstack launched Agent OS in September 2025 and introduced what it calls the “Context Economy” as its positioning. Kontent.ai shipped what it calls an Agentic CMS the following month. The Model Context Protocol that Anthropic introduced in late 2024 has become the connective tissue, adopted by OpenAI, Google DeepMind, and most of the CMS world.

The platforms are ready. The question is whether your content is.

What agents actually need

An agent doesn’t want a rendered web page. It wants structured, canonical, permissioned, versioned truth. That means:

That’s the same job a mature content platform has been doing for years, just pointed at a new kind of consumer.

We’ve seen this movie before

Every channel shift exposes whether your content was ever really structured to begin with. CD-ROM, then the web, then mobile, now agents. Each one forces organizations to untangle content from presentation. Headless CMS platforms like Drupal, Contentful, Sanity, and Strapi won that argument. Content as structured data, delivered via API, rendered wherever you need it.

Agents are the most demanding channel yet. They don’t just display your content. They consume it, reason over it, and then take action. If your content is trapped inside HTML blobs or buried in PDFs that no one’s touched since 2021, it’s not ready to be context. Structure is the whole game now.

Where context lives today

Right now, company context is scattered across:

Some of these are built for governance. Most aren’t. GitHub is hands-down great for technical content and version control, but marketing and legal teams aren’t opening pull requests to update a pricing page. Notion is excellent for collaboration, weak on structured content models and role-based delivery. Every organization I talk to has some version of this scatter, and it’s about to become a much bigger problem.

The rise of the Context Management System

The old acronym still works. CMS. New job.

Headless CMS platforms have quietly solved about 70% of what agents need. Structured content models. API-first delivery. Editorial workflows. Roles and permissions. Versioning. Audit trails. What they’re adding now is the connective tissue. Acquia is embedding AI agents directly into Drupal-powered workflows through Acquia Source, and Contentful has open-sourced its MCP server to let agents take action on content operations. Across the rest of the category, Sanity launched its Content Agent in January 2026, and Storyblok, Brightspot, and dotCMS have released MCP servers of their own. MCP servers, vector indexing, semantic metadata, agent-optimized delivery endpoints. That’s a much smaller leap than building the whole governance layer from scratch.

The “just throw it all in a vector database” approach has real merit as a retrieval layer. Retrieval is one job. Governance is a different one: who owns canonical truth, who approved the content, when it expires, and who’s allowed to see it. That’s always been the CMS job. It matters more now, not less.

For teams working on Drupal, Contentful, or Acquia Source, this is encouraging. The architectural decisions those platforms made years ago (structured data, granular revisioning, API-first design) turn out to be exactly what AI agents need. Your investment in content architecture is paying off in ways you didn’t plan for. Call it a head start.

What to do about it

If you’re building agentic products, or planning to, the content question is the quiet one that will bite you later. This is the work we’re spending most of our time on with clients right now. A few forward moves:

  1. Audit where your content actually lives and who owns it. You will be surprised.
  2. Pick a source of truth for each category of content. Don’t let five systems claim the same ground.
  3. Get your structured content models right. If your content is trapped inside HTML, it isn’t ready to be context.
  4. Build the governance layer before you need it. Versioning, permissions, approval workflows. Your legal team will thank you. So will your agent.
  5. Connect your CMS to your agents via MCP or equivalent. This is how context flows. Do it early.

Content was king when the battle was for attention. Context is king now that the battle is for correctness. Agents are only as good as the material you feed them, and that material has to be managed with the same rigor we’ve applied to code, to data, and yes, to content itself.

The organizations that treat content governance as infrastructure, not a cleanup project, will be the ones whose agents are trustworthy from day one. That window is shorter than it looks.

As direct website traffic decreases and LLMs slurp up text from multiple sources to mix together and redistribute to users, it has never been more important to maintain high-quality online content. A ROT analysis — which stands for Redundant, Obsolete, Trivial — is a framework through which we can evaluate site content to improve it for usability, SEO, retrieval, and GEO. 

This is a flexible exercise that can apply to a variety of digital properties: web pages, PDFs, intranets, social media pages, call center databases, support knowledgebases… Anywhere that you, as an organization, are speaking to your audience, you have an opportunity to share knowledge, build trust, and solidify your brand image.

Similarly, ROTten content can mislead users, seed doubt, and damage your reputation.

When you use a ROT analysis to kickstart a content clean-up project, you’re ensuring that users and bots alike find only your latest, clearest, most accurate and relevant information. When done properly, it can even set up your team for better content production and management in the future.

How Oomph Approaches Content ROT Analyses

Every ROT analysis looks a little different depending on the industry, content, and what a particular audience needs. 

Make a Plan

Before jumping into dashboards and spreadsheets, we start with a conversation. With any project, we need to understand what problems your organization needs to solve: What’s important to you and your users? Where are you struggling? This is our chance to understand the why behind your content.

As we learn more about what you need, we’ll define what ROT is for your organization. What existing policies do you have in place around archiving old or outdated content? If you don’t have policies, what makes sense for you? What key user journeys should the analysis focus on? We’ll answer these questions and more to make sure we’re going into the analysis with a clear vision of what your content should look like so we can see where it’s missing the mark.

Find the ROT

Let’s get into what ROT looks like specifically and where we look for it.

Redundant means the content communicates information in more than one place. This can result in an inefficient information architecture and messy user paths. There are times duplicate content can be helpful, like when separate task flows require some of the same information. That’s why it’s important to know upfront what journeys are most important to prioritize. In these cases, when the same content shows up in multiple places across a website or app, it’s important to have a method for keeping all content in sync. If it’s possible to edit this content in a single place while distributing it across multiple pages, that can be a great method for maintaining a single source of truth.

Redundant might also refer to several articles written over time that deal with the same topics in similar ways. This can result in the newest content on the topic having its SEO/GEO cannibalized by older content on the same topic. Users might more easily find older content when you want them to find the latest. 

Obsolete content includes outdated information, language, and (probably broken) links. This type of ROT is especially damaging when it’s related to products, services, or something users are trying to take action on. It’s important to keep in mind your entire digital landscape; Maybe you’ve updated the content on your main service page, but did you remember to update automated emails, support articles, and meta descriptions? What pages aren’t built directly into a user flow but can still be found by Google? 

Consider whether it makes sense to archive or unpublish old content, like past news and events. And consider your audience: Is there a reason users would be looking for a historical record, and is that need strong enough to justify keeping it available? If you do choose to keep outdated information published, make sure that it’s clear to users that the content is old and consider providing a link to the latest version.

Trivial content can be harder to define and is highly subjective based on the organization. This might look like “fluff” pieces shared for the sake of SEO or maintaining a publishing schedule, or excessive marketing language that ultimately doesn’t serve you or your users. It might be low-traffic fine print details that apply to a specific audience who typically finds it another way. Maybe it’s content that is related to but outside of your core business function. You’ll need to make some decisions about what is important to you. 

To find ROT, we’ll use a variety of collection and measurement tools. SortSite, Screaming Frog, and Siteimprove can locate broken links, orphaned pages, and other SEO issues. Google Analytics, Hotjar, Contentsquare, and MS Clarity can show common user flows and help identify trivial content. Data from these tools can also prioritize the analysis by surfacing what content is most important to users. If a page gets a lot of traffic, we know that it needs to be clear, up-to-date, and accurate. If a page isn’t visited much, we need to ask whether it should be more highly trafficked, consolidated with higher performing content, or removed.

Deliverables and Next Steps

After all this sorting and evaluating, you might be wondering what you’ll tangibly get out of the process. We know content teams are busy, and going through a review can feel like adding more work to the pile. How can we help prioritize meaningful progress here?

The big outcome is one of my personal favorites: a clean, annotated, actionable spreadsheet. Specifically, we’ll put together an audit of your content with links, page titles, notes on whether the content falls into any of the three ROT categories, and what to do about it: keep, modify, combine, or delete. Depending on the tools your content team uses or what you are willing to subscribe to, we might prepare dashboards and reports directly within an app that your team can use as an ongoing progress tracker. Wherever this list of to-do’s lives, we’ll help you prioritize it so you can start ticking off the most crucial items. Depending on what we decided in early scoping agreements, we can even help work through some high-impact issues, like bulk deleting content, suggesting rewrites, and fixing broken links.

We can also set up an ongoing content hygiene plan. While a dedicated content ROT analysis is a great way to identify and work through issues, an effective content plan should prevent ROT as much as possible and reduce the need for a large effort in the future. This might involve setting up policies, practices, and tools to guide future content management. We’ll help you find ways to see the bigger picture when updating or developing new content to make sure all pieces are accounted for. And when ROT falls through the cracks, you’ll have a plan to regularly review site content, setting ahead of time the when, what, and who.

One Piece in the Puzzle of Strong Content

As we continue to inspect the quality of your website and other digital properties, we can use this ROT analysis as a jumping off point. The initial audit may lead directly into a deeper content audit to evaluate URL paths, heading usage, performance metrics, reading level, and more. As we consider reworking, combining, and cutting entire pages, we may find the need to restructure your information architecture and taxonomy structures, in part or in whole, informed by research exercises like card sorts and tree tests. Depending on what we’ve found in the existing content and how it needs to change, we might suggest changes to your content model, adding, modifying, or removing content types and the relationships between them.

A content ROT analysis is a flexible and fruitful way to take a fresh look at your content ecosystem. If you need help getting started, let us know. We’d love to dig in with you!

Compliance with the California Consumer Privacy Act (CCPA), as amended by the California Privacy Rights Act (CPRA), is a mandatory legal obligation for covered businesses, with significantly increased financial and operational risks starting in 2025.

The Critical Risk: Escalating Fines and Penalties

As of January 1, 2025, the California Privacy Protection Agency (CPPA) increased monetary thresholds and fines to align with the Consumer Price Index.

Key Deadlines and New Requirements (2026–2028)

Regulators have moved from a passive to an active enforcement model, removing the mandatory “grace period” for fixing violations before penalties are applied.

Website CCPA Readiness Checklist

Does This Apply to My Business?

A for-profit business must comply if it does business in California and meets any of the following:

Operational Impact of Non-Compliance

Beyond fines, non-compliance can lead to court-ordered injunctions, mandatory regular audits, and the required deletion of valuable data assets. It also risks significant reputational damage and customer churn, as modern consumers increasingly prioritize data security when choosing where to spend.

Is your website ready for California’s evolving privacy standards? Non-compliance isn’t just a legal risk — it’s a business one that can result in millions in fines, mandatory audits, and lasting reputational damage. Our team helps organizations like yours navigate complex regulatory requirements with confidence, so you can focus on what matters most. Talk to our team today.

To avoid significant financial penalties, which increased on January 1, 2025 to up to $7,988 per intentional violation, your website must function as a compliant interface for consumer privacy rights. Use this checklist to assess your current standing.

1. Mandatory Homepage Links

2. Automated Privacy Signals (Global Privacy Control)

3. Notice at Collection

4. Consumer Rights Intake (DSARs)

5. Technical & Policy Maintenance

Explore WCAG

Is your website one missing link or undetected signal away from a costly CCPA violation? Oomph’s team can walk you through a compliance audit, identify gaps in your current setup, and help you implement the technical and content updates needed to protect your organization. Get in touch with us today to book your CCPA compliance call.

Contentful is no longer just an alternative CMS—it’s become a foundational platform for organizations navigating complexity, regulation, and rapid digital change. In 2026, the question isn’t what is Contentful? It’s why are so many organizations rebuilding their digital ecosystems around it? The answer lies in how digital experiences are built, managed, and scaled today.

Contentful Is Built for Systems, Not Pages

Traditional CMS platforms were designed around pages and templates. That model breaks down when content needs to move faster, live in more places, and remain consistent across teams and channels.

Contentful takes a different approach. It treats content as structured data, not static pages. That means teams create content once and deliver it anywhere—websites, apps, portals, email, or future channels that don’t yet exist.

In 2026, this isn’t a “nice to have.” It’s how modern digital platforms operate.

Composable Architecture Is Now the Default

Composable architecture has moved from trend to standard. Organizations want the freedom to choose best-in-class tools without being locked into monolithic platforms.

Contentful fits cleanly into this model. It integrates with design systems, analytics platforms, personalization tools, consent managers, and AI services through APIs—without forcing teams into rigid workflows.

This flexibility allows organizations to evolve their stack over time instead of rebuilding every few years.

AI Depends on Structured Content

AI-driven experiences are only as good as the content behind them. In 2026, organizations are using AI to support personalization, search, localization, content optimization, and automation.

Contentful’s structured content model makes this possible. Clean, well-defined content enables AI tools to understand, reuse, and adapt content accurately—without introducing risk or inconsistency.

For teams exploring AI responsibly, Contentful provides the infrastructure needed to scale with confidence.

Governance and Compliance Are Built In, Not Bolted On

For regulated and mission-driven organizations, governance isn’t optional. Publishing controls, audit trails, permissions, and review workflows are essential.

Contentful supports these needs at scale. Teams can define roles, control who edits or publishes content, and maintain visibility into changes across environments. This level of governance is critical in industries like healthcare, legal, finance, and the public sector.

In 2026, compliance isn’t something teams add later—it’s designed into the platform from day one.

Marketing and Development Work Better Together

One of Contentful’s biggest advantages is how it aligns marketing and engineering teams. Developers maintain design systems and integrations. Content teams manage content without breaking layouts or workflows.

This separation of concerns reduces friction, speeds up delivery, and minimizes production errors—especially as digital ecosystems grow more complex.

Ready to explore what Contentful could do for your organization? Whether you’re evaluating platforms, planning a migration, or looking to optimize your current setup, Oomph can help you build a content infrastructure designed for the long term. Let’s talk about your next move.

Why Organizations Move to Contentful Now

Organizations typically migrate to Contentful when legacy systems start holding them back. Common triggers include:

In 2026, Contentful isn’t chosen because it’s new. It’s chosen because it’s resilient.

For organizations new to the platform, getting started doesn’t have to mean a complete rebuild. Oomph’s Contentful Kickstart Package helps teams move from decision to deployment with a structured, low-risk approach—giving you the foundation to scale as your needs evolve.

The Takeaway

Contentful has evolved alongside the modern digital landscape. It’s not just a CMS—it’s a content platform designed for scale, governance, and change.

For organizations planning beyond their next website launch and toward long-term digital maturity, Contentful provides the flexibility and confidence needed to move forward.

Ready to explore what Contentful could do for your organization? Whether you’re evaluating platforms, planning a migration, or looking to optimize your current setup, Oomph can help you build a content infrastructure designed for the long term. Let’s talk about your next move.

For many organizations, privacy regulations like GDPR and CCPA seem like distant legal concerns rather than operational priorities. In practice, however, websites serve as the primary point of data collection—making compliance far more relevant than most teams assume. If your site collects user data in any form, privacy compliance isn’t optional.

Understanding When GDPR and CCPA Apply

GDPR governs the collection of personal data from users in the European Union, while CCPA applies to personal data collected from California residents.

Crucially, these regulations are triggered by user location, not company headquarters. A U.S.-based organization serving a global audience may be subject to both frameworks.

Why Websites Are at the Center of Compliance

Most modern websites collect data through multiple channels:

Each of these collection points creates compliance obligations around consent, transparency, and user control.

Moving Beyond Cookie Banners

Meaningful compliance extends well beyond footer disclaimers. Effective privacy management requires:

Legacy CMS platforms frequently lack the flexibility and governance capabilities needed to meet these requirements.

The Role of Your CMS in Privacy Compliance

Your content management system is instrumental in supporting privacy obligations. A modern, composable CMS enables organizations to:

For regulated and mission-driven organizations, CMS limitations can translate directly into compliance vulnerabilities.

The Cost of Non-Compliance

While regulatory penalties are a concern, the greater risk lies in eroding user trust.

Today’s users expect transparency and control over their personal information. Organizations unable to deliver on these expectations risk damaging their reputation with customers, donors, and partners.

Final Thoughts

GDPR and CCPA represent more than legal obligations—they present fundamental digital experience challenges. Websites built on flexible, compliance-ready platforms are better positioned to adapt as privacy expectations continue to evolve.

In today’s environment, privacy compliance shouldn’t be viewed as a constraint. It’s an essential component of delivering a modern, trustworthy digital experience.

Need help ensuring your website meets modern privacy standards? Our team specializes in building compliance-ready digital platforms that protect your users and your organization. Let’s discuss your requirements.

In recent months, Generative Engine Optimization (GEO) has been gaining attention, often positioned as the next evolution beyond traditional Search Engine Optimization (SEO). For some clients, this presents an exciting opportunity to rethink and restructure their digital content. For others, it can feel overwhelming, raising more questions than answers. As AI-powered search tools like ChatGPT, Perplexity, and Gemini change how people discover content online, clients increasingly ask: What is GEO, and how can we prepare our sites for it?

The following handy Q&A guide aims to demystify Generative Engine Optimization (GEO), explain why it matters, and provide practical steps your team can take to get started.

Q: What is GEO and how is it different from SEO?

A: GEO stands for Generative Engine Optimization. While SEO (Search Engine Optimization) focuses on getting your content to rank in traditional search engines like Google (via keywords, backlinks, and site performance), GEO focuses on getting your content mentioned, referenced, summarized, or cited in AI-generated answers from tools like ChatGPT, Gemini, and Perplexity.

Think of SEO as getting your content listed, whereas GEO is about making your brand and its content the answer.

Q: Why should my organization care about GEO?

A: AI platforms are rapidly becoming the first stop for users looking for answers, especially younger audiences and professionals. If an answer appears via Gemini on the top of a Google search, fewer people may scroll further down the page to look for other sources. They got the answer they needed from just one search. If your content isn’t optimized for these tools, you’re missing out on certain traffic data, visibility, and an opportunity to build trust. 

In 2026, ChatGPT alone sees over 4.5 billion visits per month, and Perplexity handles nearly 500 million monthly queries.

Q: How is GEO impacting my site’s analytics? 

A: Likely a lot. Generative engines often summarize content without requiring a click. That means you may see fewer impressions and clicks, even if your content is powering the AI’s answer. Most websites are seeing direct traffic declining across the board. With that said, users who do click through to sites are often engaging more deeply, leading to longer session durations and higher conversion rates. 

Because of this, it’s crucial to learn these new patterns and recognize them within your site’s analytics by setting up new reports. 

Q: How do AI engines choose which content to cite?

A: AI tools evaluate a number of factors, with the most important being:

Each tool has its own algorithm, but clear, factual, structured content with recent updates from trusted sources performs best.

Q: What kind of content works best for GEO?

A: Content that answers questions directly, especially with a conversational tone, tends to work well. Additionally, you want your content to explain not just the what, but also the why and how, since generative engines often expand on user intent. Content structures that perform well for GEO include:

Q: How can we tell if our content is being featured in AI tools?

A: While most AI platforms don’t yet provide native analytics, you can track GEO success through:

Q: Is there a way to make our site more “AI-friendly”?

A: Yes! Here are key GEO best practices:

  1. Use schema markup: Help AI models understand your content’s structure and intent. You can use schema.org to help guide you through improving your site’s markup. 
  2. Write in a Q&A or conversational format: More people are asking full questions or prompts in ChatGPT—rather than just listing keywords. Match your content with how users phrase queries in AI tools. 
  3. Optimize your About page: Make sure that your About page is thoughtfully written to answer who you are, what you do, and why. ChatGPT, for example, pulls from these pages to assess trustworthiness and authority.
  4. Refresh content: Update existing articles with new data and a clear structure (aka headings, bullets, FAQ sections, summaries). Note: You don’t need to create new URLs, just refresh the content to make sure it is relevant and current for today. 
  5. Include citations and data points: Wherever possible, add data and sources. These increase your authority and credibility.

Q: Do we need to optimize differently for each AI tool?

A: The core strategies (trustworthiness, schema, natural language, performant) apply across all platforms, but there are nuances:

Q: Can we block AI tools from using our content?

A: Yes, but be thoughtful about what you are blocking. Adding a file like robots.txt can block AI crawlers, but doing so may reduce your visibility and lead to attribution from AI tools. It could also block legitimate crawlers and thus negatively impact both SEO and GEO, so be thoughtful about how you compose and format that file. 

Note: If your brand has legal or content ownership concerns, we can help you assess what should or shouldn’t be available for AI training or citation.

Q: Do AI Tools honor authenticated access?

A: Yes, but remain mindful. Models like ChatGPT can’t “log in” or bypass authentication. If full research content is only available behind a user login, it won’t be included in training data or scraped summaries. But still pay attention to how content is displayed. If your research is behind a login or subscription paywall, ensure that:

Q: What is llms.txt and should I add it to my site?

A: llms.txt is a proposed convention for websites to provide a lightweight, machine- & human-readable summary (in Markdown) of the “important” parts of the site, to help large language models (LLMs) more easily crawl, interpret, and use content. More sites are starting to add it to their sites to help guide which pages AI should pay attention to. However, it is not yet a universally supported or enforced standard. Many LLMs or AI platforms do not currently yet automatically look for or honor llms.txt. As of now, you can think of it as a nice-to-have, not a requirement.

Q: How often should we update content for GEO?

A: Best practice recommends updating at least once a year for evergreen content. Prioritize updates for:

Even simple updates like reordering information, adding new facts, or improving layout can go a long way with AI engines.

Q: Is GEO just another passing trend?

A: Not at all. GEO is a direct response to how AI is changing digital search and content discovery. Platforms like Google are rethinking their search experience through tools like Gemini, as more people turn to these tools for answers. GEO is how brands stay visible in this new AI landscape.

Q: What’s the first step we should take for GEO Optimization?

A: Start with a content and schema audit of your top-performing pages. From there, apply structured markup, rewrite headlines for clarity, add Q&A sections where applicable, and refresh key posts. A phased approach focused on high-value content will have the biggest immediate impact.

Need help figuring out what content to prioritize for GEO? Our team at Oomph can assess your current visibility and build a roadmap tailored to AI performance.

For more insights into GEO optimization, read…

Generative Engine Optimization (GEO) is making organizations scramble — our clients have been asking “Are we ready for the new ways LLMs crawl, index, and return content to users? Does our site support evolving GEO best practices? What can we do to boost results and citations?”  

Large language models (LLMs) and the services that power AI summaries don’t “think” like humans but they do perform similar actions. They seek content, split it into memorable chunks, and rank the chunks for trust and accuracy. If pages use semantic HTML, include facts and cite sources, and include structured metadata, AI crawlers and retrieval systems will find, store, and reproduce content accurately. That improves your chance of being cited correctly in AI overviews.

While GEO has disrupted the way people use search engines, the fundamentals of SEO and digital accessibility continue to be strong indicators of content performance in LLM search results. Making content understandable, usable, and memorable for humans also has benefits for LLMs and GEO.

How LLM systems (and AI-driven overviews) get their facts

Understanding how LLMs crawl, process, and retrieve web content helps us understand why semantic structure and accessibility best practices have a positive effect. When an AI system generates an answer that cites the web, several distinct back-end steps usually happen: 

  1. Crawling — Bots visit URLs and download page content. Some crawlers execute javascript like a browser (Googlebot) while others prefer raw HTML and limit their rendering.
  2. Chunking — Large documents are split into small, logical “chunks” of paragraphs, sections, or other units. These chunks are the pieces that are later retrieved for an answer. How a page’s content is structured with headings, paragraphs, and lists determines the likely chunk boundaries for storage.
  3. Vectorization — Each chunk is then converted into a numeric vector that captures its semantic meaning. These embeddings live in a vector database and enable systems to find chunks quickly. The quality of the vector depends on the clarity of the chunk’s text.
  4. Indexing — Systems will store additional metadata (URL, title, headings, metadata) to filter and rank results. Structured data like schema metadata is especially valuable. 
  5. Retrieval — A user asks a question or performs a search and the system retrieves the most semantically similar chunks via a vector search. It re-ranks those chunks using metadata and other signals and then composes its answer while citing sources (sometimes). 

The Case for Human-Accessible Content

There are many more reasons why digital accessibility is simply the right thing to do. It turns out that in addition to boosting SEO, accessibility best practices help LLMs crawl, chunk, store, and retrieve content more accurately.

During retrieval, small errors like missing text, ambiguous links, or poor heading order can fail to expose the best chunks. Let’s dive into how this can happen and what common accessibility pitfalls contribute to the confusion.

For Content Teams — Authors, Writers, Editors

Illustration of the problem with poor alt text on images, comparing one poor example and one good example

Lack of descriptive “alt” text

While some LLMs can employ machine-vision techniques to “see” images as a human would, descriptive alt text verifies what they are seeing and the context in which the image is relevant. The same best practices for describing images for people will help LLMs accurately understand the content. 

Illustration of poor heading structure, where the poor example shows skipped heading levels while the good example shows consecutive heading levels

Out-of-order heading structures

Similar to semantic HTML, headings provide a clear outline of a page. Machines (and screen readers!) use heading structure to understand hierarchy and context. When a heading level skips from an <h2> to an <h4>, an LLM may fail to determine the proper relationship between content chunks. During retrieval, the model’s understanding is dictated by the flawed structure, not the content’s intrinsic importance. (Source: research thesis PDF, “Investigating Large Language Models ability to evaluate heading-related accessibility barriers”) 

Illustration of poor link text context, where the poor example shows Click Here and Read more links and the good example shows more descriptive and unique text samples

Descriptive and unique links

All of the accessibility barriers surrounding poor link practices affect how LLMs evaluate their importance. Link text is a short textual signal that is vectorized to make proper retrieval possible. Vague link text like “Click here” or “Learn More” does not provide valuable signals. In fact, the same “Learn More” text multiple times on a page can dilute the signals for the URLs they point to.

Using the same link text for more than one destination URLs creates a knowledge conflict. Like people, an LLM is subject to “anchoring bias,” which means it is likely to overweight the first link it processes and underweight or ignore the second, since they both have the same text signal. 

Example of the duplicate link problem: <a href=“[URL-A]”>Duplicate Link Text</a>, and then later in the same article, <a href=“[URL-B]”>Duplicate Link Text</a>. Conversely, when the same URL is used more than once on a page, the same link text should be repeated exactly.

Illustration of plain language with a poor example and a more positive example. The poor example is dense and wordy while the good example if succinct and uses a list to break the text into chunks.

Logical order and readable content

Simple, direct sentences (one fact per sentence) produce cleaner embeddings for LLM retrieval. Human accessibility best practices of plain language and clear structure are the same practices that improve chunking and indexing for LLMs

For Technical Teams — IT, Developers, Engineers

An illustration of poor semantic structure, where the left shows a potential structure made only of HTML div elements, while the good example shows semantic elements used correctly.

Poorly structured semantic HTML

Semantic elements (<article>, <nav>, <main>, <h1>, etc.) add context and suggest relative ranking weight. They make content boundaries explicit, which helps retrieval systems isolate your content from less important elements like ad slots or lists of related articles. 

Illustration of data in written form as one way to parse information, but contrasted with schema markup which can make it easier for robots to collect correct information about a subject.

Lack of schema

This is technical and under the hood of your human-readable content. Machines love additional context and structured schema data is how facts are declared in code — product names, prices, event dates, authors, etc. Search engines have used schema for rich results and LLMs are no different. Right now, server-rendered schema data will guarantee the widest visibility, as not all crawlers execute client-side Javascript completely. 

How to make accessibility even more actionable

The work of digital accessibility is often pushed to the bottom of the priority list. But once again, there are additional ways to frame this work as high value. While this work is beneficial for SEO, our recent research uncovers that it continues to be impactful in the new and evolving world of GEO.

If you need to frame an argument to those that control the investments of time and money, some talking points are: 

Staying steady in the storm

Let’s be clear — this summer was a “generative AI search freak out.” Content teams have scrambled to get smart about LLM-powered search quickly while search providers rolled out new tools and updates weekly. It’s been a tough ride in a rough sea of constant change.

To counter all that, know that the fundamentals are still strong. If your team has been using accessibility as a measure for content effectiveness and SEO discoverability, don’t stop now. If you haven’t yet started, this is one more reason to apply these principles tomorrow. 

If you continue to have questions within this rapidly evolving landscape, talk to us about your questions around SEO, GEO, content strategy, and accessibility conformance. Ask about our training and documentation available for content teams.

Additional Reading