# Underneath

> Underneath is a generative engine optimization agency. We help businesses become easier for AI-powered search and answer systems to understand, retrieve, describe and reference, including ChatGPT, Google AI Overviews and AI Mode, Gemini, Claude, Perplexity and Microsoft Copilot.

Canonical entity record for Underneath (Underneath Agency). Updated 2026-10-11. HTML: https://underneath.agency/agent · JSON-LD: https://underneath.agency/.well-known/entity.json · Site index: https://underneath.agency/llms.txt

## Identity

- Name: Underneath
- Legal name: Underneath Agency
- Also known as: Underneath Agency, Underneath GEO agency, underneath.agency
- Category: Generative engine optimization (GEO) agency
- Website: https://underneath.agency
- Founded: 2026
- Serves: United States, United Kingdom, Australia, Canada
- Headquarters: 2846 Simons Hollow Road, Bloomsburg, PA 17815, United States
- Contact: hello@underneath.agency. Email only; no phone number is published. Replies within one business day (Monday to Friday, US business hours).
- Organization @id: https://underneath.agency/#organization

## What Underneath does

Underneath finds out why people choose a business, then makes sure AI assistants and AI search give that answer: the pages, structured data, entity facts and third-party sources that AI engines retrieve, trust and cite. It sells generative engine optimization only, as one practice under one plan, and reports mentions, citations, share of voice and leads from AI referrals.

Mission: We help businesses, from single locations to enterprises, get mentioned, cited and recommended by AI assistants and AI search, and chosen because of it, with senior specialists, measurable methods and no shortcuts.

### What Underneath does not sell

- SEO as a standalone service
- paid advertising (PPC, social ads)
- social media management
- web design or development
- branding or creative production

### AI engines tracked in every program

ChatGPT, Claude, Gemini, Google AI Overviews, Google AI Mode, Perplexity, Microsoft Copilot, Bing.

## Services

### [Generative Engine Optimization](https://underneath.agency/services/generative-engine-optimization)

GEO services: get your business understood, cited and recommended in ChatGPT, AI Overviews, AI Mode, Gemini, Perplexity and Copilot answers. (updated 2026-09-27)


## Who it is for

Audiences: Single-location businesses; Franchises and multi-location brands; Growing companies; Enterprises; Ecommerce and Shopify merchants; SaaS companies.

Industries: Home and local services; Franchises and multi-location brands; Healthcare and dental; Legal and professional services; B2B software and technology; Financial services and insurance; Retail and ecommerce; Hospitality and travel.

## Pricing

Programs are priced in three monthly bands, set by scope: $5,000 to $10,000, $11,000 to $20,000, and $20,000 and above. Projects and audits are priced to scope. The Marketing & AI Visibility Audit starts every program and its fee is credited if you continue.

## How an engagement starts

1. Describe the problem through the contact form.
2. A free 30-minute strategy call, then a short written read: what we heard, what we would look at first, whether we are the right fit.
3. One of three ways of working: a monthly program (month one is the Marketing & AI Visibility Audit and the plan built from it), a fixed-scope project, or the audit on its own.

## How the work stays visible: The weekly 30

Every program has a 30-minute review in the same slot every week, with a decision-maker from your side in the room. Four things, always in this order.

1. The work completed, with links
2. The evidence across AI visibility, search and business metrics
3. What the data and research are telling us
4. The next priorities and decisions required

A written note follows within one business day. If a decision-maker cannot join the weekly review, we may not be the right fit.

## Evidence policy

No client results are published until a client has approved them. Scenarios on the site are labeled illustrative. No guaranteed rankings or AI mentions by a date; accounts, data and domains stay in the client’s name.

## Research

- [Do websites serve Markdown to AI agents? 2026 data](https://underneath.agency/research/agent-readable-web-study): 3.2% of top websites serve Markdown when an AI agent asks for it, at a median 96.1% smaller than the HTML; 45.8% of homepages carry JSON-LD. (updated 2026-10-08)
- [Do ChatGPT, Gemini, Perplexity and Claude agree on brands?](https://underneath.agency/research/ai-assistants-brand-agreement-study): 80 buyer questions, four AI assistants: a third of picks shared, the same top pick 10% of the time. They mostly choose differently, not read differently. (updated 2026-10-08)
- [Do AI answers match a business’s Google profile?](https://underneath.agency/research/ai-business-facts-accuracy-study): We asked 4 AI engines for the address, phone, website and hours of 159 local businesses. 18.9% of answers had a fact that differed from the Google profile. (updated 2026-10-08)
- [Do ChatGPT, Gemini, Perplexity and Claude cite pages that rank?](https://underneath.agency/research/ai-citations-google-rankings-study): 8.3% of ChatGPT’s citations rank in Google’s top 10 for the question, Claude’s 25.7%. We trace where the gap opens, stage by stage, over 80 buyer questions. (updated 2026-10-08)
- [Which AI crawlers do top websites block? 10,000 sites, 2026](https://underneath.agency/research/ai-crawler-blocking-study): 15.2% of top sites block GPTBot and 7.4% OpenAI’s search crawler. A same-day check shows ChatGPT cited no page its crawler was barred from. (updated 2026-10-08)
- [The hidden searches AI assistants run before they answer](https://underneath.agency/research/ai-hidden-searches-study): ChatGPT ran 3.7 searches per buyer question before answering. When a search named a source such as NerdWallet or Avvo, the answer cited it 44.0% of the time. (updated 2026-10-08)
- [AI Mode vs AI Overviews: how different are the sources?](https://underneath.agency/research/ai-mode-vs-ai-overviews-study): Same 400 searches: AI Mode and AI Overviews shared 13.9% of cited URLs, and AI Mode cited 16.3% of top-10 pages to the AI Overview’s 29.7%. (updated 2026-10-08)
- [Do AI Overviews cite the pages that rank? 4,051 citations](https://underneath.agency/research/ai-overview-citations-study): Only 28.7% of 4,051 AI Overview citations are page-one Google results. A page-one result’s citation rate falls from 49.5% at position 1 to 15.5% at 9. (updated 2026-10-08)
- [What pages cited by AI Overviews have in common: 3,096 pages](https://underneath.agency/research/ai-overview-cited-pages-study): 3,096 top-10 pages on the same searches: rank predicted AI Overview citation. Compared within each search, only a machine-readable date held up. (updated 2026-10-08)
- [The YouTube videos Google’s AI cites are small](https://underneath.agency/research/ai-overview-youtube-videos-study): Within the same search, the YouTube videos Google’s AI cited came from smaller channels than the ones it showed and skipped. 800 US searches. (updated 2026-10-08)
- [When does Google show an AI Overview? 1,248 US searches](https://underneath.agency/research/ai-overviews-frequency-study): 1,248 US searches: AI Overviews followed the search more than the industry. Local packs cut the rate to 30.7%; rewording a topic lifted it from 59.4% to 93.8%. (updated 2026-10-08)
- [How faithfully do AI assistants quote software prices?](https://underneath.agency/research/ai-pricing-accuracy-study): 61.9% of 840 plan prices four AI assistants quoted for 45 software products were fully faithful to the official page. Most of the rest were real variants. (updated 2026-10-08)
- [Does rewording a question change AI brand recommendations?](https://underneath.agency/research/ai-prompt-phrasing-study): In 1,560 AI answers, a tight budget kept the original’s first brand 15.3% of the time, against 68.0% when the same question was simply asked again. (updated 2026-10-08)
- [Ask an AI the same question 5 times: do the brands change?](https://underneath.agency/research/ai-recommendation-consistency-study): Same 20 buyer questions, five runs each: a quarter of ChatGPT’s brands appeared every time. AI visibility is a distribution, and one answer is a sample. (updated 2026-10-08)
- [Same question, four countries: do AI recommendations change?](https://underneath.agency/research/ai-recommendations-by-country-study): Five runs per country, 40 questions: ChatGPT’s brand lists overlapped 0.594 within one country but 0.429 across the US, UK, Canada and Australia. (updated 2026-10-08)
- [When does a Reddit thread become evidence in Google’s AI?](https://underneath.agency/research/ai-reddit-citations-study): Google’s AI cited Reddit in 17.9% of AI Overviews. Cited threads had twice the comments of skipped ones, and 20.8% of cited claims were not supported. (updated 2026-10-08)
- [How fresh are the pages AI engines cite?](https://underneath.agency/research/ai-source-freshness-study): Held to the same questions, AI assistants cited pages first published half as long ago as Google’s top 10. AI Overviews showed no pull toward recent pages. (updated 2026-10-08)
- [Do Wikipedia and schema make AI assistants recommend a brand?](https://underneath.agency/research/brand-entity-ai-recommendations-study): Brands on Wikipedia are named by more AI assistants, but within the same question most of the gap goes once brand prominence is taken into account. (updated 2026-10-08)
- [Which Google Maps businesses does ChatGPT recommend?](https://underneath.agency/research/chatgpt-local-picks-google-profile-study): Across 898 ChatGPT answers over two days, more reviews than the local median raised a Maps business’s chance of being listed by about 20 points. (updated 2026-10-08)
- [ChatGPT local recommendations: stable details, shifting lists](https://underneath.agency/research/chatgpt-local-recommendations-study): ChatGPT’s local listings match Google Maps field by field, but which businesses it lists changes from run to run, and cited pages often do not name them. (updated 2026-10-08)
- [“Is this brand legit?” How AI assistants build a reputation](https://underneath.agency/research/is-it-legit-ai-reputation-study): Asked “Is this brand legit?”, four AI engines said yes in every complete answer, and 99.7% raised a problem. How they build a reputation from evidence. (updated 2026-10-08)
- [How many websites have an llms.txt file? 2026 adoption data](https://underneath.agency/research/llms-txt-adoption-study): We checked /llms.txt on 5,902 top websites: 11.5% publish a valid file, 2.1% an llms-full.txt, and 15.7% return an HTML page at the address instead. (updated 2026-10-08)
- [How many “best of” lists cited by AI rank their own brand first?](https://underneath.agency/research/self-promoting-best-lists-study): Of 269 AI-cited numbered “best X” lists with an identifiable publisher, 24.2% ranked their publisher first. Such lists were 1.1% of all citations. (updated 2026-10-08)

## Guides

- [How can a CPA or tax firm win better clients through AI search?](https://underneath.agency/resources/accounting-firms-clients-ai-search): By being the firm AI assistants tie to a specific niche and place. Early studies show a few large firms dominate answers, but focused smaller firms break in. (updated 2026-10-10)
- [Are AI assistants now choosing accounting software for small businesses?](https://underneath.agency/resources/accounting-software-ai-search): Partly. AI assistants now shape which accounting tools owners and their accountants consider, so vendors must be named, priced and described correctly there. (updated 2026-10-10)
- [How do AI agent companies get onto buyers’ shortlists in AI search?](https://underneath.agency/resources/ai-agent-companies-customers-from-ai-search): By making their agent easy to verify: clear use cases, public security and governance evidence, honest pricing and independent proof that assistants can find. (updated 2026-10-10)
- [Do people click fewer websites when they use AI answer engines?](https://underneath.agency/resources/ai-answer-engines-fewer-website-clicks): Yes. In a small 2024 lab study, people clicked about one source per question with AI answer engines and about four on Google. Larger studies agree. (updated 2026-10-08)
- [What do lost clicks to AI answers mean for pipeline and revenue?](https://underneath.agency/resources/ai-answers-pipeline-revenue): Fewer search clicks mean fewer prospects entering your funnel, and AI answers appear most on the research and comparison searches where buying starts. (updated 2026-10-10)
- [How often do AI answers say things their sources do not support?](https://underneath.agency/resources/ai-answers-unsupported-claims): Often: audits find roughly one in nine to one in three statements in AI search answers are not backed by the sources cited beside them. (updated 2026-10-08)
- [How often do AI assistant users never visit a website?](https://underneath.agency/resources/ai-assistant-sessions-without-website-visits): About a third of the time: in a 2026 panel, 34.1% of AI assistant sessions had no observed web visit, against 19.5% of the same people’s search sessions. (updated 2026-10-10)
- [Can our server logs show what content AI bots are looking for on our site?](https://underneath.agency/resources/ai-bot-server-logs-content-demand): Partly. Logs show which pages AI bots request, including ones you lack, but not whether any answer used or cited them. One site acted on that signal. (updated 2026-10-08)
- [Can an AI search engine cite my page for something my page does not say?](https://underneath.agency/resources/ai-citing-pages-for-claims-they-do-not-make): Yes. Audits find AI answers attach citations to pages that do not back the claim, from about 3% of Google AI Overview claims to half of one engine’s citations. (updated 2026-10-08)
- [How do AI coding tools win developers through AI search?](https://underneath.agency/resources/ai-coding-tools-developers-ai-search): By earning developer trust where AI answers look: docs, benchmarks, community threads and honest comparisons, then growing one developer into a team of seats. (updated 2026-10-10)
- [Can we trust an AI assistant’s explanation of why it recommended a competitor?](https://underneath.agency/resources/ai-explanations-for-recommendations): Only partly. In a 12-model hotel test, AI assistants’ stated reasons matched their real drivers imperfectly and left out list order and review count. (updated 2026-10-08)
- [How do AI image generators get recommended by AI assistants?](https://underneath.agency/resources/ai-image-generators-growth-from-ai-search): By being clear on what they do best and on commercial rights, then getting that into the independent reviews, galleries and lists AI assistants read. (updated 2026-10-10)
- [How much traffic would AI Mode as Google’s default cost you?](https://underneath.agency/resources/ai-mode-default-traffic-loss): In a 2026 field experiment, making AI Mode Google’s default cut clicks to outside websites by 18.8 points per search. What that means for your traffic. (updated 2026-10-08)
- [If AI Overviews cite my site, does that make up for lost clicks?](https://underneath.agency/resources/ai-overview-citations-vs-lost-clicks): Not on current evidence. Only 1% of visits to AI Overview pages led to a click on a cited source, and Wikipedia lost traffic despite heavy citation. (updated 2026-10-10)
- [Are people more likely to stop browsing after seeing an AI Overview?](https://underneath.agency/resources/ai-overviews-end-browsing-sessions): Yes. In a Pew panel of 900 US adults, browsing ended on 26% of Google pages with an AI Overview, against 16% of pages without one. (updated 2026-10-10)
- [How do AI research tools get found and trusted through AI search?](https://underneath.agency/resources/ai-research-tools-users-ai-search): By being named, with checkable proof of citation accuracy, when students, researchers and librarians ask AI which research assistant to trust. (updated 2026-10-10)
- [Do a few websites win most AI Mode and Perplexity citations?](https://underneath.agency/resources/ai-search-citation-concentration): Yes. A small group of sites takes most AI citations, most of all in Google AI Mode in one Swiss study, least in Perplexity, with a long tail behind them. (updated 2026-10-10)
- [Will AI search reduce our dependence on marketplaces and aggregators?](https://underneath.agency/resources/ai-search-marketplace-dependence): Not yet. Booking sites still supplied most of Gemini’s hotel citations, but their share fell for experience-led questions, where brands’ own pages can compete. (updated 2026-10-08)
- [What should our AI search strategy be for the next five years?](https://underneath.agency/resources/ai-search-strategy-next-five-years): Nobody can forecast five years of AI search. Invest in what has held across engines, measure properly, and keep options open on ads and Google’s AI Mode. (updated 2026-10-07)
- [How can AI software companies stand out in AI search when every rival claims AI?](https://underneath.agency/resources/ai-software-companies-in-ai-search): By earning independent, current proof of what the product does, because AI assistants and buyers both discount the hype that fills the AI software category. (updated 2026-10-08)
- [How can an AI startup win customers through AI search?](https://underneath.agency/resources/ai-startups-customers-from-ai-search): By earning independent coverage and reviews that assistants can find, so the startup reaches shortlists that turn into trials, team use and contracts. (updated 2026-10-08)
- [How do AI video companies win customers through AI search?](https://underneath.agency/resources/ai-video-generator-customers-from-ai-search): By being the tool AI assistants name when creators, marketers and training teams ask what to use, then turning that intent into free plans and upgrades. (updated 2026-10-08)
- [How can AI voice companies turn AI search into revenue?](https://underneath.agency/resources/ai-voice-software-revenue-from-ai-search): By being named, with clear consent and pricing facts, when developers, creators and contact center leaders ask AI which voice tool to try. (updated 2026-10-08)
- [Should our reputation priorities for AI assistants differ from those for human customers?](https://underneath.agency/resources/ai-vs-human-reputation-priorities): Partly. AI assistants share customers’ focus on rating and price, but in testing they overweighted eco-certification and ignored replies to reviews. (updated 2026-10-08)
- [How can an AI writing tool win users when ChatGPT also writes?](https://underneath.agency/resources/ai-writing-tools-users-from-ai-search): By owning a specific writing job general assistants do poorly, and getting that strength into the independent lists and reviews assistants cite. (updated 2026-10-08)
- [How do analytics and BI software companies get shortlisted when buyers ask AI?](https://underneath.agency/resources/analytics-bi-software-ai-search): By being named on the short lists AI answers build for BI comparisons and alternatives, with public proof on governance, accuracy and cost. (updated 2026-10-08)
- [Can we monitor ChatGPT and Gemini visibility through their APIs?](https://underneath.agency/resources/api-ai-visibility-monitoring): Not reliably. In a 2026 test, ChatGPT’s app and its API shared only 12.0% of cited websites for the same question asked a fraction of a second apart. (updated 2026-10-08)
- [How do API companies win developer demand when AI agents choose the integration?](https://underneath.agency/resources/api-companies-ai-search): By being the API that AI assistants and coding agents choose and can integrate on their own, so usage starts in code and grows into enterprise contracts. (updated 2026-10-08)
- [How can an apparel brand win new customers through AI search?](https://underneath.agency/resources/apparel-brands-customers-ai-search): By publishing the fit, size, fabric, care and returns facts AI tools need to match clothes to a shopper, and earning independent mentions that confirm them. (updated 2026-10-08)
- [How do application security vendors win demand when developers and CISOs ask AI?](https://underneath.agency/resources/application-security-demand-ai-search): AppSec vendors earn AI-driven demand by being verifiable to two buyers at once: developers who try tools and security leaders consolidating them. (updated 2026-10-08)
- [Are ads coming to ChatGPT and other AI assistants?](https://underneath.agency/resources/are-ads-coming-to-ai-assistants): They are already arriving. Researchers report OpenAI and Google have put ads into their AI products, and ChatGPT ads reached 31 European markets in August. (updated 2026-10-08)
- [Are AI search engines citing AI-generated content?](https://underneath.agency/resources/are-ai-search-engines-citing-ai-content): Yes. A 2026 audit of four AI search engines flagged about 16% of readable cited pages as AI-generated, from 7.3% for ChatGPT to 27.8% for Copilot. (updated 2026-10-08)
- [How does an asset manager get its funds considered when investors and advisers ask AI?](https://underneath.agency/resources/asset-managers-fund-demand-ai-search): By being the fund family assistants find in trusted fund research, with fees and facts consistent everywhere. Five firms already hold 58% of US fund assets. (updated 2026-10-08)
- [Will AI search point plant leaders to our firm when they plan an automation project?](https://underneath.agency/resources/automation-integrators-industrial-buyers-ai-search): It can, if assistants find proof of your platforms, industries, region and results, so plant leaders put you on the integrator longlist early. (updated 2026-10-08)
- [How can B2B SaaS companies generate revenue from AI search?](https://underneath.agency/resources/b2b-saas-revenue-from-ai-search): By getting onto the shortlists AI assistants build for category, alternatives and comparison questions, then turning that intent into trials and renewals. (updated 2026-10-08)
- [How do baby product brands get recommended when parents ask AI?](https://underneath.agency/resources/baby-product-brands-ai-recommendations): By giving AI assistants accurate, safety-complete product facts and independent proof that parents trust, so the brand is named when registries are built. (updated 2026-10-08)
- [How does a banking technology vendor get onto a bank’s shortlist when the first research happens in AI?](https://underneath.agency/resources/banking-technology-enterprise-deals-ai-search): By being on the first list a bank builds before its RFP. Core, payments and fraud deals are rare and long, so early AI answers carry unusual weight. (updated 2026-10-08)
- [How do beauty retailers turn AI product recommendations into sales?](https://underneath.agency/resources/beauty-retailers-ai-search): By being the store AI names when shoppers ask what to buy and where: accurate prices and stock, clear loyalty value, and content that answers routine questions. (updated 2026-10-08)
- [Which pages should we target to show up in AI recommendations?](https://underneath.agency/resources/best-of-lists-ai-recommendations): Ranked “best of” lists on other sites are the page type AI engines cite most when naming brands; your own site is a small share of citations. (updated 2026-10-11)
- [How does a beverage brand get its drinks recommended by AI assistants?](https://underneath.agency/resources/beverage-brands-ai-recommendations): By being the clear answer to need-state questions: 27% of US consumers already took AI drink advice, and shoppers check ingredients, price and reviews. (updated 2026-10-08)
- [Can AI search help a biotech company find partners, licensees and investors?](https://underneath.agency/resources/biotech-partnering-ai-search): Partly: AI tools now help investors and scientists screen biotech deals, so assets with public, cited science are easier to find before partnering talks. (updated 2026-10-08)
- [Will architects and contractors find our products when they ask AI what to specify?](https://underneath.agency/resources/building-materials-specification-ai-search): Increasingly, yes: product research is moving online and into AI, so specifiers can only name materials whose performance data AI can find and check. (updated 2026-10-08)
- [How does a business advisory firm get found when owners ask AI for help?](https://underneath.agency/resources/business-advisory-firms-clients-ai-search): Owners now ask AI about exits, cash and growth before calling anyone. Advisory firms get named by being specific, credentialed and independently covered. (updated 2026-10-08)
- [Could AI search engines filter out manipulative GEO content?](https://underneath.agency/resources/can-ai-search-filter-manipulative-geo): In lab tests, yes: one defense cut manipulative GEO’s success from 50.32% to 6.20% while keeping most useful sources. No live engine is known to use it yet. (updated 2026-10-08)
- [Can a competitor game AI assistants into recommending their product first?](https://underneath.agency/resources/can-competitors-game-ai-recommendations): In lab tests, yes: hidden text on a product page pushed a product to the top of AI recommendations. Stronger, defended AI engines resist far better. (updated 2026-10-08)
- [Can optimizing for AI search backfire on your brand?](https://underneath.agency/resources/can-geo-backfire-on-your-brand): Yes. In controlled tests, common rewriting tricks often lowered rankings and hype got flagged, but careful optimization did not harm answer quality. (updated 2026-10-08)
- [Can GEO be used to push false information into AI answers?](https://underneath.agency/resources/can-geo-push-false-information-into-ai-answers): Yes. In tests, one planted page made AI systems recommend a fake product in up to 27% of cases, and fabricated evidence was treated as real. (updated 2026-10-08)
- [Can smaller, lower-traffic websites get cited by AI search engines?](https://underneath.agency/resources/can-small-websites-get-cited-by-ai): Yes. AI engines cite a long tail of little-known sites, but being cited is not being recommended, and well-known brands still win most answers. (updated 2026-10-10)
- [Can sellers game AI shopping rankings with manipulative product copy?](https://underneath.agency/resources/can-you-game-ai-shopping-rankings): Sometimes, on weaker AI systems. In tests, GPT-5 and Claude flagged and demoted manipulative product copy, and any gain vanished once rivals copied it. (updated 2026-10-08)
- [What do commercial health sites cited by ChatGPT have in common?](https://underneath.agency/resources/chatgpt-cited-commercial-health-sites): In a 615-source ChatGPT audit, cited commercial health sites mostly showed medical review (71.1%), schema (86.8%) and pages over 1,500 words (68.4%). (updated 2026-10-11)
- [Will optimizing content for ChatGPT hurt our Google rankings?](https://underneath.agency/resources/chatgpt-optimization-google-rankings): Not if you protect pages already earning Google clicks. In one field test, a guarded rewrite lifted ChatGPT referrals with no Google loss beyond the site trend. (updated 2026-10-08)
- [Our ChatGPT referral traffic is growing; can we credit our GEO work?](https://underneath.agency/resources/chatgpt-referral-growth-and-geo): Not from growth alone. In one site’s logs, ChatGPT referrals grew 5.7 times, but untouched pages grew 3.5 times. Credit needs a comparison group. (updated 2026-10-08)
- [Where does ChatGPT fit in the customer journey compared with Google?](https://underneath.agency/resources/chatgpt-vs-google-customer-journey): Later than Google: in a 2026 US and British panel, search usually opened the journey, while people more often browsed before using an AI assistant than after. (updated 2026-10-10)
- [When a formulator asks AI who supplies a chemical, does our company come up?](https://underneath.agency/resources/chemical-suppliers-b2b-buyers-ai-search): Chemical suppliers reach B2B buyers in AI search when grades, specifications, safety data sheets and regulatory status are public and easy to verify. (updated 2026-10-08)
- [Does my brand show up differently in Chinese AI models than in Western ones?](https://underneath.agency/resources/chinese-vs-western-ai-brand-visibility): Likely yes. One study found Chinese AI models named brands far more often than Western ones, though its authors have a stake. Language clearly matters. (updated 2026-10-08)
- [How can a cloud provider win enterprise deals when buyers ask AI first?](https://underneath.agency/resources/cloud-infrastructure-enterprise-demand-ai-search): By being the cloud AI answers name for real architecture, cost and sovereignty questions, with proof architects can check before they commit. (updated 2026-10-08)
- [How do cloud security vendors get onto enterprise shortlists when buyers ask AI first?](https://underneath.agency/resources/cloud-security-enterprise-buyers-ai-search): Cloud security vendors win AI shortlists with proof assistants can check: clear category fit, cloud coverage, marketplace listings and third-party evidence. (updated 2026-10-08)
- [Can AI search bring our CNC shop the RFQs that now go to instant-quote platforms?](https://underneath.agency/resources/cnc-machine-shops-rfqs-ai-search): Yes, if assistants can verify your machines, materials, tolerances, certifications and location; instant-quote platforms are already built for that. (updated 2026-10-08)
- [How do collaboration software companies win teams that ask AI first?](https://underneath.agency/resources/collaboration-software-ai-search): By being named when teams ask AI which tool to try, then turning those free signups into seat growth and company-wide contracts that renew. (updated 2026-10-08)
- [Can we combine ChatGPT, Gemini and Perplexity into one AI visibility score?](https://underneath.agency/resources/combine-ai-engines-visibility-score): Only with care. Score each engine separately first, then combine with stated weights; pooling raw counts lets the most talkative engine dominate. (updated 2026-10-08)
- [Do contractors find our construction software when they ask AI what to use?](https://underneath.agency/resources/construction-software-customers-ai-search): Often they do: contractors use AI at work, and the shortlist for estimating, bidding and project software forms before your demo request arrives. (updated 2026-10-08)
- [Can AI search put our jobsite technology on a contractor’s shortlist?](https://underneath.agency/resources/construction-technology-demand-ai-search): It can, when AI can verify your accuracy, compliance and service record, because contractors now shortlist drones, robots and sensors before calling. (updated 2026-10-08)
- [How does a specialist consulting firm get on the shortlist when clients ask AI for advice?](https://underneath.agency/resources/consulting-firms-clients-ai-search): By being the firm AI can tie to a specific problem, industry and place, and confirm in outside sources, because clients now research consultants before calling. (updated 2026-10-08)
- [Can AI assistants win new subscribers for a consumer app?](https://underneath.agency/resources/consumer-app-subscribers-from-ai-search): Yes. One in ten US adults first heard of their latest app from an AI assistant, but the app store listing and the first hour of use still close the sale. (updated 2026-10-08)
- [How do consumer electronics brands and retailers win shoppers who compare specs in AI?](https://underneath.agency/resources/consumer-electronics-sales-from-ai-search): By being the model AI names when shoppers compare specs and prices: AI answers lean on independent lab reviews, retailer data and accurate prices. (updated 2026-10-08)
- [How do consumer fintech apps win customers when people ask AI about money?](https://underneath.agency/resources/consumer-fintech-apps-customers-ai-search): By being named, and described accurately, when people ask AI which app to use and whether it is safe. In money questions, trust facts decide the answer. (updated 2026-10-08)
- [Will AI assistants put our devices on the shortlist when buyers choose a brand and ecosystem?](https://underneath.agency/resources/consumer-tech-brands-ai-search): By being named, with correct compatibility and launch facts, when households ask AI which device or ecosystem to buy into, a choice that lasts years. (updated 2026-10-08)
- [Is my content business at risk from Google’s AI summaries?](https://underneath.agency/resources/content-business-risk-from-ai-search): It depends on your content. Factual pages lose visits to Google’s AI summaries; opinion and experience content gained, until AI Mode eroded the gain. (updated 2026-10-10)
- [Will AI search put our plant on the supplier shortlist before the RFQ goes out?](https://underneath.agency/resources/contract-manufacturers-rfqs-from-ai-search): It can, when your capabilities, certifications and location are written where AI can verify them, because engineers now use AI to build supplier longlists. (updated 2026-10-08)
- [How do makeup brands get recommended when shoppers ask AI?](https://underneath.agency/resources/cosmetics-brands-product-discovery-ai-search): By giving AI assistants shade, undertone and finish facts they can match to a shopper’s question, backed by reviews and independent coverage. (updated 2026-10-08)
- [Why do AI assistants keep naming the same CRMs, and how do we join them?](https://underneath.agency/resources/crm-software-ai-search): Because the leaders dominate the sources AI search reads. Challenger CRMs get named by owning specific needs: an industry, a team size, a budget. (updated 2026-10-08)
- [How do customer support software companies get chosen when buyers ask AI?](https://underneath.agency/resources/customer-support-software-ai-search): By being named while support leaders re-shop for AI agents, with clear outcome pricing and proof of resolution rates that buyers and AI answers can check. (updated 2026-10-08)
- [How can cybersecurity software companies generate revenue from AI search?](https://underneath.agency/resources/cybersecurity-software-revenue-from-ai-search): By getting named in the AI answers that now build security shortlists, backed by third-party proof that CISOs and AI assistants can both check. (updated 2026-10-08)
- [How do data platforms get shortlisted when data leaders ask AI?](https://underneath.agency/resources/data-infrastructure-enterprise-demand-ai-search): By being named, accurately, in the AI answers data leaders use to frame architecture choices, then winning the proof of concept on the buyer’s own data. (updated 2026-10-08)
- [How do data science platforms win enterprise buyers who research with AI?](https://underneath.agency/resources/data-science-platforms-ai-search): By being named and described accurately in the AI answers that frame platform comparisons, governance checks and cost estimates, then winning the pilot. (updated 2026-10-08)
- [How does a database company get chosen when AI helps pick the stack?](https://underneath.agency/resources/database-companies-developers-ai-search): By being the database AI coding tools and assistants reach for by default, then backing it with docs, benchmarks and proof enterprise architects can check. (updated 2026-10-08)
- [Can a site with deep content but no technical SEO get cited by Gemini?](https://underneath.agency/resources/deep-content-without-technical-seo-gemini): Sometimes. In a 14-hotel Gemini audit, an independent hotel with no schema but a 33-question FAQ was cited directly; a polished rival was not. (updated 2026-10-11)
- [Can AI search put your dental technology on a dentist’s shortlist?](https://underneath.agency/resources/dental-technology-ai-search): Dentists already use AI at work. To make their shortlist, a dental technology company needs clear, checkable facts on cost, clearance and fit. (updated 2026-10-08)
- [How do developer tool companies win users when developers ask AI first?](https://underneath.agency/resources/developer-tools-ai-search): By being the tool AI coding agents and assistants pick for a developer’s stack, then turning that free signup into team plans and enterprise contracts. (updated 2026-10-08)
- [How do DevOps companies get recommended when engineers ask AI which tool to use?](https://underneath.agency/resources/devops-platforms-ai-search): By being the tool AI answers and coding agents find in docs, GitHub and engineer communities, then converting team adoption into platform deals. (updated 2026-10-08)
- [How do we know whether a GEO campaign actually improved our AI citations?](https://underneath.agency/resources/did-geo-improve-ai-citations): Only with repeated measurement before and after, a margin of error, and an untreated comparison. A 3-point gain in AI citation share can be pure noise. (updated 2026-10-08)
- [Our AI visibility went up: was it because of our GEO work?](https://underneath.agency/resources/did-geo-work-raise-ai-visibility): Maybe. A rise can also come from chance, AI platform growth, engine updates, competitors or a changed prompt set, so you need a baseline and a control. (updated 2026-10-08)
- [How does a digital bank win depositors when savers ask AI where to put their money?](https://underneath.agency/resources/digital-banks-depositors-ai-search): By making rates, conditions and deposit insurance facts current and checkable, because savers now ask AI where to put money and AI often gets rates wrong. (updated 2026-10-08)
- [How do digital health apps win new users when people ask AI about their health first?](https://underneath.agency/resources/digital-health-apps-users-ai-search): By giving assistants proof they can check: plain privacy terms, honest evidence and app store trust, now that 32% of US adults ask AI about health. (updated 2026-10-08)
- [Do AI agents make up facts about my business, or just leave them out?](https://underneath.agency/resources/do-ai-agents-invent-or-omit-business-facts): Mostly they leave facts out. Answering from the web instead of a business’s site, agents left out 45% of facts, up from 29%; wrong facts rose only 4% to 6%. (updated 2026-10-08)
- [Do AI agents recommend businesses whose websites they can read more often?](https://underneath.agency/resources/do-ai-agents-recommend-readable-websites): In one large vendor study, AI agents clearly recommended businesses with agent-readable websites 20% of the time, against 11% for sites they struggled to read. (updated 2026-10-11)
- [Do AI assistants still answer from training data or from live web pages?](https://underneath.agency/resources/do-ai-assistants-answer-from-training-data): Mostly from live pages. In a 2026 test of four AI agents, memory supplied 7–10% of answers, though some AI answers still cite no source at all. (updated 2026-10-10)
- [Do AI assistants favor big brands over smaller competitors?](https://underneath.agency/resources/do-ai-assistants-favor-big-brands): Mostly yes when products look alike: AI assistants default to market leaders, but in tests a small, clearly stated advantage beat a famous name. (updated 2026-10-08)
- [Do pages need author bylines to be cited by ChatGPT?](https://underneath.agency/resources/do-ai-cited-pages-need-author-bylines): No. In a ChatGPT health audit, 64.7% of cited pages named no author and only 7.0% gave a name with credentials; publisher authority mattered more. (updated 2026-10-11)
- [Do AI engines cite local-language sources in other languages?](https://underneath.agency/resources/do-ai-engines-cite-local-language-sources): Mostly yes: ChatGPT and Perplexity switch to local-language sources, while Claude kept citing English sites in one study. Translation alone is not enough. (updated 2026-10-10)
- [Do ChatGPT and other AI engines cite my own website or third-party reviews?](https://underneath.agency/resources/do-ai-engines-cite-your-own-website): Mostly third-party sources. Brand-owned sites are a small share of AI citations in most studies, though the share rises for buying-ready questions. (updated 2026-10-10)
- [Do Gemini, GPT and Claude prefer the same kind of content?](https://underneath.agency/resources/do-ai-engines-prefer-same-content): Mostly yes on content quality: Gemini, GPT and Claude share most preferences. But they cite different sources, so one good page will not show up everywhere. (updated 2026-10-11)
- [Do Google AI Overviews downplay negative content?](https://underneath.agency/resources/do-ai-overviews-downplay-negative-content): In one audit of 11,000 Google questions, AI Overviews drew less from negatively toned sources. Brand reviews were not tested, and criticism still appears. (updated 2026-10-08)
- [Do Google AI Overviews reduce clicks to my website?](https://underneath.agency/resources/do-ai-overviews-reduce-clicks): Yes. In a 2026 field experiment, hiding Google AI Overviews raised clicks to outside sites by 8.8 points; browsing data and Wikipedia traffic agree. (updated 2026-10-10)
- [Do AI search engines just agree with however a question is phrased?](https://underneath.agency/resources/do-ai-search-engines-agree-with-leading-questions): Often, on loaded questions. In a 2024 audit, AI answer engines gave one-sided answers that agreed with a charged question 50% to 80% of the time. (updated 2026-10-08)
- [Do AI search engines cite the same websites as Google?](https://underneath.agency/resources/do-ai-search-engines-cite-the-same-sites-as-google): Mostly not. Studies find AI engines and Google share 18% to 38% of sources, and only 8.3% of ChatGPT’s citations ranked in Google’s top 10 in our test. (updated 2026-10-10)
- [Do ChatGPT, Copilot, Google and Perplexity cite the same sources?](https://underneath.agency/resources/do-ai-search-engines-cite-the-same-sources): No. Studies find ChatGPT, Copilot, Google and Perplexity cite largely different pages for the same question, so success on one engine rarely carries over. (updated 2026-10-10)
- [Do backlinks still matter for visibility in Perplexity?](https://underneath.agency/resources/do-backlinks-matter-for-perplexity-visibility): Probably, but the evidence is thin. In a test of 112 startups, linking websites were the strongest predictor of Perplexity visibility; no study proves cause. (updated 2026-10-10)
- [Does being cited by ChatGPT boost our Google rankings?](https://underneath.agency/resources/do-chatgpt-citations-boost-google-rankings): No evidence says so. The one field test found ChatGPT gains left Google clicks flat to falling, and ChatGPT mostly cites pages Google does not rank. (updated 2026-10-08)
- [Do citations make people trust AI answers more, even when the links are wrong?](https://underneath.agency/resources/do-citations-make-ai-answers-more-trusted): Yes. In a US experiment, reference links raised trust in AI answers, and broken or irrelevant links raised it just as much as valid ones. (updated 2026-10-08)
- [Do comparison pages help B2B brands get cited by AI?](https://underneath.agency/resources/do-comparison-pages-help-b2b-ai-citations): Partly. Comparison content gets used more once a page is cited, but brands’ own sites drew only 2.9% of AI citations in one large vendor dataset. (updated 2026-10-08)
- [Do FAQ pages help you get cited by AI search engines?](https://underneath.agency/resources/do-faq-pages-help-ai-citations): Not by format alone. In the research so far, FAQ formatting did not make AI engines use a page more; the evidence inside the page did. (updated 2026-10-08)
- [Do GEO content tactics like adding statistics, quotes or citations actually get you cited more by AI?](https://underneath.agency/resources/do-geo-content-tactics-work): Mostly not. A 2025 benchmark found most AI rewriting tactics did nothing or lowered a page’s rank; adding statistics hurt in 19 of 24 settings. (updated 2026-10-08)
- [Do people actually check the sources AI assistants cite?](https://underneath.agency/resources/do-people-check-ai-sources): Rarely. In a Pew panel of 900 US adults, only 1% of visits to Google pages with an AI Overview led to a click on a cited source. (updated 2026-10-08)
- [Do people trust AI search answers less than regular Google results?](https://underneath.agency/resources/do-people-trust-ai-search-less): Slightly. In US experiments, the same text was trusted a little less when labeled AI, and a week of Google’s AI Mode lowered trust in Google. (updated 2026-10-08)
- [Does advertising spend help our brand get recommended by AI, or does something else matter more?](https://underneath.agency/resources/does-ad-spend-help-ai-brand-recommendations): Not directly, on current evidence: search interest and online conversation tracked AI brand prominence far more closely than advertising spend did. (updated 2026-10-08)
- [Does AI search actually use the pages it cites?](https://underneath.agency/resources/does-ai-search-use-the-pages-it-cites): Not evenly. AI search engines lean heavily on some cited pages and barely use others, so a citation count overstates how much a page shaped the answer. (updated 2026-10-07)
- [Does showing up in AI answers actually drive business results?](https://underneath.agency/resources/does-ai-visibility-drive-business-results): Possibly, but not yet proven. A 2026 review of 45 studies found that claims about GEO’s return outstrip the evidence; none tied AI visibility to sales. (updated 2026-10-08)
- [Does blocking Google-Extended in robots.txt reduce our visibility in AI Overviews?](https://underneath.agency/resources/does-blocking-google-extended-hurt-ai-overviews): Possibly. A large 2025 study linked blocking Google-Extended to fewer AI Overview citations; our 2026 check saw no clear effect. Weigh it as a real risk. (updated 2026-10-10)
- [Can restructuring existing content, without changing what it says, increase AI citations?](https://underneath.agency/resources/does-content-structure-increase-ai-citations): Possibly, but the evidence is split. One lab test found a 17.3% lift from restructuring alone; a 252,000-trial test found formatting had no effect. (updated 2026-10-08)
- [Is it true that GEO increases AI visibility by 40%?](https://underneath.agency/resources/does-geo-increase-ai-visibility-by-40-percent): Not as usually stated. The 40% comes from a 2023 lab test in which a page already given to the AI won more of the answer’s text, not more traffic. (updated 2026-10-08)
- [Does keyword stuffing still work in AI search?](https://underneath.agency/resources/does-keyword-stuffing-work-in-ai-search): No. In published tests, repeating query keywords did nothing or lowered a page’s share of AI answers; relevance and search ranking mattered more. (updated 2026-10-08)
- [Does list order change what an AI assistant recommends?](https://underneath.agency/resources/does-list-order-change-ai-recommendations): Yes, in controlled tests: options shown first are picked more often. One hotel study valued first place at $11.7 a night; the size varies widely by AI model. (updated 2026-10-10)
- [Does the way customers phrase a question change which sources AI search cites?](https://underneath.agency/resources/does-question-phrasing-change-ai-sources): Yes. In a Gemini hotel test, experience-style questions drew 55.9% of citations from non-booking sites, against 30.8% for booking-style ones. (updated 2026-10-10)
- [If we rank on Google, will we show up in AI Overviews?](https://underneath.agency/resources/does-ranking-on-google-get-you-into-ai-overviews): Not reliably. In our test only 28.7% of AI Overview citations were page-one results, though the top result was cited in 49.5% of AI Overviews. (updated 2026-10-10)
- [Does readable writing help your content appear in AI answers?](https://underneath.agency/resources/does-readable-writing-help-ai-visibility): Somewhat. AI engines cite easier-to-read pages, and fluent rewrites helped in early lab tests, but later tests found polishing text alone rarely helps. (updated 2026-10-08)
- [Does Reddit and forum content actually shape Google AI Overview answers?](https://underneath.agency/resources/does-reddit-shape-google-ai-overviews): Less than citations suggest. Google’s AI Overviews cite Reddit, but one study found they drew 22.1 points less content from forums than from other sources. (updated 2026-10-10)
- [Does social media content help your brand show up in AI search?](https://underneath.agency/resources/does-social-media-help-ai-search-visibility): Mostly on Google. AI Overviews often cite Facebook, Instagram and YouTube; ChatGPT and Claude rarely cite social platforms at all. (updated 2026-10-07)
- [Can AI recommendations lower what a DTC brand pays for each new customer?](https://underneath.agency/resources/dtc-brands-ai-search-sales): Possibly. AI-referred shoppers bring a high share of new buyers and convert well, but the channel is still small next to paid social and search. (updated 2026-10-08)
- [How does an ecommerce brand get its products chosen by AI shopping assistants?](https://underneath.agency/resources/ecommerce-brands-ai-shopping-assistants): By making every listing, retailer page and review source agree, so ChatGPT, Google, Rufus and Perplexity can find, compare and trust your products. (updated 2026-10-08)
- [How do ecommerce platforms and apps win merchants who ask AI?](https://underneath.agency/resources/ecommerce-platform-ai-search): By being named when merchants and their agencies ask AI which platform or app to use, and by proving you can put merchants’ products in AI shopping. (updated 2026-10-08)
- [Can AI search bring an EdTech company more schools, teachers and families?](https://underneath.agency/resources/edtech-ai-search): Partly. Teachers, students and families already ask AI; districts still buy on evidence, privacy and state lists, so AI answers must find that proof. (updated 2026-10-08)
- [How do endpoint security vendors win enterprise deals when CISOs ask AI which EDR to buy?](https://underneath.agency/resources/endpoint-security-enterprise-customers-ai-search): Endpoint security vendors earn AI shortlist places through public test results, incident transparency and consistent facts, not self-ranked lists. (updated 2026-10-08)
- [Will clients find our engineering firm when they ask AI who can design their project?](https://underneath.agency/resources/engineering-firms-project-inquiries-ai-search): Engineering firms get named by AI when their project types, licenses, people and locations are stated clearly and confirmed by sources clients already trust. (updated 2026-10-08)
- [When engineers ask AI which CAD, simulation or PLM tool to use, will they hear our name?](https://underneath.agency/resources/engineering-software-ai-search): Engineering software gets recommended by AI when its workflows, file formats, plans and limits are stated plainly and confirmed by sources engineers trust. (updated 2026-10-08)
- [Is an English-only AI visibility check enough if you sell abroad?](https://underneath.agency/resources/english-only-ai-visibility-audits): No. In a 12-language European test, AI named local brands far more often when asked in their home language, so English-only checks undercount them. (updated 2026-10-08)
- [Are AI assistants now shaping which enterprise software gets shortlisted?](https://underneath.agency/resources/enterprise-software-shortlists-ai-search): Increasingly, yes. Enterprise buyers research vendors with AI assistants, often private ones at work, before any seller hears about the deal. (updated 2026-10-08)
- [How do ERP vendors get on the shortlist when buyers ask AI?](https://underneath.agency/resources/erp-software-ai-search): By being on the long list AI assistants draw from analyst, consultant and comparison sources, with clear industry fit, partners and implementation facts. (updated 2026-10-08)
- [Will AI search change how boards choose an executive search firm?](https://underneath.agency/resources/executive-search-firms-clients-ai-search): Referrals still decide most mandates, but directors now use AI to check firms and partners. Public, verifiable track records decide what those checks find. (updated 2026-10-08)
- [Can fake reviews on the web make AI assistants recommend a fake brand over ours?](https://underneath.agency/resources/fake-reviews-fake-brands-ai-recommendations): Yes. In tests of 12 AI models, planted review pages got a non-existent brand recommended in up to 73.8% of products, and no defense tested fixed it. (updated 2026-10-08)
- [How does a fashion brand get recommended when shoppers ask AI what to wear?](https://underneath.agency/resources/fashion-ecommerce-ai-recommendations): By describing products in shoppers’ style and occasion language, with clear images and consistent details everywhere AI shopping tools and visual search look. (updated 2026-10-08)
- [How can a financial consulting firm win CFO, deal and restructuring work through AI search?](https://underneath.agency/resources/financial-consulting-firms-clients-ai-search): By being the firm AI names for a specific finance problem, company size and deal stage, with a reputation, credentials and current commentary that hold up. (updated 2026-10-08)
- [Are AI assistants changing how banks and finance teams choose fintech software?](https://underneath.agency/resources/fintech-software-customers-from-ai-search): Yes. Finance teams and bank staff research vendors with AI assistants, and in finance those answers lean on trusted, regulated and well-ranked sources. (updated 2026-10-08)
- [How can a fintech startup get recommended by AI when established brands own the answers?](https://underneath.agency/resources/fintech-startups-customers-ai-search): By owning the category explainer, earning independent coverage early and stating fees plainly. AI favors known brands until it finds a clear reason not to. (updated 2026-10-08)
- [How do you fix wrong information about your brand in AI answers?](https://underneath.agency/resources/fix-wrong-brand-information-in-ai-answers): Trace the error to the page it came from and fix that page. AI answers are built mostly from live web sources, often your own old pages. (updated 2026-10-08)
- [How can food ecommerce brands turn AI meal planning into orders?](https://underneath.agency/resources/food-ecommerce-sales-from-ai-search): By being the products AI names when shoppers plan meals and build lists: complete food data, compliant claims, reviews and presence where carts are filled. (updated 2026-10-08)
- [How can a footwear brand get its shoes found when shoppers ask AI what to buy?](https://underneath.agency/resources/footwear-brands-ai-search): By making each shoe model easy for AI to match to a fit, use and price question, so it lands on the shortlist before shoppers pick a size and a store. (updated 2026-10-08)
- [How do freight carriers, brokers and forwarders win shippers who research with AI?](https://underneath.agency/resources/freight-companies-shippers-ai-search): Shippers check freight providers through rankings, reviews and quotes; AI answers now sit on that path, so carriers and brokers must be easy to verify by lane. (updated 2026-10-08)
- [How can a furniture brand turn AI shopping answers into more high-ticket orders?](https://underneath.agency/resources/furniture-brands-sales-from-ai-search): By being named when shoppers ask AI to narrow a sofa or bedroom search, with product facts, reviews and delivery terms an assistant can read and check. (updated 2026-10-08)
- [GEO agency vs SEO agency](https://underneath.agency/resources/geo-agency-vs-seo-agency): An SEO agency ranks pages in search results; a GEO agency gets a brand cited and recommended in AI answers. What each does, what they cost, and which you need. (updated 2026-09-28)
- [How should global brands approach GEO across languages?](https://underneath.agency/resources/geo-strategy-for-global-brands-across-languages): Plan market by market: AI answers change with language and country, so earn local coverage, write native content and measure each market separately. (updated 2026-10-08)
- [How much control do we have when Google changes how search results look?](https://underneath.agency/resources/google-search-design-changes-traffic-risk): Very little: Google can shift traffic with design changes publishers cannot control. AI Overviews cut English Wikipedia’s search visits by about 5%. (updated 2026-10-08)
- [How can a health insurer win new members when shoppers ask AI to explain their plan choices?](https://underneath.agency/resources/health-insurers-members-ai-search): By making plan facts, networks and ratings accurate and public before enrollment opens, because shoppers now ask AI tools to explain and compare coverage. (updated 2026-10-08)
- [How can a hospital, clinic or practice win new patients from AI search?](https://underneath.agency/resources/healthcare-providers-patients-ai-search): By making providers easy for AI to name and describe correctly: full physician profiles, consistent facts, strong reviews. Patients now ask AI who to see. (updated 2026-10-08)
- [Can AI search help a healthcare software company win health system deals?](https://underneath.agency/resources/healthcare-software-ai-search): Yes, if AI answers can find proof of EHR integration, HIPAA security and peer results, because hospitals now narrow long vendor lists before any demo. (updated 2026-10-08)
- [Can AI search help a healthcare startup compete with established health brands?](https://underneath.agency/resources/healthcare-startups-demand-ai-search): It can, if AI answers find independent proof of outcomes; buyers now pay for results and patients ask AI first, but new companies are rarely named unprompted. (updated 2026-10-08)
- [How does a healthtech company get named when employers and health plans ask AI for vendors?](https://underneath.agency/resources/healthtech-employers-payers-ai-search): With independent evidence, plain outcome and cost facts, and coverage buyers trust. Employers are cutting vendors, and AI is an early place they look. (updated 2026-10-08)
- [How do home decor brands get found when shoppers ask AI for a look, not a product?](https://underneath.agency/resources/home-decor-brands-ai-product-discovery): By showing up in the visual product sets AI tools build from style questions, with images, colors, materials and reviews an assistant can read. (updated 2026-10-08)
- [How do brands build authority that AI search recognizes?](https://underneath.agency/resources/how-brands-build-authority-for-ai-search): By being written about by sources AI engines already trust, in the outlets each engine reads, and by keeping a site AI agents can read. (updated 2026-10-07)
- [How many prompts do we need to track to measure AI visibility?](https://underneath.agency/resources/how-many-prompts-to-track-ai-visibility): There is no fixed number. Studies suggest about 40 to 150 or more prompts per topic per AI engine, each asked several times, before a ranking is trustworthy. (updated 2026-10-08)
- [How many sources does each AI search engine cite per answer?](https://underneath.agency/resources/how-many-sources-ai-search-engines-cite): Anywhere from about 3 to 40 sources per answer, depending on the engine, the topic and how it was measured. The engine ranking flips between studies. (updated 2026-10-11)
- [How much web content is already optimized for AI search?](https://underneath.agency/resources/how-much-web-content-is-optimized-for-ai-search): About 9% of pages Google and Gemini returned for 1,000 real searches showed signs of AI-search optimization, rising to 16.36% for pages updated in 2026. (updated 2026-10-08)
- [How can a small brand get recommended by AI assistants?](https://underneath.agency/resources/how-small-brands-get-recommended-by-ai): Small brands start far behind in AI answers. The best evidence says to get written about by independent sites and make your advantages easy to verify. (updated 2026-10-08)
- [How to choose a GEO agency](https://underneath.agency/resources/how-to-choose-a-geo-agency): How to choose a generative engine optimization (GEO) agency in 2026: eight criteria, a scorecard, the red flags, and the questions to ask before you sign. (updated 2026-09-28)
- [How should you design prompts and runs to track AI visibility?](https://underneath.agency/resources/how-to-design-ai-visibility-tracking): Spread your budget across languages, assistants and wordings before buying more repeats of one prompt, and fix the question set and run count in advance. (updated 2026-10-08)
- [How can an HR consulting firm win clients through AI search?](https://underneath.agency/resources/hr-consulting-firms-clients-ai-search): By being the firm AI answers connect to a named people problem, such as pay transparency or work redesign, backed by research and coverage others cite. (updated 2026-10-08)
- [Is AI search deciding which HR software gets the demo?](https://underneath.agency/resources/hr-software-ai-search): Increasingly it shapes who gets considered. HR software vendors win demos when AI answers can verify their fit, security and compliance. (updated 2026-10-08)
- [Will AI assistants name your identity platform when a credential breach sends buyers looking?](https://underneath.agency/resources/iam-enterprise-demand-ai-search): Identity vendors get named in AI answers when their standards support, security record and facts are public, consistent and confirmed by independent sources. (updated 2026-10-08)
- [Will AI help put our pumps, compressors and valves on the bid list?](https://underneath.agency/resources/industrial-equipment-leads-ai-search): Equipment makers reach more bid lists when AI can verify their duty ranges, efficiency data, standards and local service, the facts engineers specify on. (updated 2026-10-08)
- [When an engineer asks AI for a part, will it name ours?](https://underneath.agency/resources/industrial-manufacturers-ai-search): By making every part number, spec and equivalent easy for AI to verify, on your site and in distributor catalogs, before engineers design it in. (updated 2026-10-08)
- [Will plant leaders find our MES or IIoT platform when they ask AI for options?](https://underneath.agency/resources/industrial-software-plants-ai-search): When plant teams research MES, maintenance or industrial data platforms with AI, the vendors named are those whose plant-level facts are easy to verify. (updated 2026-10-08)
- [How can an insurance carrier or agency win more quotes when shoppers ask AI?](https://underneath.agency/resources/insurance-companies-customers-ai-search): By being named, correctly and state by state, when shoppers ask AI to explain and compare coverage. 32% of auto shoppers already use AI tools. (updated 2026-10-08)
- [How can an insurtech win customers from incumbent insurers when shoppers ask AI?](https://underneath.agency/resources/insurtech-customers-ai-search): By being the clear, well-reviewed alternative AI can explain when shoppers ask about switching. Insurify says AI-sourced new customers rose from 0.15% to 5.7%. (updated 2026-10-08)
- [How do investing apps and robo-advisors get chosen when people ask AI which platform to use?](https://underneath.agency/resources/investment-platforms-customers-ai-search): By being the clear answer for a specific use case, with fees and safety facts stated plainly and confirmed by the comparison sites AI cites. (updated 2026-10-08)
- [Is a GEO agency worth it?](https://underneath.agency/resources/is-a-geo-agency-worth-it): Is a GEO agency worth it? When it pays back, when it does not, what it costs in 2026, and how the AI engines themselves answer the question. (updated 2026-09-27)
- [Is ChatGPT replacing Google search?](https://underneath.agency/resources/is-chatgpt-replacing-google): Not on current evidence. People use ChatGPT alongside Google: search appeared in 46.0% of assistant sessions in a 2026 US and UK browsing panel. (updated 2026-10-10)
- [Is traditional SEO still important for visibility in AI search?](https://underneath.agency/resources/is-seo-still-important-for-ai-search): Yes. Pages that rank higher are cited more often, but Google rankings explain only part of AI citations, and less for ChatGPT than for Google’s own AI. (updated 2026-10-11)
- [Is tracking ChatGPT alone enough to measure our AI search visibility?](https://underneath.agency/resources/is-tracking-chatgpt-enough): No. AI engines cite largely different sources and pick different brands for the same question, so ChatGPT alone shows less than half the picture. (updated 2026-10-08)
- [How can an IT consulting firm win more advisory clients when mid-size companies ask AI first?](https://underneath.agency/resources/it-consulting-firms-clients-ai-search): By being the independent adviser AI names for a company’s size, industry and decision, such as an ERP choice or a fractional CIO, with proof buyers can check. (updated 2026-10-08)
- [How can an IT services firm win more projects and support calls from AI search?](https://underneath.agency/resources/it-service-firms-leads-ai-search): By being the local or specialist firm AI names for a specific problem, place and project, backed by reviews, listings and clear proof of what you fix. (updated 2026-10-08)
- [How can a jewelry brand get recommended when shoppers ask AI?](https://underneath.agency/resources/jewelry-brands-ai-product-recommendations): By being the jeweler AI names for ring and gift questions, with current facts on certification, sourcing, price and where to buy, before peak seasons. (updated 2026-10-08)
- [How can legal software companies win law firms and in-house teams through AI search?](https://underneath.agency/resources/legal-software-ai-search): By being named, accurately, when lawyers ask AI which tools fit their practice, with public proof of accuracy, confidentiality and ethics that buyers can check. (updated 2026-10-08)
- [What separates legitimate GEO from manipulation?](https://underneath.agency/resources/legitimate-geo-vs-manipulation): Truth and transparency. Legitimate GEO makes real, verifiable facts easier for AI to find; manipulation fakes evidence, hides commands or hides motives. (updated 2026-10-08)
- [How do lending platforms reach borrowers who ask AI where to get a loan?](https://underneath.agency/resources/lending-platforms-borrowers-ai-search): By being named, with accurate rates, terms and reputation, when borrowers ask AI where to get a loan. Visibility brings applications, never approvals. (updated 2026-10-08)
- [When a shipper asks AI for a 3PL, does our warehouse network make the list?](https://underneath.agency/resources/logistics-providers-shipper-contracts-ai-search): 3PL and fulfillment providers reach more shippers in AI search when locations, capabilities, integrations and outside proof are easy to verify. (updated 2026-10-08)
- [How can logistics software companies win more customers through AI search?](https://underneath.agency/resources/logistics-software-ai-search): By being named, accurately, when shippers, carriers and 3PLs ask AI assistants which TMS, WMS or fleet tool fits, then turning that into demos. (updated 2026-10-08)
- [How do luxury brands win the high spenders who now ask AI what to buy?](https://underneath.agency/resources/luxury-brands-ai-search): By shaping the sources AI cites when high spenders ask open questions: most luxury AI prompts name no brand, and official sites are a minority of citations. (updated 2026-10-08)
- [How do machine learning companies get found when enterprise buyers ask AI who to hire?](https://underneath.agency/resources/machine-learning-companies-ai-search): By earning the independent proof AI assistants draw on, so they name you when enterprises ask who can build or supply a machine learning solution. (updated 2026-10-08)
- [Will AI steer a plant’s next machine purchase toward us or a rival?](https://underneath.agency/resources/machinery-companies-buyers-ai-search): Machinery makers get shortlisted when AI can verify their applications, cycle times, prices, financing and dealer support, the facts a capital case rests on. (updated 2026-10-08)
- [When a CEO asks AI which consulting firm to hire, will ours be named?](https://underneath.agency/resources/management-consulting-firms-clients-ai-search): Only if its reputation is visible in sources AI reads: rankings, client perception studies, case evidence and independent coverage, not just its own claims. (updated 2026-10-08)
- [How do marketing software companies win buyers who ask AI first?](https://underneath.agency/resources/marketing-software-ai-search-growth): By being named when marketing leaders ask AI to map a category: most CMOs now start software searches in AI tools, and they shortlist about three vendors. (updated 2026-10-08)
- [How does an online marketplace win more buyers and sellers when people ask AI?](https://underneath.agency/resources/marketplace-buyers-sellers-ai-search): By being the answer on both sides: where buyers should shop and where sellers should list. AI traffic is still small for marketplaces, but it converts. (updated 2026-10-08)
- [How can a mattress brand win customers who ask AI which mattress to buy?](https://underneath.agency/resources/mattress-brands-customers-from-ai-search): By being on the short list when shoppers ask AI to compare mattresses, with verifiable facts, real reviews, clear trial terms and no unsupported sleep claims. (updated 2026-10-08)
- [How can MDR and managed SOC providers win qualified leads from AI search?](https://underneath.agency/resources/mdr-providers-qualified-leads-ai-search): By being named, with defined response metrics and independent proof, when stretched IT and security leaders ask AI which MDR or managed SOC fits them. (updated 2026-10-08)
- [How should we measure share of citations in AI search tools?](https://underneath.agency/resources/measuring-share-of-citations-in-ai-search): Count every answer, including those that never searched the web, report each engine separately, and use enough repeated answers to tell real gaps from noise. (updated 2026-10-08)
- [Can AI search build demand for a medical device before the value analysis committee meets?](https://underneath.agency/resources/medical-device-demand-ai-search): It can shape what clinicians and hospital committees read about your device, if peer-reviewed evidence and clear FDA facts are easy for AI tools to find. (updated 2026-10-08)
- [How can a medical equipment company win more hospital RFQs when buyers ask AI first?](https://underneath.agency/resources/medical-equipment-leads-ai-search): Hospital and clinic buyers are adopting AI fast. Equipment sellers win RFQs when assistants can verify their prices, contracts, service and evidence. (updated 2026-10-08)
- [Can AI search bring independent practices and specialty clinics to a medical software vendor’s demo?](https://underneath.agency/resources/medical-practice-software-ai-search): Yes, if assistants find proof of specialty fit, certification, pricing and peer ratings; 23% of medical groups expect to switch or overhaul their EHR in a year. (updated 2026-10-08)
- [How do mortgage lenders and marketplaces win qualified borrowers when buyers start with AI?](https://underneath.agency/resources/mortgage-platforms-leads-ai-search): By being named, with accurate loan facts, when borrowers ask AI to estimate payments and compare lenders, and by keeping rate claims compliant. (updated 2026-10-08)
- [Which AI engine gives the most consistent answers about brands?](https://underneath.agency/resources/most-consistent-ai-engine-for-brands): No AI engine is steadiest on every measure. Perplexity repeated brand lists most in two tests; Gemini and ChatGPT described brands most steadily in another. (updated 2026-10-08)
- [How can a managed service provider win more clients from AI search?](https://underneath.agency/resources/msps-customers-from-ai-search): By being named when owners and IT leads describe their size, region and needs to AI, with pricing, coverage and security proof they can check. (updated 2026-10-08)
- [How does a neobank turn AI answers into accounts that receive a paycheck?](https://underneath.agency/resources/neobanks-account-openings-ai-search): By answering the safety and fee questions AI users ask about app-only banks, so the account they open becomes the one their paycheck goes to. (updated 2026-10-08)
- [How do network security vendors win buyers who ask AI about zero trust?](https://underneath.agency/resources/network-security-zero-trust-customers-ai-search): By being named, in NIST and CISA terms, in AI answers about firewall refreshes, SASE and zero trust access, with a public security record buyers can check. (updated 2026-10-08)
- [Do off-site mentions help once an AI agent is reading about you?](https://underneath.agency/resources/off-site-mentions-vs-site-readability-ai-agents): Less than you might think. Once an AI agent researches a named business, whether it can read the site mattered far more than off-site mentions. (updated 2026-10-08)
- [What on-page signals are linked to citations in Google AI Overviews and Perplexity?](https://underneath.agency/resources/on-page-signals-linked-to-ai-citations): Dates, clean HTML structure and structured data track with AI citations, but ranking matters more, and only a machine-readable date held up on Google. (updated 2026-10-08)
- [Can I trust a one-time AI visibility report for my brand?](https://underneath.agency/resources/one-time-ai-visibility-report): Not on its own. AI answers vary between runs, so a one-off report can mistake noise for a lead. Reliable figures need repeated runs over two to four weeks. (updated 2026-10-08)
- [How does an outdoor brand get onto the shortlist when people ask AI what gear to buy?](https://underneath.agency/resources/outdoor-brands-ai-search): By matching gear to specific activities and conditions, and earning the expert reviews and retailer guides AI draws on when it builds an outdoor shortlist. (updated 2026-10-08)
- [When a brand asks AI for a packaging supplier, will our name come up?](https://underneath.agency/resources/packaging-suppliers-quote-requests-ai-search): Brand owners now use AI in packaging sourcing. Packaging suppliers win quote requests when assistants can verify their materials, compliance and minimums. (updated 2026-10-08)
- [When a new business asks AI how to take payments, how does a payment company get chosen?](https://underneath.agency/resources/payment-companies-merchants-ai-search): By being named, with the right fees, when a new business asks AI how to take payments. Owners choose on cost and simplicity, and rarely switch later. (updated 2026-10-08)
- [How does a payment processor get onto the shortlist when merchants research with AI?](https://underneath.agency/resources/payment-processors-merchant-demand-ai-search): By being named and described accurately when payments teams, engineers and procurement research processors with AI, from pricing models to local acquiring. (updated 2026-10-08)
- [How does a payroll provider get named when businesses ask AI?](https://underneath.agency/resources/payroll-software-ai-search): By being the clear, verifiable answer for a business’s size, states and countries. AI answers on tax-bound questions shift by location and misquote prices. (updated 2026-10-08)
- [How can a penetration testing firm win more scoped leads from AI search?](https://underneath.agency/resources/pentest-firms-leads-ai-search): By being the firm AI names when a buyer describes their audit, scope and deadline, with credentials, methods and pricing logic others can verify. (updated 2026-10-08)
- [How do pet product brands get recommended when owners ask AI what to buy?](https://underneath.agency/resources/pet-product-brands-ai-search): By being the brand AI names, with accurate facts and claims, when owners ask what to buy for their pet; one recommendation can start years of repeat orders. (updated 2026-10-08)
- [How can a drugmaker keep its brands visible and accurate in AI answers?](https://underneath.agency/resources/pharma-brand-visibility-ai-search): By making the label, the safety story and the brand’s facts easy for AI to find and repeat, within FDA rules, since patients and doctors now ask AI first. (updated 2026-10-08)
- [How do you sell procurement software to buyers who run RFPs for a living?](https://underneath.agency/resources/procurement-software-ai-search): Professional buyers still run RFPs, but AI answers now shape which procurement tools reach the longlist, so vendors need proof buyers and AI can check. (updated 2026-10-08)
- [What kind of product content do AI shopping assistants prefer?](https://underneath.agency/resources/product-content-ai-shopping-assistants-prefer): Factual, specific, easy-to-compare content matched to what buyers ask. In tests, hard product facts drove 82.4% of AI rankings; hype and gimmicks backfired. (updated 2026-10-08)
- [How do productivity tools win users when AI assistants compete?](https://underneath.agency/resources/productivity-software-ai-search-growth): By being the tool AI assistants name for one job done better, then converting on day one. AI already sends productivity apps real traffic; suites raise the bar. (updated 2026-10-08)
- [How do professional services firms win clients when buyers ask AI who to hire?](https://underneath.agency/resources/professional-services-firms-clients-ai-search): Clients of law, accounting, design and consulting firms now research with AI first. Firms get named when directories, reviews and coverage confirm expertise. (updated 2026-10-08)
- [How do project management tools win customers from AI answers?](https://underneath.agency/resources/project-management-software-customers-ai-search): By being named for the use-case and team-size questions buyers ask AI, then turning free signups into paid seats that expand. Here is the evidence. (updated 2026-10-08)
- [How do we prove that GEO caused a change in sales or conversions?](https://underneath.agency/resources/prove-geo-caused-sales): You need a comparison group that got no GEO work, ideally chosen at random, because AI traffic grows for everyone. Before-and-after numbers overstate it. (updated 2026-10-08)
- [Can AI search bring a recruiting agency more employer clients?](https://underneath.agency/resources/recruiting-agencies-employer-clients-ai-search): Yes, if employers can see what roles you fill, where, how fast and how well. AI answers lean on local listings, reviews and specialty pages to name agencies. (updated 2026-10-08)
- [What are the risks of native ads blended into AI chatbot answers?](https://underneath.agency/resources/risks-of-native-ads-in-ai-answers): Native ads in AI answers are hard to tell apart from advice, so they can mislead trusting users and damage the advertiser’s brand once spotted. (updated 2026-10-08)
- [Will AI assistants put our robots on a manufacturer’s shortlist before an integrator is called?](https://underneath.agency/resources/robotics-companies-customers-ai-search): It can, if assistants find your applications, specs, integrators and real deployments in sources they trust, as robot buying spreads beyond automotive. (updated 2026-10-08)
- [How can sales software companies turn AI answers into pipeline?](https://underneath.agency/resources/sales-software-pipeline-from-ai-search): By being named when sales leaders ask AI to shortlist tools, then answering the demo requests that follow fast, with proof on data, security and fit. (updated 2026-10-08)
- [How do security awareness training vendors win customers through AI search?](https://underneath.agency/resources/security-awareness-training-customers-ai-search): By being named, with checkable proof, when IT leads, MSPs and CISOs ask AI which training meets their insurer, auditor and budget. (updated 2026-10-08)
- [How do parcel carriers and shipping platforms win small businesses that ask AI how to ship?](https://underneath.agency/resources/shipping-companies-customers-ai-search): Small sellers now ask AI how to ship cheaply, so parcel carriers and shipping platforms need current, citable rates, coverage and reviews to be named. (updated 2026-10-08)
- [Should ads in AI search be clearly labeled, and does labeling help the advertiser?](https://underneath.agency/resources/should-ai-search-ads-be-labeled): Yes. Clear labels protect users, and research argues they protect advertisers too, since disguised ads breed distrust once people spot them. (updated 2026-10-08)
- [How do SIEM vendors get onto enterprise shortlists through AI search?](https://underneath.agency/resources/siem-enterprise-pipeline-ai-search): By being named, with checkable proof, when SOC leaders use AI to build a SIEM long list during a migration, a consolidation or a cost review. (updated 2026-10-08)
- [How do skincare brands get their products found when shoppers ask AI about their skin?](https://underneath.agency/resources/skincare-brands-ai-search): By answering concern and ingredient questions with claims you can substantiate, real expert backing and honest reviews, the signals AI and shoppers both check. (updated 2026-10-08)
- [How do sportswear brands win the sale when shoppers ask AI what to train in?](https://underneath.agency/resources/sportswear-brands-ai-search): By being the product AI names for specific activity and attribute questions, with data clean enough to buy from your site, a retailer or the chat itself. (updated 2026-10-08)
- [How do subscription brands win new subscribers through AI search?](https://underneath.agency/resources/subscription-ecommerce-customers-ai-search): Be the clear, honest answer when shoppers compare plans, prices and cancellation terms; 55% of consumers would welcome AI reordering everyday items. (updated 2026-10-08)
- [How do supplement brands win new customers from AI assistants?](https://underneath.agency/resources/supplement-brands-customers-ai-search): By being the brand assistants can verify: third-party testing, honest labels and strong reviews, now that 32% of US adults ask AI chatbots about health. (updated 2026-10-08)
- [How does a supply chain consultancy get on the shortlist when executives research with AI?](https://underneath.agency/resources/supply-chain-consulting-firms-ai-search): AI answers now help shape which supply chain consultancies get invited to bid, so firms need proof that independent sources repeat, not just a good website. (updated 2026-10-08)
- [Can AI search put our supply chain software on the RFP longlist?](https://underneath.agency/resources/supply-chain-software-ai-search): Partly: AI answers now shape the vendor longlist before an RFP is written, so supply chain vendors must be named and verifiable there first. (updated 2026-10-08)
- [How does a technology consultancy get onto enterprise shortlists when buyers ask AI first?](https://underneath.agency/resources/technology-consulting-firms-ai-search): By being the firm AI answers tie to proven production results in a named platform and industry, backed by analyst, partner and client evidence. (updated 2026-10-08)
- [How do telehealth companies get chosen when patients ask AI where to get care?](https://underneath.agency/resources/telehealth-patients-from-ai-search): By being the service assistants can verify on price, insurance, licensing and reputation, as 58% of AI health users follow up with a provider. (updated 2026-10-10)
- [How do toy brands get found when shoppers ask AI for gift ideas?](https://underneath.agency/resources/toy-brands-product-discovery-ai-search): By making age, play value and safety facts easy for AI to match to gift questions, and earning the reviews and gift-guide coverage AI answers draw on. (updated 2026-10-10)
- [How does AI search bring a virtual care platform new contracts and more patient visits?](https://underneath.agency/resources/virtual-care-platforms-ai-search): By getting the platform onto health system, plan and employer shortlists, and by helping clients’ patients find the virtual front door they paid for. (updated 2026-10-10)
- [Are AI assistants deciding which warehouse automation vendors make the enterprise shortlist?](https://underneath.agency/resources/warehouse-automation-enterprise-leads-ai-search): Partly: buyers now use AI to learn options and build longlists years before an RFP, so AS/RS, AMR and sortation vendors must be named and verifiable there. (updated 2026-10-10)
- [How do wealth management firms win new clients when prospects check them with AI before calling?](https://underneath.agency/resources/wealth-management-firms-clients-ai-search): Referrals still lead, but prospects increasingly check a firm with AI. Firms win by making their niche, fees, fiduciary status and reviews easy to verify. (updated 2026-10-10)
- [Are websites hiding instructions to manipulate how AI search presents them?](https://underneath.agency/resources/websites-hiding-instructions-for-ai-search): Yes. A scan of 1.2 billion web addresses found 1,521 hidden instructions trying to make AI systems promote, cite or praise a page. Most rarely work. (updated 2026-10-08)
- [Will AI assistants recommend our wellness products when people ask how to sleep, train or recover better?](https://underneath.agency/resources/wellness-brands-customers-ai-search): By being the product assistants can describe and verify when people ask how to sleep, train or recover better, with wellness claims that stay lawful. (updated 2026-10-10)
- [What does content that AI engines prefer look like?](https://underneath.agency/resources/what-content-do-ai-engines-prefer): AI engines favor pages that state the conclusion first, cover the topic fully, explain how and why, and pack checkable facts such as definitions and figures. (updated 2026-10-08)
- [What does a GEO agency do?](https://underneath.agency/resources/what-does-a-geo-agency-do): A GEO agency gets a brand cited, mentioned and recommended in AI answers. The nine things it does, what it costs, and how the AI engines describe the job. (updated 2026-09-27)
- [What actually drives which product an AI assistant recommends?](https://underneath.agency/resources/what-drives-ai-product-recommendations): When AI assistants can see ratings, prices and reviews, that data decides the pick. Brand name mostly breaks ties when products look the same. (updated 2026-10-08)
- [What GEO practices are actually supported by research?](https://underneath.agency/resources/what-geo-practices-does-research-support): Research supports a conservative GEO playbook: relevant, complete, verifiable, well-structured pages that AI can reach, measured stage by stage. Tricks fail. (updated 2026-10-08)
- [What happens to brands that skip GEO when competitors adopt it?](https://underneath.agency/resources/what-happens-if-you-skip-geo): In controlled tests, brands that did nothing while rivals optimized got zero AI recommendations. The early lead shrinks fast, so copying rivals is not enough. (updated 2026-10-08)
- [What is GEO, and how is it different from SEO?](https://underneath.agency/resources/what-is-generative-engine-optimization): GEO is the work of getting your information found, used and cited in AI-written answers. SEO competes for a rank; GEO competes for a share of the answer. (updated 2026-10-08)
- [What should we measure to know whether we are visible in AI answers?](https://underneath.agency/resources/what-to-measure-ai-visibility): Measure how often each AI engine names and cites you in final answers, over repeated runs, and whether what it says is right, not rankings or one screenshot. (updated 2026-10-08)
- [If AI agents can’t read my website, where does their answer about us come from?](https://underneath.agency/resources/when-ai-agents-cant-read-your-site): From web searches and other people’s pages. In one test, 42% of an AI agent’s answer came from outside a business whose site it could not read. (updated 2026-10-11)
- [Which Google searches trigger an AI Overview?](https://underneath.agency/resources/which-searches-trigger-ai-overviews): Long, specific questions trigger Google AI Overviews most; local, brand and breaking-news searches rarely do. Overall rates run from 13.7% to 67%. (updated 2026-10-10)
- [Which types of websites do AI search engines rely on most?](https://underneath.agency/resources/which-websites-do-ai-search-engines-cite): It depends on the question. Official, reference and news sites lead on public-interest topics; review publishers lead on buying questions. (updated 2026-10-11)
- [Who in my organization should own AI search visibility?](https://underneath.agency/resources/who-should-own-ai-search-visibility): No single team: SEO owns being found, content and product own what pages say, PR owns third-party coverage, and one person should own measurement. (updated 2026-10-08)
- [Why does ChatGPT give a different answer about my brand each time?](https://underneath.agency/resources/why-ai-answers-about-your-brand-change): AI assistants pick each answer partly by chance, so a brand named once may vanish on the next run. Judge visibility by rates over many runs. (updated 2026-10-08)
- [What decides whether an AI engine cites my page over a competitor’s?](https://underneath.agency/resources/why-ai-cites-competitor-page-first): In a 252,000-trial test, topic match, list position, a stated price and a recent date decided which page AI cited first. Formatting barely mattered. (updated 2026-10-08)
- [What makes ChatGPT or Gemini recommend one hotel over another?](https://underneath.agency/resources/why-ai-recommends-one-hotel-over-another): Guest rating and price decide most AI hotel picks; eco-certification and review count help, list order matters, and replying to reviews did nothing. (updated 2026-10-08)
- [Why do AI search engines cite different sources every time I check?](https://underneath.agency/resources/why-ai-search-citations-change): Because AI search engines pick sources partly at random each time. In one 45-day study, about 65% of cited sources changed from one day to the next. (updated 2026-10-08)
- [Why can’t our marketing analytics see our visibility in AI answers?](https://underneath.agency/resources/why-analytics-miss-ai-visibility): Analytics log clicks and paid impressions. Most AI answers are read without a click, so brand mentions in them leave little or no trace in your data. (updated 2026-10-08)
- [Why doesn’t ChatGPT mention our newly launched product?](https://underneath.agency/resources/why-chatgpt-misses-new-products): ChatGPT often answers from training data that predates your launch, and even assistants that search rarely surface new products. Here is what helps. (updated 2026-10-08)
- [Why is our well-known brand missing from AI product recommendations?](https://underneath.agency/resources/why-well-known-brands-miss-ai-recommendations): AI assistants often know a famous brand but fail to call it up for a plain category question. Research shows why, and how to check your own brand. (updated 2026-10-08)
- [Why does Wikipedia matter so much for AI search visibility?](https://underneath.agency/resources/why-wikipedia-matters-for-ai-search): Wikipedia is the most cited site in many AI search studies and shapes answers beyond its citations. For buyer questions its share is small. (updated 2026-10-11)

## Machine-readable formats

- Entity page (HTML): https://underneath.agency/agent (text/html)
- Entity page (Markdown): https://underneath.agency/agent.md (text/markdown)
- Entity record (JSON-LD): https://underneath.agency/.well-known/entity.json (application/ld+json)
- Site index for language models: https://underneath.agency/llms.txt (text/plain)
- Every page, concatenated as Markdown: https://underneath.agency/llms-full.txt (text/markdown)
- Sitemap: https://underneath.agency/sitemap.xml (application/xml)
- Practice page (HTML): https://underneath.agency/services/generative-engine-optimization (text/html)
- Every HTML page has a Markdown twin at the same path with `.md` appended (for example https://underneath.agency/about.md), and answers `Accept: text/markdown` with it.

Tagline: Underneath every answer is a better question.