Your homepage is now an AI landing page, not your citation engine
AI traffic and AI citations are splitting apart. Homepages are becoming the place where AI visits land, while deeper comparison, reference, and evaluation pages win the citations. That changes how smart teams should structure content and measure AI search.
Niels Kaspers
TL;DR
Most teams are still treating AI visibility like one KPI. That is a mistake. AI clicks, AI citations, and AI brand mentions behave differently, and the homepage is increasingly where AI journeys land, not where they are sourced.
Most teams are still measuring AI search like it is one thing.
It is not.
The cleanest signal I found this week is that AI traffic and AI citations are splitting apart.
That sounds subtle. It is actually a pretty big content architecture problem.
Aleyda Solis's May 27, 2026 research on AI traffic versus AI citations showed something a lot of teams are going to miss if they only stare at referral numbers. Across 40 major sites, organic search was still about 108 times larger than AI traffic. But the more interesting split was inside the AI layer itself: AI traffic concentrated on homepages and action pages, while AI citations spread across discovery and evaluation pages.
That lines up with what SEO operators kept repeating on X from June 19 through June 21. The sharp phrasing was different, but the pattern was the same: your homepage increasingly behaves like an AI landing page, while the pages that actually earn citations are often comparisons, docs, explainers, or deep reference content.
I think a lot of teams are still treating those as the same job.
They should not.
The homepage is where the click lands, not where the answer gets built
This is the mistake.
A lot of marketers still assume the page that gets the click is also the page doing the real visibility work.
In AI search, that is often false.
The answer frequently gets assembled from deeper pages that help the model understand a category, compare options, or extract a clean fact pattern. Then, when the user finally clicks, the click lands on the safest and most canonical place: the homepage, the product page, or the action page.
That is why I think the homepage is turning into an AI landing page.
Not the citation engine.
The citation engine is more likely to be:
- the comparison page
- the long-form explainer
- the FAQ cluster
- the integration guide
- the benchmark, tool page, or reference asset
That should sound familiar if you read AEO is the new SEO. I argued there that AI systems reward content that is easy to extract, structure, and cite. What feels newer now is the measurement clarity around where the click lands versus where the answer gets built.
Citations do not automatically create brand recall either
This is the second mistake.
Even when you win the citation, you might still lose the brand moment.
Semrush's June 9, 2026 ghost citations study found that 61.7% of AI citations do not produce a brand mention in the answer itself. The model links to you, but it never really says your name.
That is a brutal metric if your team is celebrating citations as if they are the same thing as awareness.
They are not.
A citation can help credibility, traffic, and distribution. But it does not guarantee memory.
That is why the distinction now matters so much:
- AI traffic measures where visits land
- AI citations measure which pages the model leaned on
- AI brand mentions measure whether the user actually heard your name
Those are three different outcomes.
If you collapse them into one dashboard tile called "AI visibility," you are going to make bad decisions.
Why this changes content strategy
Once you accept that the homepage, the citation source, and the brand mention are different jobs, the site architecture gets much clearer.
Your homepage should do homepage work.
It should convert branded curiosity, sharpen positioning, and route the visitor to the right next step.
Your deeper content should do citation work.
It should answer the specific question cleanly, compare the options honestly, and give the model enough structure to quote or summarize without confusion. This is also where strong internal links matter. If a deep page earns the citation but the click lands later, you still want that page to hand authority and users toward the action surface. That is the same reason I care so much about internal routing in automating SEO with AI and LP conversion optimization.
Your brand layer should do memory work.
That means repeating the positioning clearly enough that the AI can connect the source to the brand instead of only extracting facts from an anonymous page.
The old SEO instinct still matters more than people want to admit
A lot of people want AI search to be a full reset.
It is not.
SE Ranking's ChatGPT citation research still found that authority signals matter a lot: backlinks, domain trust, strong overall visibility, and even homepage traffic still correlate with citation likelihood. It also found that Quora and Reddit mentions keep helping, which fits the broader community-signal story we have been seeing for months.
So no, the homepage is not irrelevant.
It is just doing a different job than a lot of teams think.
The homepage helps you look canonical and trustworthy when the AI decides where to send the click.
The deeper page helps you become the thing the AI actually learns from and cites.
That is a split, not a contradiction.
The reporting model needs to change too
This is where most teams are probably under-measuring the opportunity.
Kevin Indig's June 10, 2026 piece on prompt tracking accuracy made the measurement problem plain: AI visibility behaves more like polling than rank tracking. The citations drift. The prompts vary. One run is noise.
I would add one more layer to that.
Even if your measurement gets cleaner, you still need to separate the outcomes by journey stage.
I would track at least these three buckets separately:
- AI landing traffic to homepages and action pages
- AI citations to discovery, comparison, and reference pages
- AI brand mentions for the topics that actually matter commercially
If you only track referrals, you will miss a lot of influence.
If you only track citations, you will overestimate brand impact.
If you only track mentions, you will miss the pages doing the evidence work.
What I would do if I were fixing this right now
I would make four moves.
1. Split your AI pages by job
Stop expecting one page to do everything.
Treat the homepage like the branded landing surface. Treat deep pages like citation assets. Treat product and action pages like conversion surfaces.
2. Upgrade the pages that answer evaluation questions
If you have pages that compare, explain, benchmark, or walk through tradeoffs, those deserve more attention than generic thought-leadership filler.
They are closer to where AI systems seem to build the answer.
3. Add stronger internal routes from citation pages to action pages
If the citation starts on a deep page, the next move should feel obvious.
That can be a product page, a report, a case study, a demo CTA, or another tightly related asset. The point is to turn answer-surface visibility into a cleaner journey.
4. Stop using one blended AI KPI
Break the reporting apart by clicks, citations, and mentions.
You need to know which pages attract the visit, which pages earn the evidence slot, and whether your brand is actually being named.
My broader take
The AI search conversation still has too much false drama in it.
People want one clean headline: AI is replacing search, AI traffic is exploding, citations are the new rankings, Reddit is the whole game, homepages are dead.
Most of those are too simple to be useful.
The more useful truth is messier.
Organic search is still much bigger.
AI traffic is still small.
But inside the AI layer, the split between landing pages, citation pages, and brand-memory pages is becoming much clearer.
That is the change I would design around.
Your homepage is not dead.
It is just increasingly the place where AI journeys land, not the only place where AI trust gets built.
FAQ
What is the difference between AI traffic and AI citations?
AI traffic measures the pages users visit from AI systems. AI citations measure the pages those systems cite or reference in their answers, even when those pages do not get the click.
Why are homepages getting more AI traffic?
Because AI systems often prefer to send the click to a canonical, branded, lower-risk destination once they have already handled some of the discovery and evaluation inside the answer.
Why are deeper pages still important?
Because those are often the pages the AI uses to understand the topic, compare options, extract facts, and build the answer in the first place.
What is a ghost citation?
A ghost citation is when an AI system links to your page as a source but does not mention your brand name in the answer text.
What should teams track now?
Track AI landing traffic, AI citations, and AI brand mentions separately. They reflect different parts of the journey and they call for different content fixes.