We analysed 14,728 queries across 63 sectors and 389 content types. The headline number is 67.7% composite absorption — but the real story is where that absorption lands. Here's which sectors are uninhabitable, which are defensible, and what to do about it.
In Part 1 of this series, we showed that the overall composite absorption across our dataset is 67.7 — with 41.6 percentage points from non-AI SERP features and 26.1 points from AI Overviews1. The double hit: your organic results are being crowded off the page, and then the AI answers the query before the user ever needs to scroll.
In Part 2, we explored intent absorption — the sub-index where AI doesn't just push you down the page, it removes the need to click entirely. We found that 42.3% of queries have fully satisfied intent, meaning the AI Overview answers the question so completely that the user never needs to visit a website2.
Now, in Part 3, we answer the question every content strategist is actually asking: which industries are getting hammered, and which are surviving?
The answer, drawn from the keyword catalog merged with our dataset (14,710 of 14,728 records matched, a 99.88% hit rate3), is that absorption is wildly uneven. Some industries face composite scores above 84. Others sit comfortably below 52. The difference is structural — and it tells you exactly where to reallocate budget.
Before we get to sectors, let's look at the funnel. When we mapped our dataset against purchase stage — awareness, consideration, decision, retention — the pattern was stark and unforgiving.
| Purchase Stage | Queries | Composite | AI-Only | Visibility | Click | Intent |
|---|---|---|---|---|---|---|
| Awareness | 5,922 | 76.4 | 52.2 | 92.3 | 84.3 | 63.3 |
| Consideration | 4,474 | 72.1 | 46.6 | 84.0 | 74.4 | 58.7 |
| Decision | 2,914 | 52.5 | 24.2 | 59.5 | 46.9 | 53.4 |
| Retention | 1,400 | 48.7 | 16.0 | 56.3 | 42.3 | 51.2 |
Absorption increases up-funnel: awareness-stage queries are most absorbed.
Awareness-stage queries average 76.4 composite. That's critical territory. These are the "what is," "how to," and "best" queries that content teams have been trained to chase for a decade. The logic was sound: capture people early, educate them, move them down the funnel. But that logic assumed the user would actually reach your site to be educated. At 76.4 composite — with visibility absorption at 92.3 and click absorption at 84.3 — they won't.
Consideration-stage queries fare only slightly better at 72.1. This is the comparison content — "best fuel injector cleaner," "lifelock vs aura," "best drawing tablet" — that affiliate marketers and review sites have built businesses around. The AI is now doing the comparing. The incentive to visit four different review sites evaporates.
Decision-stage queries drop to 52.5. Still high, but defensible. The AI can recommend a product, but it can't process a transaction or handle a complex checkout. There's still a journey to complete.
Retention-stage queries sit at 48.7 — the safest harbour in the funnel. These are your "login," "support," "track order" queries. The user already has a relationship with a brand. They're looking for a destination, not information.
The strategic implication is brutal: your top-of-funnel content is being systematically intercepted before users ever reach your site. The question is whether your business model can survive without TOFU traffic, and what you replace it with.
If the funnel tells us where absorption hits hardest, the sector analysis tells us who is taking the pain. We categorised every query into one of 63 sectors. The variance is enormous.
Here are the sectors facing the most severe absorption:
| Sector | Queries | Composite | AI-Only | Visibility | Click | Intent |
|---|---|---|---|---|---|---|
| Cryptocurrency & blockchain | 36 | 84.8 | 64.7 | 96.9 | 91.5 | 70.9 |
| Auto parts & accessories | 50 | 83.7 | 60.6 | 93.8 | 87.3 | 74.6 |
| Cybersecurity & privacy | 38 | 83.7 | 64.2 | 95.0 | 91.1 | 68.5 |
| Gardening & landscaping | 78 | 82.3 | 57.0 | 96.7 | 90.1 | 67.4 |
| Energy & utilities | 46 | 82.1 | 64.9 | 91.5 | 86.0 | 74.4 |
The sectors facing the most severe absorption.
These aren't random. They're information-dense, technical, or comparison-driven — exactly the territory where AI Overviews excel.
Cryptocurrency and blockchain leads the pack at 84.8 composite. Crypto queries are definition-heavy and explanation-heavy — "what is impermanent loss," "how does proof of stake work," "what's the difference between a hot wallet and a cold wallet." An AI Overview answers all of these comprehensively.
Cybersecurity and privacy scores 83.7. Again, definition and comparison territory. "What is a zero-trust architecture," "NordVPN vs ExpressVPN," "how does end-to-end encryption work." The AI gives a confident, sourced summary. The user gets what they need.
Gardening and landscaping at 82.3 reveals a different pattern. This sector is saturated with how-to content — "how to prune roses," "when to plant tomatoes," "how to aerate a lawn." AI Overviews give step-by-step instructions directly in the SERP. The content is perfect for summarisation: procedural, factual, and universally applicable.
But the real danger lies in the big verticals — sectors with hundreds of queries and massive audience overlap:
| Sector | Queries | Composite |
|---|---|---|
| DIY projects | 571 | 78.6 |
| Health & medical | 480 | 73.7 |
| Food & cooking | 547 | 73.6 |
| Finance & money | 540 | 71.9 |
| Consumer tech | 532 | 71.6 |
High-volume verticals in the absorption danger zone.
These five sectors alone represent 2,670 queries — 18.1% of our entire dataset. If you run a site in DIY, health, food, finance, or consumer tech, your keyword universe is sitting in the danger zone. Not because you're bad at SEO — because the SERP has been re-architected to answer the queries your content targets.
Not every sector is bleeding. Some are naturally defensible because their query mix resists AI absorption. The pattern is consistent: service-based, local, brand-heavy, or visual-intent sectors survive best.
| Sector | Queries | Composite | AI-Only | Visibility | Click | Intent |
|---|---|---|---|---|---|---|
| Furniture & decor | 584 | 52.4 | 16.9 | 66.6 | 47.6 | 51.5 |
| Personal services | 502 | 51.5 | 23.9 | 54.7 | 43.2 | 60.2 |
| Travel & hospitality | 576 | 60.7 | 38.4 | 74.1 | 64.9 | 49.4 |
| Toys & baby | 466 | 61.3 | 28.5 | 76.4 | 61.3 | 53.9 |
| Clothing & fashion | 580 | 62.6 | 36.0 | 73.5 | 63.3 | 57.1 |
The sectors with the lowest absorption — natural defensive harbours.
Furniture and decor at 52.4 composite is the safest sector in the dataset. Why? Visual intent. Users searching for "grey velvet sofa" or "scandi dining table" don't want a text summary — they want to see products, compare prices, and browse collections. An AI Overview describing a sofa in prose is no substitute for a product image gallery.
Personal services at 51.5 survives on local and service-booking intent. "Hairdresser near me," "plumber emergency callout," "wedding photographer London." The user wants to find a provider and book. The AI can list options, but the transaction still happens on a local site.
Travel and hospitality at 60.7 is defensible because the booking layer sits beneath the research layer. The AI can summarise what to do in Lisbon, but it can't book the flight or reserve the hotel. The research query gets absorbed; the booking query survives.
Toys and baby at 61.3 is brand-driven and gift-driven. Parents search for specific products by brand — "Lego Star Wars Millennium Falcon," "Huggies size 3 nappies." The AI might summarise a product category, but the purchase decision still requires a product page.
The pattern is clear: sectors where the user needs to do something — book, buy, browse visually, find a local provider — resist absorption. Sectors where the user wants to know something — how, what, why, which is best — get swallowed.
Sector analysis tells you which neighbourhood is dangerous. Semantic cluster analysis tells you which houses are on fire.
We collapsed our queries into 389 semantic clusters. The most and least absorbed:
Most absorbed content types:
| Semantic Cluster | Queries | Composite |
|---|---|---|
| Recipe guide | 10 | 88.5 |
| Maintenance & repair | 48 | 87.3 |
| Product recommendation | 14 | 86.8 |
| Brand comparison | 16 | 86.2 |
| Symptom & condition | 82 | 84.3 |
| Care guide | 386 | 80.7 |
| Technique & tutorial | 384 | 80.0 |
| Comparison | 1,084 | 79.9 |
| How-to guide | 1,036 | 78.2 |
| Product review | 2,440 | 74.7 |
Least absorbed content types:
| Semantic Cluster | Queries | Composite |
|---|---|---|
| App & software | 68 | 41.2 |
| Brand search | 578 | 44.6 |
| Deal & sale | 568 | 46.7 |
| Local search | 1,120 | 47.6 |
| Service booking | 722 | 54.9 |
| Setup guide | 132 | 57.2 |
| Buying guide | 1,514 | 60.3 |
Semantic content types ranked from most to least absorbed.
The pattern is brutally simple. Uninhabitable clusters are those where the AI can give a direct answer: how-to, comparison, symptom/condition, recipe, maintenance/repair. These are the content types SEOs have been producing for years — and the content types AI Overviews are designed to replace.
Resilient clusters are those where the user needs to take action: brand search, deal/sale, local search, service booking, app/software. The AI can mention that a sale exists, but it can't process the checkout.
The surprise is buying guide at 60.3 composite — lower than how-to despite being in the same research phase. Why? Buying guides require visual comparison, pricing, and availability checks that the AI can't fully replace. The AI gives a shortlist; the site closes the deal.
A note on the how-to figures: the semantic cluster "how_to_guide" (78.2) is a subset of the broader "how to" regex pattern (81.8 from Part 2). The gap reflects that some "how to" queries are classified into other clusters like "technique_tutorial" or "setup_guide".
If sectors tell you the neighbourhood and clusters tell you the house, the combination tells you which rooms are burning.
| Category | Semantic Cluster | Queries | Composite |
|---|---|---|---|
| Health & wellness products | Care guide | 62 | 87.6 |
| Home & kitchen | Maintenance & repair | 48 | 87.3 |
| DIY projects | Troubleshooting | 24 | 85.6 |
| Food & groceries | General info | 112 | 85.5 |
| Consumer tech | Comparison | 68 | 85.3 |
| Software & SaaS | General info | 68 | 84.8 |
| Home & kitchen | How-to guide | 50 | 84.7 |
| Health & medical | Symptom & condition | 82 | 84.3 |
| Beauty & personal care | How-to guide | 56 | 84.2 |
| Sports & outdoor | Care guide | 54 | 83.8 |
The most lethal category × semantic cluster combinations.
These aren't abstract numbers. They represent real content strategies being executed right now.
Health and wellness products + care guide: 87.6 composite. This is the "how do I care for [condition/product]" content that health publishers produce by the thousand. The AI answers it directly — a query like "how to care for dry eczema" gets a complete answer in the Overview. The content team wrote 2,000 words. The user read zero.
Home and kitchen + maintenance/repair: 87.3 composite. The "how to fix a leaking tap," "how to descale a kettle" content. AI gives step-by-step instructions. The DIY site that monetises through affiliate links sees the click evaporate.
Food and groceries + general info: 85.5 composite (112 queries). Nutritional information, ingredient substitutes, shelf-life questions. "How many calories in a banana," "what can I use instead of buttermilk." The AI summarises it instantly. The food blog loses the traffic that funded its recipe development.
Health and medical + symptom/condition: 84.3 composite (82 queries). This one is critical. The AI is effectively diagnosing. Users search "why does my knee hurt when I run" and get a medical-style summary. Google's YMYL policies theoretically restrict this, but the data shows AI Overviews are appearing and satisfying intent on symptom queries at high rates. For health publishers, this is an existential threat.
If your content calendar is packed with these combinations, you're not just facing competition. You're facing replacement.
Here's the pattern that should make content strategists rethink their budget.
We tagged each keyword with business value: high, medium, or low. The results invert conventional wisdom.
| Value Tier | Queries | Composite |
|---|---|---|
| Medium | 6,004 | 76.4 |
| High | 7,296 | 64.3 |
| Low | 1,410 | 48.7 |
The value paradox: medium-value awareness content is more absorbed than high-value decision content.
Medium-value keywords are the most absorbed. At 76.4 composite, they sit in critical territory. These are the "educational" and "authority-building" queries that content teams love to target: how-to guides, explainers, comparison pieces, thought leadership. The logic has always been that this content builds trust, earns links, and moves users down the funnel. But if the user never reaches the site — if the AI answers the question before they click — that logic collapses.
High-value keywords are 64.3 composite. Still high, but 12 points safer than medium-value. These are the decision-stage and transactional queries: product pages, pricing lookups, brand searches, checkout flows. The user has intent to convert. The AI can influence the choice, but it can't complete the transaction.
Low-value keywords are 48.7 composite. Mostly navigational and support queries — "login," "track order," "contact us." These were never content priorities, but they're the most defensible keywords in the dataset.
The paradox is this: content teams are over-investing in the funnel stage that's most absorbed. The content that "builds authority" is being systematically intercepted. The content that "drives conversions" is still accessible. The keywords businesses most want to rank for — the high-value, decision-stage terms — are more defensible than the keywords they invest heavily in.
This isn't an argument for abandoning top-of-funnel content entirely. Brand awareness still matters, and citation in AI Overviews still matters. But it is an argument for reallocating budget. If 60% of your content spend goes to awareness-stage how-to guides in a sector with 80+ composite absorption, you're building a library for an audience that will never visit.
The data from all three posts points to the same conclusion: the old keyword research framework is broken because it ignores absorption. Volume × difficulty = opportunity only works on a clean SERP. It isn't.
Five adjustments:
1. Audit your keyword portfolio by sector and content type.
Use the catalog framework: sector, semantic cluster, purchase stage, and value. Identify which of your sectors and content types are in the danger zone. If you're in DIY, health, food, or consumer tech with a content mix heavy on how-to and comparison, you're targeting the most absorbed territory in search.
2. Deprioritise awareness-stage, how-to, and comparison content in high-absorption sectors.
Not because it's worthless — it may still serve brand awareness — but because its direct traffic potential is minimal. At 76.4 composite for awareness-stage queries and 79.9 for comparison queries, the organic click-through environment is hostile. Shift that budget elsewhere.
3. Double down on decision-stage, local, and brand-driven queries.
Decision-stage queries average 52.5 composite. Local search averages 47.6. Brand search averages 44.6. These are the most defensible territories in the SERP. If your business has a local component or strong brand recognition, build content that serves those intents.
4. Shift from "answer questions" to "enable actions."
The AI answers questions. Your site should enable actions. Booking tools, product configurators, visual browsers, community features, availability checkers — content that does, not content that tells. If a user can get the answer from an AI Overview, don't compete on the answer. Compete on what happens next.
5. Build owned channels as insurance.
The absorption trend isn't reversing. Google AI Mode, launched in March 2025, represents the next phase: a full conversational interface that keeps the user inside Google's ecosystem4. Plan for a world with fewer organic clicks. Newsletter, community, direct traffic, and brand search are your insurance policy. The sites that survive won't be the ones with the best SEO. They'll be the ones with the strongest direct relationships.
We've shown you the aggregate data across 63 sectors and 389 content types. But aggregate data only gets you so far. The real question is: what does this look like for your vertical?
The Vertical Absorption Report is a custom run of the Absorption Index on your keyword portfolio, with full catalog enrichment. Here's what's included:
This isn't a generic SEO audit. It's keyword research for the AI era — a methodology, not just a dataset. If you run a SaaS company, we can analyse your 500 target keywords and tell you which 40% are effectively uninhabitable.
Email consulting@wayfinderai.tools to get your Vertical Absorption Report.
The AI Search Absorption Index, built from 14,728 queries across two devices and four intent categories, tells a three-part story.
In Part 1, we mapped the double hit: 41.6% from SERP features, 26.1% from AI Overviews, 67.7% total. The SERP was re-architected twice — first gradually, then suddenly.
In Part 2, we showed intent absorption: 42.3% of queries have fully satisfied intent, and the correlation with estimated CTR is -0.935.
In Part 3, we've shown where that absorption lands. The funnel is being eaten from the top down: awareness at 76.4, consideration at 72.1, decision at 52.5, retention at 48.7. The most absorbed sectors are information-dense and answerable; the safest harbours require action, visual browsing, or local booking. The value paradox reveals that medium-value, awareness-stage content — where teams over-invest — is the most absorbed, while high-value, decision-stage content is the most defensible.
The AI isn't going away. The SERP isn't going back. The question is whether your strategy evolves.
The interactive Absorption Index Browser is live at /resources/tools/absorption-index-browser — explore the full dataset, filter by sector and semantic cluster, and see how your keywords stack up.
If you want us to run the Absorption Index on your vertical, email consulting@wayfinderai.tools.
The data is the map. The action is yours.
Wayfinder AI Search Absorption Index, June 2026. 14,728 queries across desktop and mobile. Composite absorption combines visibility absorption (80.0), click absorption (70.0), and intent absorption (58.8). See Part 1 of this series for full methodology. ↩
Wayfinder AI Search Absorption Index, June 2026. Intent satisfaction assessed via LLM evaluation of AI Overview completeness. 6,224 of 14,728 queries (42.26%) classified as "Yes" — fully satisfied. See Part 2 of this series for full analysis. ↩
Keyword catalog enrichment merged 14,710 of 14,728 records (99.88% match rate), adding sector, semantic cluster, purchase stage, value tier, and persona classifications. 63 sectors, 389 semantic clusters, 4 purchase stages, 3 value tiers. ↩
Search Atlas, "Google AI Mode: How It Works, Features, SEO Impact & Access Guide," May 2026. ↩
Pearson correlation coefficient: -0.9309 across 14,728 queries. As intent absorption increases, estimated CTR decreases in a near-linear relationship. See Part 2 of this series for category-level breakdown. ↩
Explore the full research
