Key findings
16.6
average weekly hours enterprise teams spend improving AI visibility
74%
of enterprise decision-makers rank AI discoverability and attribution as a main or significant priority
60%
of consumers say AI in a brand’s messaging is a turnoff
86%
of consumers always or sometimes check the original source after receiving an AI summary
42%
rank AI answers without clear attribution as the least trustworthy content online, below medical bills and airline fees
A citation in an AI answer means almost nothing if the consumer can’t trace it back to a brand they trust. That’s the math most enterprises are still working around. They treat AI visibility as a marketing problem and brand trust as a separate brand-team problem, but the two have been the same job for a year.
Brands are investing heavily in AI visibility.
The category went from speculation to a standing budget line in just under two years. AI discoverability is now a top-three priority at most enterprises, and the budgets reflect it. What hasn’t caught up is the internal definition of “winning.” Most teams are still spending against a goal nobody has defined.
You can track LLM referral traffic and conversions the way you track attributed conversions. Or you can keep reporting activity and hope someone stops asking what it earned.
AI visibility
of enterprise decision makers say AI discoverability & attribution are a main or significant priority
69%
say content that is not open and structured risks becoming invisible to AI engines
16.6
Average weekly hours enterprise teams spend improving AI visibility
How is AI changing SEO?
AI is changing SEO by adding a second discovery layer. Brands now need to be visible in two places: traditional search results on Google, and AI-generated answers from engines like ChatGPT, Perplexity, Claude, and Gemini. Traditional search still rewards keyword relevance and backlinks.
AI engines look for something different: structured content with clean schema, original data, and editorial credibility the AI can trace back to a source. Most enterprise teams are now running both jobs at once, which is why WordPress VIP’s 2026 survey found teams spend an average of 16.6 hours per week on AI visibility alone.
Performative AI is actively turning consumers off.
The brands spending the most on AI visibility are running into a problem they didn’t account for: The people they’re trying to reach can tell when AI is in the room. Most of them don’t want it there.
A citation in an AI answer reaches a consumer in less than a second. What they think of the brand behind it takes longer. Most of the decision happens after they’ve left the AI summary entirely.
60%
of consumers say AI in a brand’s messaging is a turnoff
86%
of consumers always or sometimes explore the original source after getting an AI summary
Consumer trust
rank AI answers without attribution as the least trustworthy content
Turn this data into an action plan with us in our upcoming webinar.
The two activities are pulling against each other.
The split happens because the work was assigned to different teams two years ago, before anyone knew the work was one job. AI visibility went to whichever team owned SEO, which meant brand trust stayed with whoever owned content quality.
The people closing that gap are rebuilding their website as the place where both jobs get done at the same time. The website is where AI extracts the content cleanly and it’s also where a person goes to decide whether the brand is worth coming back to. The same foundation has to accomplish both.
“Brands cannot afford to treat visibility and trust as separate things. If people can’t trace information back to a brand they trust, being visible is not enough. The website is where a brand provides context and earns trust.”
— Steph Yiu, CEO, WordPress VIP
See it in action: Pew Research Center
When AI assistants started becoming a front door to information, Pew Research Center had two problems to solve at once: making sure LLMs could find their research, and making sure those LLMs got the numbers right.
Working with a WordPress VIP Forward Deployed Engineer, Pew built the structured-content layer that lets ChatGPT and other AI systems cite their work accurately. The editorial integrity Pew is known for stayed intact.
ChatGPT went from invisible to Pew’s #2 referrer in 30 days.
“The money spent on the FDE is already paying off because we are 3-4X ahead of where we should be, and 10X where we would’ve been this time last year with just us developing by code.”
— Seth Rubenstein, Head of Engineering, Pew Research Center
What this means for 2027
Before the next budget cycle wraps up, the dashboards measuring AI citations will start being checked against revenue. Companies that built the connection early will have the data to defend their spending. Everyone else will spend the year re-litigating their AI strategy in budget meetings the way some teams had to re-litigate social spend in 2018.
Pew’s open-source SEO schema is available for other WordPress VIP customers. The approach maps directly to the Content Intelligence and Enterprise CMS Guide — a practical framework for connecting content structure to AI discoverability.
By 2027, AI citations will be tracked the way attributed conversions are tracked today. The teams investing in that measurement layer now are quietly compounding. Everyone else is still trying to define what success looks like.
The infrastructure question (i.e. where that measurement layer lives) is what Chapter 4 addresses. Spoiler: it’s the same platform that governs your editorial.
What this means for SEO teams
The shift toward AI-driven discovery is changing what SEO teams measure and what they stop measuring. Some of the old work still pays out, but not all of it does. The teams getting ahead are the ones rebuilding their approach around what AI engines reward, with the report’s central argument (visibility and trust as one job) built into how they operate.
What’s changing
Citations add a new dimension of visibility alongside rankings. AI engines synthesize one answer and credit whichever sources informed it. The number SEO teams care about now is how often a brand appears in that answer when a user asks AI about the brand’s topic.
Attribution is shrinking at the top and concentrating at the bottom. AI summaries reduce overall click-through from the search surface, and most teams have absorbed that part. What’s less discussed: the clicks coming through are higher-intent. Consumers clicking past an AI answer are usually verifying a specific claim or returning to a brand they recognize.
Schema markup is doing more work than it used to. Structured data was a nice-to-have for rich snippets when Google was the only consumer. Now AI engines lean on schema to figure out what a page is about and whether to cite it. Brands seeing meaningful AI referral growth treat schema as foundational, not a checkbox.
Freshness matters more than it used to. AI engines prefer recently-updated sources, especially when the answer can change over time. SEO teams in “set it and forget it” mode are losing ground to teams maintaining an active refresh schedule.
What to stop doing
Stop reporting impressions on AI answer surfaces as a success metric. An impression without attribution is a leak. Track the citation, the click-through, and what happened after.
Stop optimizing for SERP features AI engines don’t use. Featured snippets and “People also ask” still matter for Google traffic. They just don’t translate to AI engines. Time spent gaming Google-specific surfaces at the expense of structured content is a single-channel bet.
Stop running AI visibility and brand trust as separate workstreams. This is the report’s central point and it applies operationally. The team optimizing for citations and the team writing the content being cited should be one team with one dashboard.
What to measure instead
A workable AI-era SEO dashboard tracks four things:
- Citation frequency across AI engines. How often the brand appears in ChatGPT, Perplexity, Claude, and Gemini answers for the queries that matter. The tooling here is still settling. Most enterprise teams either build their own using LLM APIs or partner with an AI visibility platform.
- AI-driven referral traffic. Distinct from total organic traffic. Most analytics platforms now segment AI referrers, and the brands paying attention watch that segment grow even when overall organic traffic stays flat.
- Post-click behavior from AI referrals. Time on page, pages per session, conversion rate. AI-referred visitors should behave differently from search-referred visitors. If they don’t, something is wrong with the content being cited.
- Attribution accuracy. When AI engines cite the brand, are they citing it correctly? Pew Research’s work with WordPress VIP is the cleanest example of a publisher making the AI get the numbers right.
Who should own this
The teams handling this well have one person or function accountable for both citation tracking and content credibility. In smaller organizations, that role often sits with the head of content or the head of SEO. In enterprises, it’s increasingly a “head of discoverability” role that didn’t exist 18 months ago. The reporting line varies. The same person should be able to answer “are we being cited” and “is the citation leading to a trusted experience” using one dashboard.
Traditional SEO vs. AI SEO
Both disciplines run side by side in most enterprise teams. AI SEO doesn’t replace traditional SEO; it adds a second axis the team has to measure.
| Dimension | Traditional SEO | AI SEO |
|---|---|---|
| Unit of visibility | Page rank on a SERP | Citation in an AI-generated answer |
| Primary surfaces | Google, Bing | ChatGPT, Perplexity, Claude, Gemini |
| What gets rewarded | Keyword relevance, backlinks, page authority | Structured content, schema, citation-worthy data, editorial credibility |
| Measurement | Ranking position, organic traffic, CTR | Citation frequency, AI-referred traffic, attribution accuracy |
| Time horizon for measurable results | Weeks to months | Days to weeks (AI engines update faster) |
| Common pitfall | Keyword stuffing, over-optimization | Publishing AI-generated content with no editorial review |
| Owned by | Marketing / SEO team | Marketing / SEO team, but increasingly requires content + analytics + engineering |
Continue reading
Chapter 1
The internet feels less human.
Chapter 3
Consumers are wary of gatekeeping. More than marketers are.
Chapter 4
The website is still the default trust layer.
Chapter 5
The next website doesn’t look like a website.
FAQs about how AI is changing SEO
How much time do enterprise teams spend on AI visibility?
Enterprise teams spend an average of 16.6 hours per week trying to improve AI visibility, according to WordPress VIP’s 2026 survey of 800 enterprise decision-makers. That’s roughly two working days. Most teams are spending that time without a clear definition of what success looks like, which is why 74% of leaders rank AI discoverability as a main or significant priority but few can name what they’re measuring it against.
Do consumers trust AI-generated brand content?
Most don’t. WordPress VIP’s 2026 survey found that 60% of consumers say AI in a brand’s messaging is a turnoff. When AI does surface a brand in a summary, 86% of consumers say they always or sometimes go check the original source. The clearest signal in the data: 42% of consumers rank AI answers without clear attribution as the least trustworthy content they encounter online, below medical bills and airline fees.
What is the relationship between AI visibility and brand trust?
They have the same job. A citation in an AI answer is close to worthless if the consumer can’t trace it back to a brand they recognize and trust. Most enterprises still treat AI visibility as a marketing problem and brand trust as a separate brand-team problem, and the two have been leaking value into each other for the past year. The teams pulling ahead are building both on the same foundation, with the website doing the work AI engines and human visitors both need.