# Daavid Christa I help teams understand AI search and make it a growth channel. I started coding at 12 to try to build the next facebook (took me 7 months to get chat running, I hate php), joined Tesla to build a future in engineering/robotics and ended up in tech sales. Previous roles: Salesforce, Navan, Peec AI, Aircall and Tesla. Now: Founding Account Executive at Searchable. Website: https://daavidchrista.com/ Book a free conversation: https://calendly.com/daavid-searchable/senior-living-ai-search-review LinkedIn: https://www.linkedin.com/in/daavid-christa/ X: https://x.com/daavidchrista YouTube: https://www.youtube.com/@daavidchrista ## Projects # Collection and external sources External articles and social posts are summarized here; follow the original source for the complete publication. Customer outcomes are attributed to their publisher, not claims of personal results. This file covers the published site content, not private visitor data, local caption edits or unpublished drafts. ## Prompt universe planner Map the questions across your business. Source: https://daavidchrista.com/senior-living/prompt-planner/ ## Move-in economics Put your own numbers behind it. Source: https://daavidchrista.com/senior-living/economics/ ## Community facts workbook Make your community facts clear. Source: https://daavidchrista.com/senior-living/facts-check/ ## GEO. SEO. AEO. What changes. What stays. Author: Kate Starr Publisher: searchable Published: 2026-09-11 Strong content and technical accessibility support all three. Measure rankings, answers and citations separately. Source: https://www.searchable.com/blog/geo-vs-seo-vs-aeo ## Less planning. More doing. Less content-planning time 70% — Less content-planning time · reported by Searchable Quote: The shift from monitoring to execution is everything. Publisher: searchable Source: https://www.searchable.com/customers/1mind ## Build the business case. Model three scenarios. Make your assumptions visible. Source: https://daavidchrista.com/tools/roi-studio/ ## Measure the right questions. Plan prompts, markets, engines, and the review work behind them. Source: https://daavidchrista.com/tools/prompt-budget/ ## From AI search to pipeline. Reported pipeline £1m+ — Reported pipeline · Searchable customer story Quote: Searchable is a must-have for any agency. Publisher: searchable Source: https://www.searchable.com/customers/303 ## A signal. Or just noise? Put uncertainty next to your visibility numbers. Source: https://daavidchrista.com/tools/visibility-confidence/ ## Visibility is not the same as traffic. Know which question each metric actually answers. Publisher: searchable Source: https://www.searchable.com/features/aeo-insights/visibility ## Less tool-switching. More doing. Saved in 30 days 170h — Saved in 30 days · reported by Searchable Publisher: searchable Source: https://www.searchable.com/customers/blackbird ## Does the source support the claim? Keep a reviewable trail for your next content decision. Source: https://daavidchrista.com/tools/citation-evidence/ ## Look beyond your own website. Explore which sources shape how AI describes a brand. Publisher: searchable Source: https://www.searchable.com/blog/improve-ai-brand-visibility ## Map the questions across your business. Senior living · Choose your locations, care services, markets, and topics. Build a custom question inventory and a focused starting panel. Source: https://daavidchrista.com/senior-living/prompt-planner/ ## Put your own numbers behind it. Senior living · Explore additional demand and direct-versus-referral acquisition. Source: https://daavidchrista.com/senior-living/economics/ ## Make your community facts clear. Senior living · Review services, fees and sources. Export an evidence-based action list. Source: https://daavidchrista.com/senior-living/facts-check/ ## How do families find senior living now? Senior living research · November 2026 · 9 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-global-field-guide/ ## AI knows your name. Does it recommend your communities? Senior living discovery research · 8 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-who-owns-the-answer/ ## What would one extra move-in be worth? AI search and resident economics · 8 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-answer-to-move-in/ ## What are senior living providers seeing from AI search? Evidence from senior living AI search · 8 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-case-study-evidence/ ## What I’d do in your first 90 days of AI search Research on time to results · 8 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-first-90-days/ ## What is AI getting wrong about your communities? The evidence behind community facts · 8 min read Author: Daavid Christa Published: 2026-11-02 Source: https://daavidchrista.com/research/senior-living-community-facts/ ## How I’d get your business into AI answers AI search: business evidence · 8 min read Author: Daavid Christa Published: 2026-11-03 Source: https://daavidchrista.com/research/geo-implementation-playbook/ ## Is your AI visibility actually improving? Measurement research · 8 min read Author: Daavid Christa Published: 2026-10-19 Source: https://daavidchrista.com/research/measure-ai-search-visibility/ ## AI cited your business. What did it actually say? AI citation research · 8 min read Author: Daavid Christa Published: 2026-11-03 Source: https://daavidchrista.com/research/ai-citations-evidence/ ## Should you pay for AI search ads? AI advertising research · 9 min read Author: Daavid Christa Published: 2026-11-03 Source: https://daavidchrista.com/research/ai-search-ads-brand-playbook/ ## Your data. Inside ChatGPT. Meet the Searchable connector. Author: Sam Hogan Publisher: searchable Published: 2026-08-19 Ask about visibility, citations and competitors using your Searchable data. Bring connected GA4 and Search Console signals into the conversation. Source: https://www.searchable.com/blog/introducing-searchable-chatgpt-connector ## Beyond the web search. Claude Lite is here. Author: Ivan Slobodin Publisher: searchable Published: 2026-10-05 Track how Claude describes a brand without browsing. Separate memory-based answers from web-assisted answers. Source: https://www.searchable.com/blog/claude-lite-tracking-searchable ## Life, in weeks. this is my life in weeks so far. i'm 22 and have two conflicting realities: 1. time is running out 2. i have my whole life ahead of me Author: Daavid Christa Publisher: daavid Published: 2026-01-09T17:42:59.770Z Source: https://www.linkedin.com/posts/daavid-christa_this-is-my-life-in-weeks-so-far-im-22-activity-7415448017652084736-ThN_ ## The buying journey has changed. the buying journey is no longer: google > your website > demo > decision it’s chatgpt > google > chatgpt > your website > demo > chatgpt > decision Author: Daavid Christa Publisher: daavid Published: 2026-01-05T08:45:53.769Z Source: https://www.linkedin.com/posts/daavid-christa_its-january-2026-youre-excited-and-ready-activity-7413863300544942080-A0pd ## The questions before your brand. Most teams track the prompts buyers ask last. Not the ones they ask first... Author: Searchable Publisher: searchable Published: 2026-10-19T13:34:52.505Z Source: https://www.linkedin.com/posts/trysearchable_most-teams-track-the-prompts-buyers-ask-last-activity-7517941341901754369-E1FX ## Meet Prompt Universe. Brands are tracking a few hundred prompts they don't know are right or wrong. Author: Searchable Publisher: searchable Published: 2026-10-17T12:28:15.618Z Source: https://www.linkedin.com/posts/trysearchable_brands-are-tracking-a-few-hundred-prompts-activity-7517199802011418624-yw0F ## When referral tracking changes. ChatGPT referral traffic fell 65% last week. And this week it's already back... Author: Searchable Publisher: searchable Published: 2026-10-16T09:26:10.506Z Source: https://www.linkedin.com/posts/trysearchable_chatgpt-referral-traffic-fell-65-last-week-activity-7516791590905012224-pFF6 ## Searchable in ADWEEK. ADWEEK has named Searchable one of the 11 best-funded startups racing to win AI search. Author: Searchable Publisher: searchable Published: 2026-10-15T12:28:51.584Z Source: https://www.linkedin.com/posts/trysearchable_adweek-has-named-searchable-one-of-the-11-activity-7516475177132777472-OR3I ## The missing work behind the software. Tool → partners → working workflow. Author: Daavid Christa Publisher: rampyear Published: 2026-10-06 Source: https://rampyear.com/research/clay-ecosystem/ ## The enterprise sale continues after the demo. Capability → evaluation → ownership → adoption. Author: Daavid Christa Publisher: rampyear Published: 2026-10-06 Source: https://rampyear.com/research/openai-enterprise/ ## From individual enthusiasm to company rollout. Individual use → team context → rollout. Author: Daavid Christa Publisher: rampyear Published: 2026-10-06 Source: https://rampyear.com/research/cursor-adoption/ ## Coding since 12. I’ve been coding since I was 12 and it’s time to use my technical + sales skills, combine them and create the app I wish I had as an Account Executive/BDR. Author: Daavid Christa Publisher: daavid Published: 2025-09-09 Source: https://x.com/daavidchrista/status/1965474668007280823 ## My first business. first “job” was selling my custom made Minecraft plugins to all types of big servers at 14 Author: Daavid Christa Publisher: daavid Published: 2025-09-10 Source: https://x.com/daavidchrista/status/1965678757815222283 ## Who owns the AI shopping shelf? Who is the most popular retailer in ChatGPT in 2026? Author: Searchable Publisher: searchable Published: 2026-07-22 Source: https://x.com/trysearchable/status/2079868871427641394 ## A more visual Agent. Searchable’s Agent just got a major upgrade. Author: Searchable Publisher: searchable Published: 2026-07-20 Source: https://x.com/trysearchable/status/2079223104635879459 ## See the answers behind the score. Visibility, mentions and sources in one place. Publisher: searchable Source: https://www.searchable.com/features/aeo-insights ## Start with your buyers’ questions. A map of topics, intent and opportunity. Publisher: searchable Source: https://www.searchable.com/features/prompts ## Turn the next gap into content. Move from research to a useful first draft. Publisher: searchable Source: https://www.searchable.com/features/content ## Follow the visit. Find the source. Understand AI referrals and crawler activity. Publisher: searchable Source: https://www.searchable.com/features/llm-analytics # Research — full articles # How do families find senior living now? URL: https://daavidchrista.com/research/senior-living-global-field-guide/ Author: Daavid Christa Published: 2026-11-02 Families are asking AI about care, costs, and which communities to visit. Here’s where I’d start if I ran your AI search strategy. Start with the questions a family asks before they know your name. Then check whether your communities appear, whether the information is right, and whether it helps the family take the next step. The research below shows why that matters, what providers are already seeing, and how I’d approach it across different markets. ## Start with the family’s shortlist A family is looking for somewhere their parent can live. They’re comparing care, location, costs, and whether the place feels right. By the time they call you, they may already have ruled out most of the market. Search Influence surveyed 300 people researching senior living or who had searched in the previous year. In that group, 57% seriously considered only two or three communities. The same survey says 44% removed a community because of something they read online. It also says 43% started with a search engine or an AI tool. That last figure combines the two; it doesn’t tell us how many started with AI alone. Cost transparency was both the leading comparison factor and the information people struggled most to find. So here’s where I’d start: what does a family learn about your community before anyone on your team speaks to them? If the answer on fees is vague, or your care services are hard to understand, someone else gets to explain your offer for you. That could be a directory, a competitor, or an AI answer using an old page. Your admissions team sees the families who get in touch. They don’t see every family who quietly removes you from their list. That’s the gap I want an AI search strategy to look at first. The survey is useful context, though its public summary doesn’t give us enough detail to treat the figures as a worldwide benchmark. Published family research | Reported result Seriously considered two or three communities | 57% Ruled out a community because of online information | 44% Started with search or AI combined | 43% ## Then look beyond website clicks Pew studied browsing from 900 US adults. People clicked a traditional Google result in 8% of visits with an AI summary, compared with 15% without one. They clicked a source inside the summary in just 1% of those visits. The browsing was from March 2025, with results reconstructed in April. This was general search behaviour, not a senior living test. Think about what a family can do without clicking. Learn what assisted living means. Compare fee structures. Remember a provider’s name. Decide a service sounds unsuitable. The study doesn’t measure those specific care decisions, but it gives you a good reason to look at the answer as well as the visit. I’d still track website traffic. It helps connect the work to real inquiries. I’d also save the answers to important family questions and check what they say about each community. You need both views to understand what’s happening before and after someone reaches your site. ## Does AI know your name, or actually introduce you? Here’s the result that made this distinction concrete for this site. An anonymised Canadian retirement-operator report had 100 generated ChatGPT questions. The operator appeared in 17 answers. In all 13 questions that already named the operator, it appeared. In the other 87 questions, it appeared four times. That’s 100% for the questions with its name and about 4.6% for the questions without it. We checked the report on 3 November 2026; its original collection date wasn’t shown. Some questions in the second group could name a competitor. This tells us about one sample, rather than Canadian market share. If you only asked about your own brand, you’d probably feel pretty good about that result. But a family looking for options hasn’t necessarily heard of you yet. I’d want to know whether AI introduces you when the question describes a need your community can meet. The topic breakdown gives us somewhere to look: 17 cost questions had no operator mentions, while 23 assisted-living questions produced three. That doesn’t prove the operator belonged in every answer. Open the actual questions, check the fit, and find out what information the answer needed. You can explore the counts below. ## Look at what providers are already doing Search Influence reports that Liberty Senior Living’s AI-sourced visits grew 487% over twelve months. It reports a 17.3% conversion rate for those visits, compared with 11.3% across the site. Its account describes wider marketing work too, so I wouldn’t credit AI alone for the occupancy results. Arc Intermedia reports more than sevenfold year-over-year AI traffic growth for Acts Retirement-Life Communities. It says those visitors converted 36% better than average. That means a relative improvement, not a 36% conversion rate. The work included content, site structure, listings, and outside sources. Those are interesting signals from named providers. They give us a reason to test the channel properly. They don’t tell us what return your communities will get, and the published website conversions aren’t the same as completed move-ins. What I’d take from both cases is the range of the work. Your website matters, and so does the information around it. Check the directories, reviews, official records, and comparison pages that appear in answers. Fixing your homepage won’t necessarily fix the explanation a family sees elsewhere. Provider | Publicly reported signal | Business interpretation Liberty Senior Living | 17.3% AI conversion rate; 11.3% sitewide | A website conversion result, not a move-in rate Acts Retirement-Life Communities | >7× AI traffic year over year; 36% better conversion | Reported growth and relative conversion improvement ## Build the strategy around each market An international strategy needs more than a translated prompt. The care category, payment route, and evidence a family needs can change by country. If AI describes the wrong kind of service, a fluent answer is still a bad introduction to your business. The US National Institute on Aging distinguishes assisted living, nursing homes, and other residential settings. In England, CQC publishes reports and ratings for individual homes. France distinguishes résidence autonomie from résidence services. Australia separates retirement villages from aged care facilities. Those differences belong in the questions you track and the pages you publish. I’d build one local question set with the people who understand that market. Ask them what families call the service, what they misunderstand, and what they need to check before a visit. Then use the same approach in the next market, with its own facts. The comparison below shows examples from official sources. Use it to spot where your international content needs local input. It isn’t a ranking of countries or a claim that we tested AI answers in every market. Market | Evidence that changes the answer | Why it matters in AI search United States | Different residential care categories [7] | An assisted-living option should not be described as equivalent to a nursing home England | A specific home’s CQC record [8] | A group reputation cannot substitute for that home’s inspection evidence France | Distinct housing status and resident arrangements [9] | Similar-looking senior residences can have different contracts and support routes Australia | Retirement villages differ from aged care facilities [10] | Housing and ongoing care are not the same offer Germany | Care insurance is partial protection; other costs remain [11] | An insurance benefit is not the resident’s final bill Singapore | Subsidized residential care has eligibility and means-testing rules [12] | A quoted subsidy ceiling is not every family’s entitlement ## Keep cost answers up to date Australia is a good example of why dates matter. Its official guidance separates fee arrangements by entry date and circumstances. Arrangements introduced on 1 November 2025 include a hotelling contribution and a non-clinical care contribution for some residents, while older arrangements and protections can still apply. One undated price explanation can mislead different families in different ways. Singapore enhanced residential long-term care subsidies from 1 July 2026. The published maximum for eligible citizens born in 1969 or earlier is 80%, subject to conditions and a household means test. Copying “80% subsidy” into a page without those conditions would give families the wrong expectation. I’d put cost and eligibility pages on a regular review list. Give each page someone who can confirm the facts, show when it was checked, and explain what a family still needs to ask. This is useful for people reading your site and for systems using it as a source. Update the tracked questions when the business changes too. A new care offer, a funding change, or a different admission process can change what a good answer looks like. Your AI search work should follow those changes. ## Give families somewhere to check the answer HumanGood’s guidance for families, published on 18 August 2026, treats AI as a way to narrow options and compare care. It tells readers to confirm changing details such as prices and availability with community teams and visits. That’s a provider explaining how families can use the technology, rather than an independent accuracy test. I like that approach because it gives the family a useful next step. Let them research, then make it easy to check the details with you. A clear community page can support an AI answer and give a cautious reader somewhere to verify it. Your public pages should explain the care you offer, who it may suit, fees and inclusions, and how to ask a specific question. Your team can then handle what depends on the individual: assessment, current availability, personal preferences, and the proposed agreement. The goal I’d set is a better first conversation. The family knows why your community might fit, understands what needs confirming, and can reach the right person. That is a more useful aim than trying to make an AI answer sound as though it has settled the whole decision. ## Choose questions that will change what you do We’ve now got two useful starting points. The Canadian sample shows why brand and discovery questions need separate tracking. The market comparison shows why the same wording can mean something different locally. Put those together and build around a real family decision in a real market. For example: can a couple stay together if their care needs differ? What does the monthly fee include? What happens after an assessment? These questions can reveal different gaps at the same community. Asking one generic question for every postcode gives you a lot of prompts, but may teach you very little. Searchable’s Prompt Universe helps map buyer questions from your market context and choose a useful set to track. These are modelled questions, not a feed of people’s private AI chats or a measured search-volume count. I’d judge the set by the decisions it covers and the actions it helps you prioritise. That’s where I can help: work through your locations, care services, and audiences with your team, then build a custom Prompt Universe and get you set up in Searchable. The point is to see where you’re missing, understand why, and agree what to improve first. ## Turn the findings into a plan Here’s the path I’d follow. Pick a community where suitable demand matters. Write down the family decisions you need to show up for. Check the answers, separate discovery from brand questions, and review the pages supplying the information. Then choose a small set of changes you can actually ship. Bring admissions into that conversation early. More interest only helps if the care fits and there’s staffed capacity. If families already inquire but misunderstand the fees, your first priority may be clearer information rather than more exposure. Review both discovery and the quality of the resulting conversations. Are suitable communities appearing? Are the details right? Are inquiries reaching the team with better expectations? Follow those inquiries through to visits and move-ins when the records allow it. This guide brings together published family research, provider cases, official care information, and one anonymised Searchable report. They show different parts of the picture. We haven’t run a global consumer survey or proved how many residents AI search adds. The source notes explain each study’s limits. If you want to apply this to your own communities, bring me one market and the questions your team keeps hearing. We can work out where to start. For the next step, look at the discovery breakdown: it shows how a reassuring headline score can hide the questions you really need to answer. ## Sources - Search Influence: 2026 Senior Living Search Behavior Report: https://info.searchinfluence.com/2026-ai-search-senior-living-research-study/ Public summary of a 300-person family survey, reviewed November 3, 2026. Full sampling methodology was not accessed. Search-or-AI is a combined category; these self-reports are not a global adoption estimate. - Pew Research Center: Google users are less likely to click when an AI summary appears: https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ Published July 22, 2025. Browsing data from 900 U.S. adults in March 2025, with reconstructed search results collected in April. General-search observational evidence, not a senior living causal study. - Anonymized Searchable report: Canadian retirement operator: https://daavidchrista.com/research/senior-living-who-owns-the-answer/ One ChatGPT report, 100 generated questions, reviewed November 3, 2026. 13/13 named-operator questions and 4/87 other questions mention the operator. Collection date unavailable; identity and original report link withheld. - Search Influence: Liberty Senior Living AI Search Case Study: https://www.searchinfluence.com/case-studies/senior-living-marketing/liberty-senior-living-ai-search-strategy/ Agency-reported sessions and conversion figures. A broader marketing program; not controlled evidence of AI-caused occupancy or admissions. - Arc Intermedia: Acts Retirement-Life Communities AI search case: https://www.arcintermedia.com/proof/case-studies/ai-search-increase-for-retirement-community-network/ Agency-reported prose claims of more than sevenfold AI traffic and 36% better conversion. Relative conversion improvement is not an absolute conversion rate. - Google Search Central: AI features and your website: https://developers.google.com/search/docs/appearance/ai-features Official explanation of related-query exploration, supporting links and normal search eligibility. It does not guarantee a provider recommendation. - National Institute on Aging: Assisted living and nursing homes: https://www.nia.nih.gov/health/assisted-living-and-nursing-homes U.S. government information distinguishing residential care settings. Used for category context, not AI performance. - Care Quality Commission: Find a care home: https://www.cqc.org.uk/care-services/find-care-home Official England-specific inspection and rating information. Its coverage should not be described as a worldwide quality standard. - CNSA: Résidence autonomie et résidence services, quelle différence ?: https://www.pour-les-personnes-agees.gouv.fr/changer-de-logement/vivre-dans-une-residence-avec-services-pour-seniors/residence-autonomie-et-residence-services-quelle-difference French official guidance updated January 30, 2026, distinguishing these housing categories, contracts and support arrangements. - My Aged Care: Non government-funded providers: https://www.myagedcare.gov.au/non-government-funded-providers Australian official explanation distinguishing retirement villages and independent living units from aged care facilities. - German Federal Ministry of Health: Questions about care financing: https://www.bundesgesundheitsministerium.de/pflegesichern/faq-pflegefinanzierung Official explanation of partial care insurance coverage and additional resident costs. No individual fee estimate is provided here. - Singapore Ministry of Health: Residential long-term care subsidies: https://www.moh.gov.sg/managing-expenses/schemes-and-subsidies/subsidies-for-residential-long-term-care-services/ Framework effective July 1, 2026. Eligibility, citizenship or residency, age and means testing affect support. The 80% figure is a conditional maximum. - My Aged Care: Aged care home costs and fees: https://www.myagedcare.gov.au/aged-care-home-costs-and-fees Official explanation of fee arrangements and protections. Personal circumstances and entry dates matter; this report is not individual financial advice. - HumanGood: The do’s and don’ts for using AI to research senior living: https://www.humangood.org/resources/senior-living-blog/the-dos-and-donts-for-using-ai-to-research-senior-living Provider-authored guidance dated August 18, 2026. Evidence of its public response to AI-assisted research, not a measured test of AI accuracy. - Searchable: Prompt Universe: https://www.searchable.com/prompt-universe Official product description of modeled buyer questions and representative tracking. Daavid works at Searchable; proposed onboarding and support are agreed with the provider. # AI knows your name. Does it recommend your communities? URL: https://daavidchrista.com/research/senior-living-who-owns-the-answer/ Author: Daavid Christa Published: 2026-11-02 One operator appeared in every question that named it, but in only four of 87 questions that didn’t. Here’s what I’d look at before trusting a visibility score. Ask AI about your own brand and you might like the answer. Ask it to help a family find suitable care, and the result can look very different. In this report I break down that gap, show which questions were missing, and explain how I’d use the findings to choose the next piece of work. ## First, look at what the score is counting The headline was 17 mentions across 100 generated ChatGPT questions for one Canadian retirement operator. Open the breakdown and you find that 13 of those mentions came from 13 questions that already named the operator. The other 87 questions produced four mentions. That means 76.5% of the mentions came from just 13% of the questions. So I’d show the two rates separately: 100% for questions naming the operator, and about 4.6% for questions without its name. One tells you how AI talks about a known option. The other tells you whether it brings that option into the conversation. Your strategy needs to care about both. This is one generated-question sample, checked on 3 November 2026. Its collection date wasn’t available. It doesn’t count real families, tell us whether each mention was positive, or show move-ins. It does give us a clear reason to open the questions behind the score. Question group | Questions | Mentions | Mention rate Operator named | 13 | 13 | 100% Operator not named | 87 | 4 | 4.6% Whole sample | 100 | 17 | 17% ## Why I care about the questions before your name Search Influence surveyed 300 families researching senior living or who had searched in the previous year. Its October 2026 summary says 57% seriously considered only two or three communities; 77% considered four or fewer. Those are the publisher’s survey results, rather than a national admissions benchmark. Imagine a family building that shortlist. If they haven’t heard of your community, they won’t start by asking for your name. They might ask about care near a relative, fees, or options for a couple. I’d want to know whether your business enters those answers when it’s a suitable option. The survey and our ChatGPT sample cover different groups and places, so we can’t turn them into a lost-resident estimate. The strategic point is simpler: check whether you’re being introduced while the family still has options to compare. ## Find the decisions where you’re missing In this sample, five topics produced no operator mentions: costs, choosing a community, transitions, memory care, and respite. Together they covered 45 questions. That gives us a much better research brief than “improve awareness.” Each topic helps a family make a different decision. The four discovery mentions came from amenities, independent living, and assisted living. I’d look at those answers too. What made the operator relevant there? Were the right services described? Which sources did the answer use? The overall 17% can’t tell you that. Before recommending new content, I’d check whether the operator actually belonged in the missing answers. Service fit, location, wording, source quality, and variation between answers can all matter. A zero in a small test is a reason to investigate, not a diagnosis on its own. Topic without mentions | Sampled questions | Operator mentions Costs | 17 | 0 Choosing a community | 12 | 0 Transitions | 8 | 0 Memory care | 6 | 0 Respite | 2 | 0 Combined | 45 | 0 ## Start with costs if families can’t get a clear answer The same family survey names pricing transparency as the most important comparison factor and the hardest information to find online. It says 53% use AI to compare costs, care levels, or amenities, and 46% use it to narrow a shortlist. That puts a practical question on the table: can families understand your fees before they call? The sample’s 17 cost questions produced no mentions. I’d review the actual questions, then look at what your pages explain: accommodation, included care, likely extras, assessment, and how to get a current quote. You may not be able to publish one final price for everyone. You can still explain how the price works. This doesn’t prove a missing fee page caused the result. It tells us that costs are important enough to check separately. I’d rather give your team one specific pricing question to answer well than a vague instruction to publish more AI content. ## Check brand and discovery inside each topic too There’s another trap in the data. The brand-services topic has 13 questions and 11 mentions. Those aren’t the same 13 questions as the brand-name group. Within that topic, 11 questions named the operator and all 11 mentioned it. The other two didn’t name it and produced no mentions. Assisted living shows the same issue. It had three mentions across 23 questions. Two came from the two questions naming the operator; one came from the other 21. If I only showed the topic total, you’d miss what was driving it. I’d keep two labels on each tracked question: what decision it’s about, and whether it names your business. Then you can ask a useful question about a service: are we being introduced here, or only discussed when someone already knows us? ## Don’t turn question groups into a customer funnel The report also groups questions by intent. Research produced two mentions from 24 questions, comparison ten from 49, and buying five from 27. That works out to about 8.3%, 20.4%, and 18.5%. Those aren’t the same people moving from research to purchase. They’re separate generated questions. We also don’t have the brand-name breakdown within each intent group. So the higher comparison rate doesn’t prove families become more likely to choose this operator as they progress. Use these groups to check the answers available at different decisions. To understand how people move towards a visit or move-in, follow actual inquiries in your admissions records. Keep the two views connected, with their labels clear. ## Use other providers’ results to ask better questions Search Influence’s Liberty case reports 98.7% average presence for questions about a named community’s services, and 90.5% for a city-based assisted-living question among communities offering memory care. Those are the agency’s reported results. It hasn’t published the full underlying set of answers. I wouldn’t put those figures beside our operator’s 4.6% and call it a ranking. The questions, locations, eligible services, dates, engines, and groups counted differ. Ask what was measured before deciding what another provider’s number means for you. What I would borrow is the focus on a specific question and a suitable community. Choose the decisions you want to compete for, keep the test consistent, and compare your own results over time. That makes a case study useful without turning it into your forecast. ## Now open the sources behind the answers Being mentioned and being cited are different. AI can name your operator while a directory explains the service. It can also cite your page without recommending your community. Searchable separates brand mentions from the source links so you can inspect both. I’d check which pages are shaping the family’s understanding of costs, care, and quality. Your own community page may explain the current service; an official record may support a rating; a comparison page may frame the alternatives. These pages have different jobs. Our sample gives us mention counts, not a complete source map, so I can’t name which rival or directory owns the missing answers. That would be the next piece of research. Check where you’re absent, where the facts are wrong, and where someone else is explaining your offer. ## Here’s how I’d use this with your team Pick a community and pull up the questions behind its score. Split brand from discovery, then break discovery down by the decisions families need to make. Start with the questions where your community is a fit and the answer would change whether a family considers you. Save the answers and their sources, rather than the score alone. Keep the topic, location, engine, date, and brand-name label with each one. Our 100-question snapshot shows why: change the mix and you can change the headline without improving discovery. Then choose an action your team can check. Clarify a fee explanation, correct a community fact, or answer a comparison question. Revisit the same question set after the change has had time to be found. The family survey and provider cases give context; your own answers tell you where to work. I can help you build that question set in Searchable and turn it into a short action plan. Once you can see the discovery gap, the next job is measurement: make sure a better-looking chart means the answers actually improved. ## Sources - Anonymized senior living visibility dataset and lesson: https://daavidchrista.com/research/senior-living-case-study-evidence/ Supplied Searchable report independently recounted for this site on November 3, 2026: one Canadian retirement operator, 100 generated ChatGPT questions, 17 mentions. Collection date and window unavailable. Derived percentages and cross-group analysis are ours; this is not a market-share study. - Search Influence: How Families Search for Senior Living in the Age of AI: https://www.searchinfluence.com/blog/research-senior-living-search-behavior-ai/ Published October 23, 2026; reviewed November 3, 2026. Publisher survey of 300 families currently searching or having searched in the prior 12 months. Self-reported behavior. The public summary does not provide enough sampling detail to establish population representativeness. - Search Influence: Liberty Senior Living AI Search Case Study: https://www.searchinfluence.com/case-studies/senior-living-marketing/liberty-senior-living-ai-search-strategy/ Agency-reported client results reviewed November 3, 2026. Presence rates refer to specified questions and eligible communities, not a matched comparison with our anonymized sample. - Searchable: AI Mentions & Citations Analytics: https://www.searchable.com/features/aeo-insights/mentions-citations Official product description reviewed November 3, 2026. Supports the distinction between brand appearances and source citations. The competitive interpretation is our analysis; I work at Searchable. # What would one extra move-in be worth? URL: https://daavidchrista.com/research/senior-living-answer-to-move-in/ Author: Daavid Christa Published: 2026-11-02 Let’s work through the fees, capacity, timing, and costs behind an AI search business case, using real market data and a clearly labelled example. I’d start the business case with your communities, rather than a visibility percentage. Where do you have staffed space? Which families can you serve? What would another resident contribute, and when? The examples below show how those answers can change the result. ## Start with the communities that need the demand NIC reported 90.4% occupancy across 31 primary US senior-housing markets in Q3 2026. Independent living was at 91.7% and assisted living at 89.1%. Its October release also said annual inventory growth had stayed below 1% for six quarters. That gives us useful market context: demand was being absorbed faster than new units were arriving. Now bring that back to your business. An additional suitable inquiry has a different value at a nearly full building and a community with staffed vacancies. I’d choose where to invest by looking at available places, care fit, and the type of demand each community needs. Even the market averages differ: NIC reported 93.3% for San Francisco and 88.7% for Dallas. Neither number tells you which rooms your team can sell. Use the wider data to understand the market, then build the decision around your own communities. ## Compare the same communities over time Brookdale’s Q2 2026 results show why this matters. Consolidated resident fees fell 8.7% to $708.5 million, while same-community resident fees rose 5.5% to $688.7 million. Brookdale linked the overall fall mainly to community dispositions. Its same-community comparison covered 515 communities present in both periods. That same-community group had 82.9% weighted average occupancy and $6,714 monthly revenue per occupied unit. These are business results, not an AI search case study. I’m using them because the comparison changes the story. If your portfolio buys, sells, or closes communities during a campaign, the total can move for reasons that have little to do with marketing. I’d keep a stable group of communities for the review and show pricing, care mix, and capacity changes beside the results. Brookdale Q2 2026 measure | Reported result | Meaning for this analysis Consolidated resident fees | −8.7% year over year | Portfolio changes affect the total Same-community resident fees | +5.5% year over year | Comparable buildings tell a different story Same-community occupancy | 82.9% | Revenue still depends on occupied capacity ## Use your actual fees and margins CareScout’s 2025 survey put the US assisted-living median at $6,200 a month, or $74,400 a year. It reported nursing-home costs per day: $315 for a semi-private room and $355 for a private room. These describe different settings and units. None is a sensible default value for every lead. Use the expected revenue for your actual room and service. Check included care, extra support, concessions, payer arrangements, and one-time charges with finance. Then ask when a resident would arrive and how long they’re expected to stay. Revenue, occupied months, and contribution after costs answer different questions. I’d show all three where you have the data. A big annual fee looks attractive, but it can hide timing and costs that change whether the investment makes sense. ## Follow the path from answer to inquiry Google’s Generative AI performance report gives site owners a view of impressions from AI Overviews and AI Mode. Google says it rolled out worldwide on 31 August 2026. You can look by page, country, date, and device, though sites without enough impressions may not see it. Its impressions also sit inside Web reporting. That helps show exposure to links. Next come visits, inquiries, suitable prospects, tours, and move-ins. I’d look for a record at each stage instead of assuming that more impressions explain the final admissions number. Track identifiable AI referrals and ask families how they found you. Keep their answer as a separate field. A recorded website source and a family saying they used AI tell you different things, and either can miss earlier research on another device. ## Work out how many occupied months you can add A practical unit is one resident in an available place for one month. If your calculation predicts three move-ins but you have two staffed places, start with two. If they arrive partway through the year, count the months after arrival. Use the same care with conversion rates. This month’s move-ins may come from inquiries made months ago. Follow a group of inquiries from first contact through tours and admissions, and keep the unresolved cases visible. Dividing unrelated monthly totals won’t tell you their conversion rate. I use this way of thinking to connect marketing activity with something finance and admissions can recognise. It isn’t an industry conversion benchmark. Your records need to supply the rates, timing, and capacity. ## Run a scenario with the limits included Here’s a fictional example. Start with 30 additional qualified inquiries. Assume 40% complete a tour and 25% of those tours lead to admission. That gives three expected move-ins. Now apply the operating limit: there are only two staffed places. Both residents arrive at the start of month three and stay through month twelve, giving 20 occupied resident-months. At an assumed $6,000 a month, that is $120,000 revenue. Assume 35% contribution after variable costs and you have $42,000 before the extra program cost. Subtract an assumed $12,000 for the program and $30,000 remains. These are example inputs to show the calculation. Replace them with your numbers. Scenario step | Illustrative result 30 × 40% × 25% | 3 expected admissions Staffed capacity constraint | 2 actual places in the scenario 2 residents × 10 months × $6,000 | $120,000 revenue $120,000 × 35% − $12,000 | $30,000 after modeled program costs ## Ask whether those residents would have arrived anyway Change one assumption: one of the two residents would have arrived through another route anyway. Now only ten resident-months are additional. At $6,000 a month, that’s $60,000 extra revenue. Apply the same 35% contribution and $12,000 program cost and $9,000 remains. Same two visible move-ins, very different business case. This is why I’d ask two questions in a review: which interactions got credit, and what business did the work actually add? A source field in your CRM helps with the first. To answer the second, you need a credible comparison with what would have happened otherwise. There can also be value in avoiding a paid referral fee when a family comes directly. Check the contract and the invoice before counting a saving. If it’s the same resident, don’t count a whole additional admission and the saved fee as though they came from two different wins. ## Choose an investment your team can explain I’d invest where better discovery or clearer information can solve a specific problem: suitable families aren’t finding a community, or inquiries arrive with the wrong understanding of care and costs. Then connect that problem to staffed space, admissions, timing, and contribution. Searchable’s LLM Analytics can help with AI referrals, landing pages, and crawler activity, depending on your setup. Your admissions and finance records still supply the business outcome. Put those views beside each other rather than turning one mention score into a revenue estimate. Bring one community’s figures to a conversation with me and we can work through the assumptions. Once you’ve got a business case you can defend, look at paid AI search too: where it’s available, where it fits your offer, and how you’d measure a test. ## Where the numbers come from The market, operator, cost, and product sources were checked on 3 November 2026. NIC and CareScout cover US markets; Brookdale is one operator. They don’t measure revenue caused by AI search. The two financial scenarios are fictional examples showing how capacity, arrival dates, and genuinely additional business change the answer. ## Sources - NIC: Q3 2026 senior housing occupancy: https://www.nic.org/blog/senior-housing-occupancy-nears-record-high-as-primary-markets-surpass-90/ Published October 9, 2026. Covers 31 primary US markets. Market occupancy and supply figures are not an AI-search outcome. - Brookdale: Second Quarter 2026 Results: https://brookdaleinvestors.com/news/news-details/2026/Brookdale-Announces-Second-Quarter-2026-Results/default.aspx Published August 10, 2026, for the quarter ended June 30. Same-community and consolidated comparisons have different populations. No AI attribution is claimed. - CareScout: 2025 Cost of Care Survey results: https://investor.genworth.com/news-events/press-releases/detail/1054/carescout-releases-2025-cost-of-care-survey-results Published March 2026. US provider-survey medians, not prices for a particular community or conversion-model defaults. - Google Search Console: Generative AI performance report: https://support.google.com/webmasters/answer/16984139 Official guidance reviewed November 3, 2026. Worldwide rollout August 31, 2026. Reports impressions for supported Google Search AI features, not admissions. - Searchable: LLM Analytics: https://www.searchable.com/features/llm-analytics Vendor feature description reviewed November 3, 2026. Implementation depends on account setup and connected data. The author works at Searchable. # What are senior living providers seeing from AI search? URL: https://daavidchrista.com/research/senior-living-case-study-evidence/ Author: Daavid Christa Published: 2026-11-02 Liberty and Acts report growing AI traffic and stronger website conversion. Here’s what I’d learn from their work, and what I’d check before expecting the same result. The useful signal in these cases is the quality of the visitors who reached the website. The useful lesson is the breadth of the work: community content, technical fixes, listings, and outside sources. I’d use the cases to shape a test for your business, then measure your own discovery, inquiries, and move-ins. ## Start with the visitors who arrive Search Influence reports that Liberty Senior Living’s AI-sourced visits grew 487% over twelve months. It reports a 17.3% conversion rate from those visits, against 11.3% across the site. The program included community content, FAQs, technical work, outside material, advertising, and email. The case also reports occupancy above 92%, without separating how much AI contributed. The conversion comparison is the part I’d want to understand first. What counted as a conversion? Were these suitable inquiries? Did they become tours? People arriving after an AI conversation may have already done some research, but this public case doesn’t prove that explains the difference. A small channel can be worth attention if the people arriving are a good fit. I’d look at inquiry quality alongside traffic volume. Keep the website conversion rate labelled as such; admissions records are needed to tell us what happened after that. ## Look at the work behind the Acts result Arc Intermedia reports more than sevenfold year-over-year AI traffic growth for Acts Retirement-Life Communities and says AI visitors converted 36% better than average. Its account covers content, site structure, internal links, business listings, outside material, citation monitoring, and branded participation on Reddit and Quora. It doesn’t publish an absolute admissions outcome. That gives you a more useful picture of the work than a claim that one new file or page format fixes AI search. The community page is one part of it. So are the facts a directory repeats, the questions other publishers answer, and whether those sources still describe the business correctly. I’d inspect the sources in your own answers before copying their channel mix. Reddit may matter for one set of questions and barely feature in another. The case describes a combined program; it doesn’t isolate which individual action produced the gain. ## Read the percentages before borrowing them Liberty’s 487% growth means the ending traffic amount was 5.87 times the starting amount. The move from 11.3% to 17.3% conversion is six percentage points, or about 53% higher relatively. Acts’ 36% improvement is relative too. For illustration, an 8% rate improved by 36% becomes 10.88%; that starting 8% isn’t an Acts figure. When you put a case study in front of your team, name the event beside the number. Visits, website conversions, tours, and occupancy describe different things. I’d rather have a modest result everyone understands than a big percentage that falls apart in the next finance meeting. Evidence in the public cases | What it supports | What remains unestablished AI traffic growth | A growing measured referral segment | The absolute size of incremental demand Above-average recorded conversion | A potentially valuable visitor segment | A universal tour or move-in rate A broad implementation program | Several credible areas of work | One isolated causal tactic ## Then check whether your own brand is being discovered Our separate anonymised Canadian report contains 100 generated ChatGPT questions. The operator appears in 17 answers: 13 out of 13 questions naming it, and four out of 87 questions without its name. Opening the question rows changes how you read the headline. A family who already knows your name needs accurate information about you. A family finding options needs AI to introduce suitable communities. I’d track those separately, even if both sit inside the same reporting tool. Excellent brand recognition can hide a weak position earlier in the decision. We checked this report on 3 November 2026, and its original collection dates weren’t available. The questions were generated, and some without the operator’s name could still name a competitor. It’s a useful worked example of a measurement problem, rather than a measure of Canadian family demand. Question group | Mentions / questions | Rate Names the operator | 13 / 13 | 100% Does not name the operator | 4 / 87 | 4.6%, rounded All generated questions | 17 / 100 | 17% ## Give the missing questions a practical owner The sample has no operator mentions in 17 cost questions or 12 community-selection questions. Assisted living produced three mentions from 23 questions. Keep topic and brand intent separate: a cost question can either name the operator or leave it out. My next step would be to open those cost and selection answers with your marketing and admissions teams. Was the operator suitable? Which pages were cited? What information would a family need to compare it fairly? The counts alone don’t identify the cause. Then brief a specific improvement. Explain the room, included support, extra charges, assessment conditions, or how to get a current quote. Check what’s actually missing before writing. Publishing the page is a testable action, rather than a guarantee that AI will start mentioning it. ## Make the next conversation easier HumanGood’s August 2026 guidance tells families they can use AI to narrow options and compare care, then confirm current prices, availability, and suitability with community teams and visits. That’s the provider’s advice, not an accuracy study. There’s a good sales lesson here. Help someone arrive with better questions and a realistic understanding of the offer. Your page needs to work for the person reading it as well as the system finding it. Be clear about what you can provide and what needs an individual conversation. I’d rather see more suitable inquiries than more mentions for care the home doesn’t offer. Track those expectations with admissions. If AI sends a family to the wrong service, a positive-sounding recommendation still creates work for everyone. ## Get the basics right before chasing tricks Google says AI Overviews and AI Mode don’t need special AI markup. Supporting pages need normal indexing and snippet eligibility, and appearing is never guaranteed. So check access and indexing, then focus on the information the page provides. “We provide excellent care” tells a family very little. Can a couple stay together? What assessment is needed? What’s included in the fee? Those answers are worth working on. They also need to be available on pages a search system can actually reach. I’d split the review into two questions: can the relevant system read this page, and does it answer the family’s question well? The public cases describe work on both. They don’t prove that every provider needs the same template or will get the same lift. ## Borrow the process from other industries Searchable’s 303 story reports more than £1 million in pipeline for a creative agency, alongside a move from manual checks to repeatable visibility and content work. Pipeline is potential business, rather than collected revenue. And a creative agency isn’t a senior living operator. The part I’d borrow is the process: save the answer, agree what needs improving, publish the change, and check what happens. I work at Searchable, so that connection should be clear. Another industry’s result is a useful example of the workflow, not a return I’d put into your admissions forecast. ## Use the cases to design your first test Choose one community, a set of relevant family questions, and a business reason to improve the answers. Keep the starting answers and their sources. Agree who will fix the gaps and how you’ll check both visibility and suitable inquiries. That gives your team a clear test instead of a collection of tactics. I can help set that up in Searchable and work through the findings with you. For the next step, let’s look at the economics. Traffic and conversion are useful signals, but the investment decision needs your capacity, fees, and actual admissions process. ## How I’ve used these sources These public cases, official guidance, and the anonymised report were checked on 3 November 2026. The agencies’ results are attributed to them and haven’t been independently audited here. We don’t have a shared control group or complete move-in records. The calculations explain published percentages; they aren’t additional customer results. ## Sources - Search Influence: Liberty Senior Living AI Search Case Study: https://www.searchinfluence.com/case-studies/senior-living-marketing/liberty-senior-living-ai-search-strategy/ Participating agency’s report, reviewed November 3, 2026. Traffic and conversion measures do not independently establish AI-caused move-ins. - Arc Intermedia: Acts Retirement-Life Communities AI search case: https://www.arcintermedia.com/proof/case-studies/ai-search-increase-for-retirement-community-network/ Uses the narrative claims of more than sevenfold AI traffic and 36% better conversion than average. Animated counters were not used. Reviewed November 3, 2026. - HumanGood: The Do’s and Don’ts for Using AI to Research Senior Living: https://www.humangood.org/resources/senior-living-blog/the-dos-and-donts-for-using-ai-to-research-senior-living Provider guidance published August 18, 2026. Evidence of its stated approach to family research, not a controlled accuracy study. - Google Search Central: AI features and your website: https://developers.google.com/search/docs/appearance/ai-features Official eligibility guidance reviewed November 3, 2026. The application to family decisions is our analysis. - Searchable: 303 customer story: https://www.searchable.com/customers/303 Vendor-published creative-agency case, reviewed November 3, 2026. Pipeline is not recognized revenue or senior living admissions. The author works at Searchable. # What I’d do in your first 90 days of AI search URL: https://daavidchrista.com/research/senior-living-first-90-days/ Author: Daavid Christa Published: 2026-11-02 A quarter is enough to build a useful baseline, fix important gaps, and see what changes. Here’s how I’d set the work up without promising someone else’s results. I’d use the first quarter to answer three questions: are the right communities being discovered, are the facts right, and is the work helping suitable families get in touch? Start with a stable set of questions, ship the important changes early, and review the answers alongside admissions data. ## First, be clear about what a quarter can show Search Influence reports that The Barclay at SouthPark in Charlotte increased AI visibility from 59.8% to 75.0% over twelve weeks in May–July 2026. That’s a 15.2-percentage-point increase. Having the starting point, endpoint, and period makes this a useful case to discuss. The agency hasn’t published the full repeated question set, all the answers, or a matched community that didn’t receive the work. So we can see the reported change during the program, without isolating the effect of each tactic. I’d take that as a reason to run a serious quarterly test. I wouldn’t use it as your target. Your starting position, local competitors, care services, and ability to publish changes will shape what the quarter can deliver. ## Agree what success means before you start A visibility percentage counts appearances in the answers being measured. It doesn’t count families, qualified inquiries, tours, or occupied apartments. Before the work starts, I’d agree which of those your team can measure and which decision each number will help you make. Give the report a simple test: could someone repeat the headline accurately after reading it? “Visibility rose during these twelve weeks” fits the Barclay evidence. Crediting the program with all the resulting occupancy goes further than the published data allows. My first deliverable would be a saved starting point: the questions, locations, engines, answers, and sources. Add the business context, including care services and available space. Then you have something concrete to compare after the changes. ## Keep quarterly and annual results separate Liberty Senior Living’s wider case reports 487% growth in AI-driven visits over twelve months. That’s a different measure and a longer period than the Barclay visibility result. Use it as annual context, rather than calling it a 90-day outcome. Acts Retirement-Life Communities has a separate agency-reported result: more than sevenfold year-over-year AI traffic growth, with conversion 36% better than average. Again, those are annual traffic and relative conversion figures, not quarterly move-in results. I’d label the time period on every chart in your review. Platform adoption, other campaigns, and changes to the business can all affect an annual result. Clear dates make it easier to judge what a case study can actually tell us about your quarter. ## Spend the quarter doing the work, not only reporting it Searchable’s Blackbird story describes 90-day optimisation sprints and a reported 30% average visibility improvement across its client base. It also reports 170 hours saved in 30 days. This is a vendor-published agency story, with no senior living subgroup identified. It’s useful as an example of how a team organises delivery. It doesn’t give us a reconstructable average for your market or a 30-point target. Thirty percent improvement and thirty percentage points mean different things. The practical lesson I’d use is speed between finding a gap and publishing a good correction. In the first weeks, choose work your team can ship: a fee explanation, an important community fact, an access issue. A dashboard that keeps identifying the same problem while nothing changes won’t teach you much. ## Publish early enough to learn something Google says recrawling can take a few days to a few weeks. Requesting a crawl doesn’t guarantee immediate inclusion, and repeatedly requesting one doesn’t speed it up. That guidance is about Google, rather than a timetable for every AI product. This is why I’d aim to publish the important changes early in the quarter. A page corrected in week two has had more time to be found than a page corrected in week twelve. Record the publication date so the review reflects that difference. Then separate three milestones: the page changed, a search system found it, and a relevant answer used it. Those may happen at different times. Keep checking the same questions, and don’t call a change unsuccessful just because the first answer after publication looks the same. ## Give admissions time to catch up WelcomeHome’s 2025 review says half the prospects in its completed-move-in analysis moved in within 25 days of their tour. The slowest quarter took more than two months. The analysis covers people who actually moved in during 2025 and completed activities. It doesn’t describe every inquiry or isolate AI referrals. That means 25 days isn’t a promise you can apply to a new lead. People who never moved in aren’t represented in that timing figure. I’d use your own inquiry history to decide when a new group has had a fair chance to progress. An inquiry arriving late in the quarter may still be useful even if it hasn’t become a move-in by the review. Equally, this quarter’s move-ins may have begun their search before the AI work started. Keep those groups clear when you report. ## Review the work on three timelines I’d give the review three timelines: what the team published, when it started appearing in search and answers, and what happened to the families who got in touch. Putting those beside each other helps explain why a project can be progressing while the revenue picture is still incomplete. The table below is a way to structure that review, not an industry benchmark. Use it to ask what changed and what evidence you have for the next stage. Don’t make the visibility chart answer a question that belongs in admissions records. For a group, keep the community context visible too. A new opening and a near-full home may need different kinds of demand. The same improvement in AI visibility won’t necessarily have the same commercial value at both. Clock | Evidence that belongs to it | What it cannot establish alone Implementation | A verified change is live on a relevant page | That an answer engine used it Search exposure | A comparable answer or source changed | That a family contacted the provider Family decision | An inquiry progresses to a tour or move-in | That AI search caused the decision ## Check whether the improvement is really discovery Our anonymised Canadian snapshot shows 13 mentions from 13 questions naming the operator, versus four from 87 questions without its name. It’s a single set of 100 generated ChatGPT questions, with no original collection date shown. It illustrates a split worth checking in your own baseline. If your quarterly score rises because you added more questions about your own name, you haven’t necessarily improved how families discover you. Keep a stable discovery set, and label any questions you add during the program. That way, new coverage doesn’t masquerade as improved performance. By the end of the quarter, I’d want a much clearer answer to where you’re missing, which facts need fixing, and which changes were followed by better answers. Even where the admissions outcome needs more time, those findings should guide what you do next. ## Leave the quarter with a decision At the review, choose what to keep doing, what to change, and where to expand. Bring the starting answers, shipped changes, repeat checks, and any suitable inquiries you can follow. A useful first quarter should leave the team better informed, rather than waiting for an arbitrary percentage to turn green. I’d prioritise the next round around relevance and capacity. Expand a question set that reveals useful gaps. Fix a recurring error. Stop spending effort on questions your communities can’t serve. Make each decision traceable to the evidence you’ve collected. The cases and documentation here were checked on 3 November 2026. I work at Searchable, and I can help build the question set, get your team onboarded, and run that review. Before you choose your first changes, read the community-facts guide next. It shows how small errors about care, ratings, or fees can change the whole answer. ## Sources - Search Influence: Liberty Senior Living AI Search Case Study: https://www.searchinfluence.com/case-studies/senior-living-marketing/liberty-senior-living-ai-search-strategy/ Reviewed November 3, 2026. Publisher image for The Barclay at SouthPark visually verified: 59.8% to 75.0% AI visibility, May–July 2026, 12 weeks. The 15.2-percentage-point difference is our calculation. Portfolio session growth has a separate 12-month window. Agency-reported results; no controlled comparison or raw answer panel published. - Arc Intermedia: Acts Retirement-Life Communities AI Search Case Study: https://www.arcintermedia.com/proof/case-studies/ai-search-increase-for-retirement-community-network/ Reviewed November 3, 2026. Uses the narrative’s more-than-sevenfold year-over-year traffic claim and 36% higher conversion comparison. No inference of quarterly performance, move-in conversion, or isolated causality. - Searchable: Blackbird Customer Story: https://www.searchable.com/customers/blackbird Reviewed November 3, 2026. Vendor-published account links 90-day sprints to reported visibility improvement. Not a senior living benchmark; underlying client-level data is not published. I work at Searchable. - Google Search Central: Ask Google to recrawl your URLs: https://developers.google.com/search/docs/crawling-indexing/ask-google-to-recrawl Official guidance reviewed November 3, 2026. Crawling may take days to weeks and is not guaranteed by a request. This guidance does not specify time to an AI recommendation. - WelcomeHome: Senior Care Insights, 2025 Year in Review—public sample: https://info.welcomehomesoftware.com/hubfs/20961659/Senior%20Care%20Insights%202025%20Year%20in%20Review%20Sample.pdf?hsLang=en Printed page 13, reviewed November 3, 2026. Sales timing describes successful completed activities for prospects who moved in during 2025. It is not an AI-channel study or the conversion probability for all inquiries. - Anonymized senior living visibility dataset and lesson: https://daavidchrista.com/research/senior-living-case-study-evidence/ Supplied Searchable report recounted November 3, 2026: 100 generated ChatGPT questions for one Canadian operator. Brand-named questions: 13 mentions in 13; other questions: four in 87. Collection date and window unavailable. No longitudinal outcome inferred. # What is AI getting wrong about your communities? URL: https://daavidchrista.com/research/senior-living-community-facts/ Author: Daavid Christa Published: 2026-11-02 The name can be right while the care, price, or location is wrong. Here’s how I’d check the details that matter to families. Start with one community and read the answer as if you were a family choosing care. Does it describe the right place, service, price, and conditions? Official guidance across five markets shows why those details need checking, even when the answer has a source link and sounds confident. ## Start with the promise the answer makes Your operator’s name appears. Great. Now read what comes after it. Does the answer promise nursing support at a home that doesn’t provide it? Quote an old fee? Use a rating from another location? A brand mention on its own won’t catch those problems. The official sources below give us concrete checks: an archived CQC record doesn’t always mean a home closed; Ontario distinguishes retirement homes from long-term care; Australia separates retirement villages from aged care; US ratings have specific components; and France’s EHPAD comparison price has a defined basis. I’d use these examples to build a review with your community teams. This isn’t a measured AI error rate or an accusation against any provider. It’s a way to spot which details would change whether a family sees you as a suitable option. ## In England, check what “archived” actually means CQC’s data guidance, updated on 2 November 2026, covers active and inactive providers and locations, registration dates, former providers, services, and report dates. It says a record can be archived after re-registration, a legal-structure change, or an address change. Archiving doesn’t always mean the service closed. Some directory-file updates can also be delayed. So if AI says a home has closed, I’d check the current location record and its history before accepting the conclusion. The source link might be genuine while the interpretation is wrong. A continuing service can appear under more than one registration record over time. Make the relationship clear on your community page: the current name and location, relevant official record, and any change a family is likely to encounter. A group-level “about us” page won’t answer every question about an individual home. ## In Ontario, explain the care the home actually offers RHRA’s public register lists licensed retirement homes, services, inspection findings, and enforcement summaries. Its guidance distinguishes retirement homes from long-term care homes. Residents pay for accommodation and purchased care, and the services depend on the home. Some care may come from outside providers. That leaves real questions to answer. Is support available overnight? What is included? Who provides additional care, and how is it arranged? “Retirement home” gives a family a category, but it doesn’t settle whether that particular home can meet their needs. I’d ask each community team to review the local service description, especially where a group-wide page is reused. A capability somewhere in the group shouldn’t accidentally read as a promise at every home. We haven’t identified that error at a named operator here; it’s a check I’d build into the process. ## In Australia, keep retirement living and aged care clear My Aged Care says retirement villages and independent living units are not aged care facilities. They serve people needing less care than aged care homes and sit under state and territory regulation, rather than the Australian Government’s aged-care arrangements. That difference matters even when a family searches broadly for senior living. If you operate internationally, I’d review the category language with someone local before translating the page. A family looking for an independent retirement setting needs a different explanation from a family arranging ongoing residential care. The question set should reflect that difference too. Clear limits help people decide whether to contact you. I’d rather explain the service accurately than attract inquiries the team has to redirect. That’s the intended benefit of the change; the official guidance doesn’t measure a conversion improvement from it. ## In the US, ask what the stars refer to CMS describes an overall nursing-home rating and separate ratings for health inspections, staffing, and quality measures. It also says a rating system can’t cover every part of choosing a home, such as specialty care or how easily relatives can visit. The guidance was last modified on 27 October 2026. If an answer calls your community “five-star,” I’d open the source. Which rating system? Overall or one component? Which facility? A review-site score and an official nursing-home rating are different claims, even when both use stars. On your own page, show the system, location, measure, and date beside the rating. That gives a family something they can check. It also reduces the chance of a group’s flattering headline being copied into a claim about the wrong property. ## In France, explain what the comparison price buys France’s official EHPAD comparison guidance defines its monthly starting figure as 30 days of basic accommodation plus 30 days at the lowest dependency tariff, GIR 5–6. The displayed calculation excludes public assistance. The explanation, updated on 14 August 2026, also lets readers compare included services. A price can be copied perfectly and still mislead if the answer drops what’s included. I’d check the room basis, dependency assumption, extras, and treatment of assistance. The number needs its explanation beside it. You can apply that habit elsewhere without copying French rules into another market. Separate a starting accommodation fee, an assessed care package, and a household’s final cost. Explain how to get a current quote rather than making a general answer sound personalised. ## Prioritise the errors that change the decision Now look across the checks. Some facts describe the operator, some the individual location, and others the service, rating, or fee package. I’d review each claim at the level it actually describes. A corporate source can be useful without proving every local detail. Give the most attention to errors about care fit, costs, and admission conditions. A correct postcode won’t cancel out a false promise about overnight support. If you use an accuracy score, keep the important errors visible beside it and explain how they affect the family’s decision. Evidence reviewed | Important distinction | Consequence if flattened CQC, England | Archived registration versus closed service | A continuing home could be excluded RHRA, Ontario | Retirement-home setting versus purchased care | Support could be overstated My Aged Care, Australia | Retirement village versus aged care facility | The wrong service category could be suggested CMS, United States | Overall rating versus component or other score | Quality evidence could be misapplied CNSA, France | Defined comparison package versus final cost | A family could misunderstand affordability ## Fix the page that should answer the question Google advises making important content available as text and keeping structured data consistent with what people can see. It doesn’t require special AI schema. That gives your website team a useful foundation, though readable content still needs to be accurate. I’d trace a wrong claim back to the page used in the answer. Is your local page vague? Is a directory carrying an old operator name? Does the price page omit the conditions? These are possible causes to investigate, rather than errors we’ve measured at a particular provider. Save the exact claim, source passage, right community, and review date. Then get the relevant team to approve the correction. You’ll have a much clearer brief than “improve our AI reputation,” and a specific answer to check again later. ## Make accuracy part of the strategy For me, useful discovery includes being described correctly. If the answer attracts families who need care you can’t provide, a higher mention rate isn’t helping the team as much as it seems. Check fit and expectations alongside presence. Searchable can help you inspect brand mentions and the sources used in answers. Your care and commercial teams confirm the underlying facts. Put them in the same review: what did the answer say, what’s correct today, and which page needs work? I can help turn recurring issues into tracked questions and a short improvement plan. Then use the first-90-days guide to organise the work: publish the important corrections early and give the answers time to change before the review. ## What this review covers The official care and search guidance was checked on 3 November 2026. These country examples help explain possible AI-answer mistakes; they aren’t an audit of named providers or a clinical assessment. Check the current individual record and local guidance before changing a care, pricing, or eligibility claim. ## Sources - CQC: Using CQC data: https://www.cqc.org.uk/about-us/transparency/using-cqc-data England. Updated November 2, 2026; reviewed November 3. Describes provider/location identity, archived records, and current directory-update delays. - RHRA: Frequently asked questions: https://www.rhra.ca/en/faq/ Ontario retirement-home regulator. Reviewed November 3, 2026. Describes register scope, the distinction from long-term care, and locally offered or external care services. - My Aged Care: Non government-funded providers: https://www.myagedcare.gov.au/non-government-funded-providers Official Australian guidance reviewed November 3, 2026. Distinguishes retirement villages and independent living units from aged care facilities. - CMS: Five-Star Quality Rating System: https://www.cms.gov/medicare/health-safety-standards/certification-compliance/five-star-quality-rating-system US nursing-home framework. Page modified October 27, 2026; reviewed November 3. Ratings have separate components and do not cover every suitability consideration. - CNSA: Understanding the EHPAD price comparison: https://www.pour-les-personnes-agees.gouv.fr/comprendre-le-comparateur-de-prix-des-ehpad Official French explanation updated August 14, 2026. Defines the 30-day accommodation and GIR 5–6 basis before public assistance. - Google Search Central: AI features and your website: https://developers.google.com/search/docs/appearance/ai-features Official guidance reviewed November 3, 2026. Supports visible text and consistent structured data; the senior living interpretation is our analysis. - Searchable: AEO Insights: https://www.searchable.com/features/aeo-insights Vendor feature description reviewed November 3, 2026. Available evidence depends on account setup. The author works at Searchable. # How I’d get your business into AI answers URL: https://daavidchrista.com/research/geo-implementation-playbook/ Author: Daavid Christa Published: 2026-11-03 Start with what your buyers ask, check the answers they get, and fix the gaps that matter. Here’s the research behind that approach. Your buyers can learn about you, compare you, and rule you out before they reach your website. I’d build an AI search strategy around those decisions: the questions, the sources used to answer them, and the next step a suitable buyer can take. The research helps explain where to focus and what to measure. ## Start before someone searches for your product OpenAI’s September 2025 study of 1.5 million consumer conversations grouped 49% of messages as asking, 40% as doing, and 11% as expressing. Around 70% of consumer usage was outside work. That doesn’t make half the messages shopping searches. It shows how much people use the product to ask for information and advice. Think about what your buyer needs to understand before choosing a supplier. What problem do they have? Which approaches are worth comparing? What could go wrong? I’d start there, then move towards pricing, alternatives, and your specific offer. If all your content starts with your company name, you may be arriving late to the conversation. ## Read the answer, even when it doesn’t send a click Pew’s study of browsing from 900 US adults found traditional-result clicks in 8% of Google visits with AI summaries, compared with 15% without. Source links in the summaries were clicked in 1%. The browsing came from March 2025 and the results were reconstructed in April, so this is a dated observational study. For your strategy, I’d check what the answer teaches someone before worrying only about the link. Does it describe the problem fairly? Is your product relevant? Are your limitations explained? Website traffic still matters, but it won’t show everything a buyer learned before getting there. ## Look at the quality of the visitors who do arrive Adobe’s 29 October 2026 retail report says AI-referred visits in September converted to purchases 90% better than non-AI visits and generated 83% higher revenue per visit. AI referral traffic was up 71% year over year. The analysis covered more than one trillion US retail visits, 100 million products, and 18 categories. Those are relative comparisons among recorded visits, not a 90% conversion rate or a forecast for your business. Still, they give you a reason to ask whether the people arriving from AI have already narrowed their choices. Check inquiry quality and purchase behaviour as well as visit count. A smaller stream can deserve attention. ## Decide which business problem you’re solving An AI answer, a website visit, and a purchase are different moments. Fewer people clicking can coexist with stronger conversion among those who do. The studies above cover different groups and periods, so you can’t multiply their figures into your own forecast. I’d ask your team a practical question: do you need more people to consider you, better information about the offer, or more suitable people reaching sales? Each problem suggests different work. A citation chart helps you investigate exposure; it doesn’t decide which problem matters most to the business. Observed unit | What it establishes | What remains unknown Brand mention | The name appeared in a sampled answer | How many buyers saw or trusted it Source citation | A page was linked as evidence | Whether the company was recommended Referred visit | Someone reached a measured website | Whether the visit was additional Completed sale | A commercial outcome occurred | What would have happened without the exposure ## Write something the answer can actually use The GEO paper accepted at KDD 2024 tested a benchmark of 10,000 queries and reported visibility improvements of up to 40% under its experimental measures. It looked at changes such as adding relevant sources, quotations, and statistics. The effect varied by domain and by how visibility was measured. The useful lesson is to improve the substance and clarity of the page. Give a real example, explain a condition, show the source for a claim, or answer the comparison the buyer is making. Don’t use the paper’s 40% as a lead forecast or fill a page with statistics just to imitate the treatment. ## Keep the SEO basics in the plan Google’s guidance explains that AI answers can use its Search index and related searches through query fan-out. It still recommends useful, original information and normal search fundamentals. You don’t need a separate page for every possible wording of the same question. I’d organise pages around the information a buyer needs: pricing, compatibility, implementation, risks, and proof. One answer may draw on several of those pages. Make each one clear enough to be useful on its own, and connect them so someone can continue their research. That’s a content strategy you can maintain. ## Check that the right crawlers can reach you OpenAI documents OAI-SearchBot for search and GPTBot separately for model training. Your policy on training can differ from your policy on appearing in search. Ask whoever manages the site to check the actual rules and access, rather than assuming all AI crawlers do the same job. When a page is absent from an answer, first rule out an access problem. A blocked page and a readable page that wasn’t selected need different fixes. Being allowed to crawl is only the starting point; it doesn’t guarantee a citation or recommendation. ## Review the platform settings too Google’s Search Console control for generative AI features rolled out worldwide on 31 August 2026. Inclusion is the default, but a property can inherit a parent property’s choice. Google distinguishes this setting from training controls and from exclusion from Search overall. I’d check the relevant property with the site owner before commissioning more content. An inherited exclusion, a crawl block, and a weak explanation can all leave you missing from an answer. You’ll save time by finding out which issue you actually have. ## Use Google’s reporting for the question it answers Google’s dedicated Generative AI performance report gives a view of impressions for supported Search AI features, grouped by page, country, date, and device. It rolled out worldwide on 31 August 2026, and sites with too few impressions may not see it. Use it to check link exposure. Pair it with saved answers, visits, and inquiries when you want to understand the wider journey. ## Use Bing to find where your sources participate Bing’s February 2026 AI Performance announcement describes citation counts, cited URLs, sampled grounding queries, and trends across supported experiences. It separates those figures from a page’s rank or placement in an answer. I’d use the report to find which pages contribute, then open the answers to understand their role. A technical source and a buying guide may each earn a citation while helping very different decisions. ## Turn the research into a plan you can run Here’s how I’d start: choose a buyer and market, write down their important decisions, save the current answers, and inspect the sources. Prioritise gaps your team can fix. Give each change a clear owner and review date, then recheck the same questions. Track visits and suitable inquiries where you can. At Searchable, I can help you build a custom Prompt Universe and get the team working through those findings. Bring your actual offer and the questions you hear in sales. If you want to see how that approach changes for a local service, read the senior living guide next: care fit, geography, and family expectations make the details especially important. ## How to read the research The sources were reviewed on 3 November 2026. These studies cover different years, products, and populations. I’ve kept their figures attributed and separated my recommendations from their findings. None establishes a standard revenue uplift for every business. Use them to ask better questions about your own market. ## Sources - OpenAI: How people are using ChatGPT: https://openai.com/index/how-people-are-using-chatgpt/ September 15, 2025 summary of the NBER working paper; 1.5 million consumer conversations. Usage classifications are not purchase-intent measures. - Pew Research Center: Google users and AI summaries: https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/ July 2025 observational study; methods and limits described above. - Adobe: AI traffic and US retail revenue per visit: https://business.adobe.com/blog/ai-traffic-higher-revenue-per-visit-retail-holiday October 29, 2026; September 2026 observed US retail results. Relative comparisons among referred visits are not incremental revenue estimates or non-retail benchmarks. - Aggarwal et al.: GEO, Generative Engine Optimization: https://arxiv.org/abs/2311.09735 Accepted to KDD 2024; version revised June 28, 2024. Experimental visibility metrics, not lead or revenue outcomes. - Google: Optimizing for generative AI features: https://developers.google.com/search/docs/fundamentals/ai-optimization-guide Official explanation of grounding, query fan-out and search fundamentals; reviewed November 3, 2026. Business implications are the author's analysis. - OpenAI: Overview of crawlers: https://developers.openai.com/api/docs/bots Official distinction between search and training crawlers; reviewed November 3, 2026. Access is not a guarantee of citation. - Google Search Console: Search generative AI control: https://support.google.com/webmasters/answer/16908024 Worldwide rollout August 31, 2026. Inclusion control, defaults and inheritance; distinct from model-training controls. - Google Search Console: Generative AI performance report: https://support.google.com/webmasters/answer/16984139 Worldwide rollout August 31, 2026. Impressions and supported dimensions; not a customer-level attribution report. - Bing: Introducing AI Performance: https://blogs.bing.com/webmaster/2026/2/Introducing-AI-Performance-in-Bing-Webmaster-Tools-Public-Preview/ February 10, 2026 preview announcement. Reports citations and sampled grounding queries across supported experiences, not customer outcomes. # Is your AI visibility actually improving? URL: https://daavidchrista.com/research/measure-ai-search-visibility/ Author: Daavid Christa Published: 2026-10-19 Before trusting a rising score, check the questions, the answers, and what changed in the reporting. Here’s how I’d review it with your team. A useful report should explain where you appear, what the answer says, and whether the change matters to your buyers. Keep the question set consistent, separate brand from discovery, and put platform data beside your own visits and inquiries. That gives you something you can act on. ## Decide what you want the number to tell you You’ve got a higher AI visibility score this month. What improved? Did more buyers have a chance to see your business? Did the answer describe you more accurately? Did you simply track different questions? I’d settle that before calling the month a win. Google can report a displayed link. A monitoring tool can save what a selected question produced. Website analytics can record a visit, and your CRM can record an inquiry. Each is useful, but each sees a different part of the journey. Start the review with the business question: are suitable buyers discovering us, getting the right information, and taking a useful next step? Then choose the evidence that can answer it. You’ll get a clearer report than putting every available number on one chart. ## Check how Google counts an AI impression Google says its Generative AI performance report became available worldwide on 31 August 2026. It covers link impressions in AI Overviews and AI Mode, excluding Search Labs experiments. It doesn’t count every mention of your name in the answer text. The default chart counts at property level. Two links from the same site in one feature count as one chart impression. Page tables count at page level, so adding their rows can give a different total. The AI report also overlaps with Web reporting. Adding both totals would count some exposure twice. Watch the export too. Values displayed as a dash or tilde become zeros in downloaded data. If a spreadsheet says zero, check whether the platform actually measured no impressions or didn’t have a usable value. I’d put the counting rule beside the chart, especially when a report goes to leadership. You shouldn’t need to be a Search Console specialist to understand what the number represents. Check the original view before you explain a jump or a disappearance. ## Treat Bing’s topics as a research aid Bing launched Topics Portfolio in beta on 13 October 2026. Topic Size approximates citations across sites for a topic. Queries counts distinct grounding queries, while Sites Cited estimates the sites participating. Opportunity volumes are given as ranges. Grounding queries are the searches used to find supporting material. They aren’t a count of people asking the same buyer question. A large topic in this report doesn’t automatically mean a large audience ready to buy from you. Some topics are left out when site presence or wider activity is too low. So a missing local topic doesn’t prove there’s no interest in it. I’d check its relevance with the people who know the market before removing it from the strategy. Use the topic view to find pages and questions worth investigating. Then check the actual answers, your commercial fit, and other evidence of demand. It gives you useful context without replacing those checks. ## Open the answer behind a citation Bing’s original AI Performance announcement describes citation counts, pages, query samples, and trends. It explicitly separates those measures from ranking, authority, or placement in an answer. That’s an important distinction when someone asks whether the brand is “winning.” Your page might be cited for a useful explanation while the answer recommends someone else. A care-funding page could help explain the rules without putting the provider on the family’s shortlist. The source did a job; it may not be the job you were measuring. The same applies to software documentation and retail buying guides. I’d check whether the business is described, compared, recommended, or criticised, and whether the source supports what’s said. A single citation count hides those differences. Keep a few saved answers in the leadership report. Show one useful improvement and one unresolved gap, with the question and source. That makes the numbers much easier to discuss and turns the review into decisions about the work. ## Split questions about your name from discovery Our anonymised Canadian senior living sample has 17 brand mentions across 100 generated ChatGPT questions. At first glance, that is a 17% mention rate. The split is more useful: 13 out of 13 questions naming the operator mention it, compared with four out of 87 questions without its name. That makes discovery about 4.6%. Around 76.5% of the observed mentions come from questions that already give the model the operator’s name. We checked the sample on 3 November 2026; the original collection period wasn’t shown. These are answer counts, not a measure of family reach. If I added more questions with the operator’s name, the overall score could improve even if discovery stayed exactly the same. I’d keep the groups separate and show how the question set changed whenever I introduced new coverage. Treat the question list as part of the result. Save the wording, topic, market, language, engine, and date. Otherwise the next review may be comparing two different tests. Question group | Mentions / questions | What this sample suggests Names the operator | 13 / 13 | Recognition when the name is supplied Does not name the operator | 4 / 87 | Limited discovery across these sampled questions Combined | 17 / 100 | A total that conceals the difference ## Don’t turn a sample percentage into a traffic forecast Picture two equal-sized test sets. One mostly asks about your company by name. The other asks people to discover suitable suppliers. You could perform the same within each type and still receive different overall scores because the questions are mixed differently. Neither test tells you how many real people ask those questions or see the resulting answers. It also doesn’t tell you whether the mention was helpful enough to prompt a click. Those are separate things to observe. I’d use the sample to test the answer, then use first-party records to follow visits and inquiries. Where the path isn’t visible, say so. You can still learn something useful about discovery without filling the missing steps with assumptions. A stronger score should give you a better question to investigate: which answers improved, for which buyers, and what happened next? It isn’t a percentage you can apply directly to your market size or website revenue. ## Explain a chart movement before reacting to it Google recorded a logging error in Generative AI Search impressions for 13–17 August 2026. Its 21 August update says the missing data was restored. The problem affected the reporting, rather than actual search visibility. If you saw that decline while it was happening, rewriting your content wouldn’t have fixed it. I’d first check reporting notices, recent data completeness, settings, and changes to the tracked questions. Then look at what changed in the answers themselves. That doesn’t mean every decline is a reporting problem. Competitors, content, the platforms, and buyer questions can all change too. It means the explanation should fit the evidence, rather than whichever marketing activity happened closest to the dip. The August incident is resolved. Keep it as an example of a check worth making, rather than an excuse to dismiss a later fall. If the same answer gaps keep appearing across repeated checks, they deserve attention. ## Connect the report to a useful business outcome Searchable separates brand mentions from linked sources and lets teams inspect the answer evidence. I’d use that to understand where you’re missing and what the system says about you. The next question is whether the change helped someone make a better decision. In senior living, that means looking at suitable inquiries, tours, and move-ins separately. A prominent recommendation for care you don’t offer can create poor inquiries. A corrected service explanation can save the team time even if lead volume doesn’t rise. Agree what matters with sales or admissions before the review. That might be clearer expectations, discovery in a priority location, or better inquiry fit. Then show the evidence for that outcome instead of relying on the most flattering chart. Keep credited revenue separate from revenue you believe the work added. A CRM source label helps describe the journey; it doesn’t prove the customer would never have arrived otherwise. The size of the claim should match the comparison you can support. ## End the review with something to do I’d keep the report to a few connected questions: what did we track, which answers changed, which pages gained exposure, and what happened to the inquiries we can follow? Keep the groups and dates visible so someone can compare the next month fairly. Then choose the next action. Fix a recurring wrong claim. Investigate why a competitor appears for a relevant comparison. Check whether a new page is being found. Each should have a named owner and a clear question for the next review. This guide uses Google and Bing documentation plus our supplied answer sample, checked on 3 November 2026. The equal-sized test sets are an illustration, not a campaign result. The source notes keep the platform definitions and limits available. I can help you set up that review in Searchable and make the findings useful to the team. Next, take a closer look at citations. Once you know where a source appears, you still need to check whether it supports what the answer says. ## Sources - Google Search Console: Generative AI performance report (Search): https://support.google.com/webmasters/answer/16984139 Official documentation reviewed November 3, 2026; global rollout dated August 31, 2026. Covers AI-feature link impressions, aggregation, overlapping Web data, and export treatment of unavailable values. - Bing: Introducing Topics Portfolio (Beta) in AI Performance: https://blogs.bing.com/webmaster/2026/10/introducing-topics-portfolio-beta-in-ai-performance/ Published October 13, 2026. Approximate citation-based topic measures and visibility thresholds; not counts of end users or consumer prompt volume. - Bing: Introducing AI Performance in Bing Webmaster Tools—Public Preview: https://blogs.bing.com/webmaster/2026/2/Introducing-AI-Performance-in-Bing-Webmaster-Tools-Public-Preview/ Published February 10, 2026. Official scope and metric definitions for supported Microsoft AI experiences. Citations do not measure rank or placement. - Anonymized senior living visibility dataset and lesson: https://daavidchrista.com/research/senior-living-case-study-evidence/ Supplied report recounted November 3, 2026. One Canadian operator, 100 generated ChatGPT questions, 17 mentions. Question composition calculations are ours. Collection date and window unavailable; not consumer research or market share. - Google Search Console: Data anomalies: https://support.google.com/webmasters/answer/6211453 Reviewed November 3, 2026. Generative AI Search impression logging issue dated August 13–17, 2026; August 21 update confirms restoration. No effect on actual visibility reported. - Searchable: AI Mentions & Citations Analytics: https://www.searchable.com/features/aeo-insights/mentions-citations Official feature description reviewed November 3, 2026. Supports examining mentions and sources separately. Commercial and causal interpretation is our analysis; I work at Searchable. # AI cited your business. What did it actually say? URL: https://daavidchrista.com/research/ai-citations-evidence/ Author: Daavid Christa Published: 2026-11-03 Open the link, check the claim, and see what a buyer would take away. That’s where I’d start before celebrating another citation. A citation gives you somewhere to look. It doesn’t settle whether the answer is accurate, whether the source is independent, or whether the business was recommended. I’d review the important claim and its evidence together, then decide which gap needs work. ## Read the sentence before counting the link Your page is cited in an AI answer. What is it being used to support? It might show that you offer a service. That doesn’t automatically establish that you’re the best option, available in the buyer’s area, or suitable for their situation. Read the claim and the page together. Five links can mean five useful sources. They can also mean five copies of one announcement, or a few pages that don’t support the sentence beside them. I’d count citations to see where to investigate, then review the claims that could affect a buying decision. ## Check whether the source supports the whole claim A 2023 Stanford study tested four generative search engines available at the time. It found that 51.5% of generated sentences were fully supported by citations, while 74.5% of citations supported their associated sentence. Those are two different checks: was the claim supported, and did the attached link actually justify it? The products were Bing Chat, NeevaAI, Perplexity, and YouChat in their then-current versions. I wouldn’t present those figures as an error rate for today’s tools. I would keep the review habit: find the exact passage, then check which parts of the answer it really supports. ## Find the original source where you can The Tow Center’s March 2025 study gave eight search-enabled AI tools excerpts from 200 news articles, creating 1,600 queries. It asked for the article, publisher, and URL. More than 60% of responses contained errors, even though the excerpts had been selected so Google returned the original in its first three results. That was a specific retrieval task, not every kind of AI question. But the issues it found are worth checking: copied versions receiving credit, wrong attribution, and broken or invented links. If your company publishes original research, I’d inspect whether the answer leads to your actual work or a page repeating it. ## Don’t confuse being published with being true An October 2026 preprint by Qi Liu and colleagues analysed 17,211 citation instances across ten AI-search platforms and 6,356 domains. In its publication experiments, eight platforms cited a fabricated concept within seven days. The researchers traced distinctive markers back to their own publications. That’s a reason to check the underlying evidence, not a tactic I’d recommend copying. The tests used designed questions and measurements from March–April 2026, and the paper is a preprint. Its useful lesson is that a source entering an answer doesn’t make the source’s claim true. ## Repeat important checks Muck Rack and Generative Pulse’s October 2026 release includes a September panel of 1,020 unbranded consumer prompts across four systems. In same-day repeat tests, it reports cited-domain overlap of 41% for ChatGPT, 45% for Gemini, and 94% for Perplexity. These are the publishers’ findings, rather than an independent audit of the vendor. A changing source list can still produce a correct answer, and a stable one can keep repeating a mistake. I’d repeat important checks to see whether the problem persists, while keeping accuracy and consistency as separate questions. One screenshot is a useful observation, not the whole picture. ## Work out which problem you’re looking at These studies test different things: sentence support, finding an original article, picking up a fabricated idea, and repeating the same domains. Different questions, years, products, and groups counted. Averaging their percentages would create a reliability score they never measured. For your business, I’d turn them into four checks. Does the source support the important claim? Is the credit right? Is the source itself credible? Does the issue happen again? You may need a content correction for one problem and a source-attribution investigation for another. Evidence problem | What can go wrong | Commercial consequence Support | A link does not justify the nearby claim | A buyer learns an unsupported product promise Attribution | The original work is credited elsewhere | The originator loses identifiable recognition or traffic Source quality | Published material is mistaken for established fact | Weak information gains an appearance of authority Repeatability | The same question produces different sources | One screenshot overstates a stable competitive position ## Check whether several sources just repeat one claim Imagine your product announcement is repeated by four publications. AI cites those pages and says the product saves a certain amount of time. There are several URLs, but the performance claim may still come from your original announcement. Independent testing would add something different. I’d follow a significant number back to whoever measured it. What was the comparison, what period did it cover, and what conditions applied? A company case study can be useful. It stays more useful when a buyer can see who reported the result and what was actually counted. ## Pay attention to the missing conditions Here’s a fictional example I’d use in a review. Your site says an integration is available on one enterprise plan. The AI answer says the product includes the integration. Both names are right, but the missing plan condition can send the wrong prospect into a demo. A brand-match score won’t catch that. For a local service, the same problem could assign a group’s care capability to every branch. Check plan, location, date, eligibility, and price conditions where they change the offer. These examples explain what to look for; they aren’t measured errors from a customer account. ## Use reporting to find the work Bing’s June 2026 AI Performance update defines citation share as a site’s share of citations for a grounding query. It says this is an observation of source use, doesn’t expose competitor domains, and isn’t traffic share or a quality score. Comparisons can show changes without identifying one certain cause. I’d use that reporting to find pages worth opening. Then ask whether the resulting answer helps the intended buyer understand the business. The chart tells you where your source participates; the review tells you what to improve. You need the actual claim to make that second decision. ## Turn the check into a clear brief Save the question, answer, claim, source passage, and the correction you believe is needed. Get the relevant team to confirm the fact. Then choose the action: improve your page, correct an outside listing, or investigate further. That’s a brief someone can pick up and complete. I can help connect those findings to a question set and action plan in Searchable. If you run local care communities, read the community-facts guide next. It takes this review into the details that matter most there: service fit, official records, ratings, and what a quoted price includes. ## A note on the studies Sources were checked on 3 November 2026. I haven’t rerun these experiments. The older results stay labelled by date, the October paper is a preprint, and the vendor release is attributed. The product and location examples are fictional. There’s no combined accuracy score because the studies measured different things. ## Sources - Liu, Zhang and Liang: Evaluating Verifiability in Generative Search Engines: https://arxiv.org/abs/2304.09848 2023 study of four then-current engines. Sentence support and citation precision are distinct measures; not a current-product benchmark. - Tow Center: AI Search Has a Citation Problem: https://www.cjr.org/tow_center/we-compared-eight-ai-search-engines-theyre-all-bad-at-citing-news.php March 6, 2025. Original researchers' account of 200 article excerpts tested across eight tools, yielding 1,600 queries. Specific news-retrieval task. - Qi Liu et al.: From Public Posts to AI-Search Citations: https://arxiv.org/abs/2610.11932 Preprint submitted October 8, 2026; measurements from March–April 2026. Experimental publication effects are not a universal ranking or conversion model. - Muck Rack and Generative Pulse: October 2026 citation research: https://rss.globenewswire.com/news-release/2026/10/28/3390134/0/en/generative-pulse-earned-media-still-drives-more-than-84-of-ai-citations.html October 28, 2026 vendor-issued research release. Broad monitoring and the separate prompt panel have different scopes; overlap is not an accuracy metric. - Bing: Intents, Topics, Citation Share and Compare: https://blogs.bing.com/search/2026/6/New-AI-Visibility-Insights-in-Bing-Webmaster-Tools-Intents-Topics-Citation-Share-Compare/ June 16, 2026 official preview announcement. Defines citation share and its limitations; reviewed November 3, 2026. # Should you pay for AI search ads? URL: https://daavidchrista.com/research/ai-search-ads-brand-playbook/ Author: Daavid Christa Published: 2026-11-03 Check where you can advertise, what you can measure, and whether the offer fits. Here’s how I’d decide if a test is worth running. Start with access and fit, then build a test around your margins and a clear outcome. Google and OpenAI offer different placements and reporting. Their case studies can help you ask questions, but they don’t give you a ready-made return forecast or buy you a favourable AI answer. ## Check the market before planning the campaign Google announced on 28 October 2026 that it was expanding AI Mode ads to Australia, Canada, India, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines, and Singapore. Brazil, Mexico, and the UK were listed as planned Q4 additions. It also announced ads in follow-up conversations in the US. Check your account before treating a planned launch as available inventory. Google says AI Mode questions averaged three times the length of traditional searches in its May 2026 US research. Longer questions can tell you more about the need, but they don’t automatically make a better sale. I’d start with the customer, market, and offer you can actually reach. ## Understand what you’re buying in Google Google’s documentation makes relevant Search, Shopping, and Performance Max campaigns eligible for these placements. You can’t exclusively target, or separately opt out of, AI Overviews and AI Mode. Their impressions, clicks, conversions, and costs feed into standard campaign reporting together with other activity. Some sponsored options also have independently generated AI explainers that the advertiser can’t manually edit or approve. Google says buying ads doesn’t influence its organic AI answers. So I’d distinguish the placement you buy from the explanation the system produces, and be careful about crediting an aggregate campaign improvement to AI placements alone. ## Check advertiser access and audience access separately OpenAI’s Ads Manager guidance bases self-service eligibility on the country of the legal entity advertising and being billed. It lists eligible countries and separates account creation from getting access. A successful signup doesn’t confirm every audience, format, or category is open to your business. I’d check three things before spending time on creative: the billing entity can advertise, the target market is available, and the specific offer is eligible. If your company operates internationally, write those answers down for the market you actually want to test. ## Keep paid placement and the AI answer separate OpenAI says ads run separately from the chat model and can’t shape or rank its answers. Its current FAQ describes ads for eligible Free and Go users, with the listed paid individual and business subscriptions remaining ad-free. It also describes testing a separate multi-image placement during image generation. Advertisers get aggregated performance information, rather than people’s chats, histories, or memories. A relevant placement gives you a chance to be noticed. You still need the person to choose a next step. I wouldn’t build a sales plan around the idea that an ad gives you access to the conversation or an AI endorsement. ## Read the outcome behind each case-study number OpenAI’s 5 October announcement reports partner findings for WeightWatchers, Dose, and Portland Leather. The table below keeps each result attached to what it measured. A lower attributed acquisition cost, the customer mix within additional purchases, and the share of new visitors are all useful questions, but they give you different answers. These are platform-published findings, not an independently audited advertiser average. Business and reporting partner | Published finding | What the number measures WeightWatchers · DV Rockerbox | 15.3% lower attributed acquisition cost than its blended paid-search benchmark | Cost assigned to an acquisition under the measurement method Dose · WorkMagic | 67% of incremental purchases came from net-new customers | The customer mix within purchases estimated to be additional Portland Leather · Triple Whale | 93% of visitors from ChatGPT Ads were new | Visitor newness, not a purchase or profit rate ## Use your margins to decide what the result is worth A new visitor may never buy. An attributed customer may have bought anyway. And a new sale may still cost too much to fulfil. Before the test, I’d agree what counts as success after advertising, delivery, and service costs, using the event your business can actually measure. For a longer sales cycle, you may need to review suitable inquiries first and completed sales later. Keep the connection clear rather than treating every form submission as revenue. The published cases show useful outcomes are possible; your offer and economics decide whether the same channel is worth your investment. ## Give the measurement window some thought OpenAI’s measurement article says 52.7% of eligible one-day view-through conversions in an early analysis happened within an hour of the matched impression. That counts a group of attributed conversions; it doesn’t say 52.7% of ad viewers bought. The article also reports a WorkMagic geographic-lift study estimating 2.3 times more additional Dose orders than last-click tracking captured. One method connects an impression with a later event inside a window; the other estimates extra business through an experiment. I’d discuss which method you’re using before the test starts. A purchase soon after an impression is useful evidence, but timing alone doesn’t prove the ad caused it. OpenAI describes this work as early-stage, so don’t apply its multiplier to your own sales. ## Check the policy for your actual offer OpenAI’s policies, updated on 10 September 2026, restrict health services outside the US and describe case-by-case approval for some US health advertisers. They also prohibit individual housing rental and sales ads. Separate rules exclude vulnerable contexts, including personal and mental health conversations. For senior living, I’d get the platform’s classification of the exact offer and market before building a budget around it. A retail case doesn’t settle whether a care-related campaign is eligible. Where paid placement isn’t available, families may still need clear public information and accurate organic answers. ## Ask what the campaign added A detailed conversation can help an ad meet a specific need. It can also put the ad in front of someone who was already close to buying. Both can look good in an attribution report, even though they may add different amounts of business. Picture two fictional buyers. One was already planning to purchase and sees an ad on the way. The other discovers a suitable offer through the ad and makes a purchase they otherwise wouldn’t have made. Both purchases could receive advertising credit; only the second adds a sale in this simplified example. Keep that question in the review, especially when existing brand demand is strong. ## Here’s how I’d make the decision Confirm access and policy fit, choose a specific offer and audience, and agree the result you’ll judge. Keep the budget limited enough to learn from. Review the available reporting, inquiry or customer quality, and contribution after costs. Expand because the evidence supports it, rather than because the platform is new. I work at Searchable and can help connect your paid test with the wider questions buyers ask about you. Accurate information, useful organic discovery, and paid exposure each have a role. If you need to put the whole strategy together, return to the implementation guide and start with your buyer’s actual decisions. ## Keep availability and results in context The sources were reviewed on 3 November 2026. Availability comes from official documentation and dated announcements; planned launches remain plans. Partner results retain their original measures and haven’t been independently audited here. The fictional buyers illustrate attribution, rather than describe real customers. Recheck access and policy in your account before running a campaign. ## Sources - Google: Ads in AI Mode expand to new markets: https://blog.google/products/ads-commerce/ads-in-ai-mode-global/ October 28, 2026. Ten announced expansion markets; Brazil, Mexico and the UK remain Q4 plans in this source. - Google Ads Help: Ads in AI-powered Search experiences: https://support.google.com/google-ads/answer/18826097 Current campaign eligibility, controls, aggregate reporting and answer independence. Reviewed November 3, 2026; product behavior can change. - OpenAI: Ads Manager Availability: https://help.openai.com/en/articles/20001245-ads-manager-availability Reviewed November 3, 2026. Legal-entity country eligibility is distinct from audience, placement and offer eligibility. - OpenAI: Ads in ChatGPT: https://help.openai.com/en/articles/20001047-ads-in-chatgpt Current user eligibility, paid-placement separation and conversation privacy. Reviewed November 3, 2026. - OpenAI: Building advertising for the way people use AI: https://openai.com/index/new-chatgpt-ads-format-and-measurement/ October 5, 2026. Partner-reported case results use different definitions; not independently audited here or representative of every advertiser. - OpenAI Ads: More ways to measure ChatGPT Ads: https://ads.openai.com/blog/more-ways-to-measure October 5, 2026. Early attribution analysis and reported geographic-lift evidence; different denominators and methods. - OpenAI: Ad policies: https://openai.com/policies/ad-policies/ Updated September 10, 2026. Offer restrictions and sensitive-placement exclusions; does not establish eligibility for every senior-living business model. # Life and media