Blog, page 23 of 28
Notes on getting recommended by AI.
The full archive. Everything we have published on what AI apps read, what they cite, and what moved the numbers.
AI visibility monitoring toolsAI Visibility Monitoring Tools Tested: A Reproducible Test Design of Coverage, Citations, and PriceMentioned once and then gone is usually a repeat count problem, so the number that matters is runs per exact prompt per service.29 September 2026page not cited by AIWhy Your Most-Visited Page Isn’t Cited by AI: A URL-Level Crawl, Indexing, and Citation TestTraffic, indexing and citation are three different measures, and the denominator that matters is answers that could have cited anything at all.29 September 2026AI confuses similar business namesAI Confuses Two Businesses With Similar Names: A Disambiguation Audit and Evidence-Based Fix GuideSix kinds of mix-up, the identity table that makes them countable, and the wrong-company rate we have not measured.29 September 2026Google rankings vs AI mentionsGoogle Rankings vs AI Mentions: A Matched-Query Study With Search Console and Answer-Level DataWhy a good Google position does not carry into an AI answer, and the matched query study that would measure it properly.29 September 2026AI mentions without linkingAI Mentions My Business Without Linking: A Citation and Link Audit MethodThree separate things get called a mention, and the unlinked mention rate only means something when all three are counted apart.29 September 2026AI quotes old priceWhy AI Quotes an Old Price: A Dated Price-Audit Test Across Your Website, Feeds, and AI AnswersThe full price consistency protocol, the four real price failures we measured, and the match rate we have not yet run.29 September 2026AI says my business is closedAI Says Your Business Is Closed: A Source-by-Source Status Check and Correction PlaybookThe order to work through a wrong closure claim, the registers to check, and the status accuracy count we have not measured.29 September 2026Kolkata marketing agenciesKolkata marketing agencies: verified services, public pricing and a reproducible comparisonThe dual evidence rule, the agreement check between two researchers, and a stated count of nil firms checked as of 29 September 2026.29 September 2026Ahmedabad marketing agenciesAhmedabad marketing agencies compared: a dated directory with public price evidenceWhy every price needs an as at date, how often to recheck, and a stated count of nil agencies with a price verified on a named primary source.29 September 2026Chennai marketing agenciesChennai marketing agencies: a source linked directory and pricing datasetWhat a source link must contain before a row counts, how to archive it, and a dated count of nil agencies verified from a primary page.29 September 2026Hyderabad marketing agenciesHyderabad marketing agencies: verified services and published pricing, updated 29 September 2026How to state a share as a count over a denominator, the three traps that make shares dishonest, and our own share of 0 of 0.29 September 2026Pune marketing agenciesPune marketing agencies compared: services, public pricing and selection methodEight checks that verify a firm's own service and location claims against its own site, and a stated count of nil as of 29 September 2026.29 September 2026Delhi marketing agenciesDelhi marketing agencies: a dated directory with verified services, prices and methodHow to draw the Delhi and NCR boundary before you count, how to deduplicate one firm with three entities, and why our count is nil.29 September 2026Mumbai marketing agency pricesMumbai marketing agency prices and services: a source linked comparison, updated 29 September 2026The fraction that matters is how many named agencies publish a price at all, and this page gives you the method to produce it.29 September 2026Bengaluru marketing agenciesBengaluru marketing agencies compared: a dated, evidence scored directoryThe inclusion rules, the ten point evidence rubric, and a plain statement that the count of agencies evaluated is nil as of 29 September 2026.29 September 2026GEO retainerWhat should a GEO retainer include? Deliverables, hours and how to price themFourteen line items a retainer must name, the pricing evidence to collect yourself, and why we publish no floor price.29 September 2026AI crawlers robots.txtWhich AI crawlers should you allow? A dated robots.txt and access control testSixteen named crawler tokens, four configurations to test, and how to count real fetches from your own server logs.29 September 2026do AI answers favour big brandsDo AI answers favour big brands? A dated test of 100 category promptsHow to define brand size groups before you measure, the rate difference arithmetic, and the only dated counts we hold.29 September 2026which businesses benefit from GEOWhich businesses benefit most from GEO? A 100 question, five category testThe count that answers this is responses where a business was both recommended and cited, and here is the test that produces it.29 September 2026does GEO replace SEODoes GEO replace SEO? What Google's own documentation requires, and a 30 page eligibility testThe number to look for is zero additional technical requirements, the document that carries it, and a test that shows whether your pages qualify.29 September 2026AI visibility auditAI visibility audit: a reproducible 50 prompt checklist, scoring rubric and sample auditThe 50 prompt design with its 600 response arithmetic printed, the scoring rubric in full, and the dated samples we actually hold.29 September 2026AI Overviews law firmCan a law firm website be cited in Google AI Overviews? A reproducible search testZero queries captured so far, so this page publishes the search test and the empty cell rather than a share.29 September 2026Claude citationsClaude web search and law firm citations: a reproducible test of what it finds and citesZero Claude answers tested for our own domain, and six dated Claude findings from the nearest sector we hold.29 September 2026Perplexity citation ratePerplexity law firm citations: a repeatable audit method for queries and resultsThe measured citation rate for our own domain on Perplexity is 0 of 6, recorded in September 2026.29 September 2026Gemini law firm recommendationsHow often does Gemini recommend law firms? A reproducible test across named markets and practice areasThe count we hold for Gemini is zero answers tested, so this page publishes the test rather than a rate.29 September 2026AI Overviews marketing agencyGoogle AI Overviews for marketing agency queries: a dated local and industry search auditLocation is the variable that decides this audit, so it goes in every row beside the date.29 September 2026Claude marketing agencyClaude web search for marketing agency recommendations: a dated prompt and citation studyPrint the repeat run count next to every prompt, and report cited mentions as a fraction of those runs.29 September 2026Perplexity marketing agencyPerplexity's marketing agency answers: which agencies and sources appear in a repeatable test?A distinct domain count means nothing without the answer count and the query count beside it.29 September 2026Gemini marketing agencyHow Gemini answers 'best marketing agency' searches: a repeatable local and industry prompt studySplit the prompts by type, repeat each one, and report named plus cited as one fraction per type.29 September 2026Google AI OverviewsGoogle AI Overviews for SaaS queries: how eligibility and links work, and how to check themThree separate denominators, and why a share of queries showing an Overview belongs to one date and one location only.29 September 2026Claude web searchClaude web search and SaaS recommendations: a dated citation and repeatability testThe fraction that matters is runs where a product was named and supported by a citation, printed per prompt with its sample size.29 September 2026Perplexity citationsWhat sources does Perplexity cite for SaaS recommendations? A reproducible prompt auditHow to count distinct source URLs properly, and why a URL total with no answer count and no prompt count means nothing.29 September 2026Gemini SaaS recommendationsWhich SaaS companies does Gemini recommend? A dated, repeatable citation studyThe repeat run count is the method. A share of runs with no run count beside it is not a measurement.29 September 2026AI describes my business wronglyAI describes my business incorrectly: a 30 query fact accuracy audit with an evidence logHow to build the fact sheet, count errors against a stated total, and why we publish no error fraction of our own yet.29 September 2026AI brand recommendation testDoes AI recommend your competitor instead? A repeatable 5 assistant brand recommendation testThe full protocol, and our own measured results printed as counts out of denominators, with the engine and the date on every one.29 September 2026Wikipedia notabilityCan a new company have a Wikipedia article? A source based notability self check and conflict of interest guideHow to count qualifying independent sources honestly, why the answer for a new company is often zero, and what to do instead.29 September 2026Quora business profileHow to create a Quora business profile and answer questions transparently: a dated, step by step checklistThe setup protocol, the fields to log per attempt, and why a single successful setup is one observation and not a success rate.29 September 2026Reddit for businessHow a new business can build a transparent Reddit presence: profile setup, subreddit rules, and a 30 day participation logThe rules check to run before you post anywhere, the log format, and an honest count of the communities whose rules we have checked.29 September 2026G2 categoriesHow to get a new software product listed and categorised on G2: eligibility, evidence and a reproducible submission checklistHow to build the evidence pack for a new category request, including how to count competing products without fooling yourself.29 September 2026Clutch profileClutch company profiles: the free basic listing, what verification means, and how to record current pricing yourselfThe difference between a free profile and a paid tier, and why we name the page to read instead of printing a figure.29 September 2026Product Hunt submissionHow to submit a product on Product Hunt: current rules, account requirements and a launch checklist you can dateHow to record a platform rule so it cannot go stale into a false claim, and the launch items that are entirely under your control.29 September 2026Trustpilot business profileTrustpilot business profile: free tools, public pricing and review rules, and how to record a price so it stays trueWhy we will not restate a platform's lowest paid plan, and the region and billing fields that make your own check reusable.29 September 2026TrustRadius vendor profileTrustRadius vendor profile: what is publicly documented about listing, reviews and pricing, and what we will not claimA page with no number in it, because no number can be responsibly named before a dated check of the current official process.29 September 2026GetApp listingGetApp listing eligibility: how to build a dated checklist for software vendors that does not go staleThe checklist format, the source version record, and why our inclusion consideration count is empty as of 29 September 2026.29 September 2026Software Advice listingHow to get a Software Advice listing: vendor submission, free profiles and pricing data, and how to record a price band properlyWhat a symbol based price band can and cannot tell a buyer, the twelve fields to record, and why we publish no scale of our own.29 September 2026Capterra listingCapterra product listing: eligibility, submission, and free versus paid options, with a dated count you make yourselfHow to count the stated eligibility criteria in a guidelines document reproducibly, and why we publish no count of our own.29 September 2026G2 listingHow to get a software product listed on G2: eligibility, submission and verification, with a dated check you run yourselfThe submission protocol, the fourteen fields to log, and why we publish no median time to a claim decision.29 September 2026dental clinic AI searchHow AI search describes dental clinics: a reproducible India test of local treatment queries, September 2026Ten treatments across six cities, the own domain citation fraction, a five field description accuracy check, and a clear statement of what has not been run.29 September 2026recruitment agency questionsQuestions employers and candidates ask before choosing a recruitment agency: the interview protocol, and why N is still zeroThirty five questions split between the two populations, the full interview and coding method, and a plain statement that no interviews have been conducted.29 September 2026recruitment agency AI searchAI recommendations for recruitment agencies: a dated, reproducible India benchmark by role and cityThe role and city grid, separate employer and candidate waves, the own domain citation definition, and an honest account of what has not been run.29 September 2026accountant buyer questionsWhat people ask AI before choosing an accountant: a published prompt study in India, and the count we do not yet haveThe sixty question set we would test, the five lawful ways to collect real prompts, the coding rules, and a plain statement that no prompts have been collected.29 September 2026accounting firm AI visibilityHow often do AI assistants recommend local accounting firms? A reproducible India benchmark, and what we have and have not run as of September 2026The city and service grid, the own domain citation definition, the 324 observation arithmetic, and the nearest X of Y we actually hold.29 September 2026questions to ask a lawyerBefore hiring a lawyer: 25 questions to ask, what to verify, and where to check itTwenty five questions in five groups of five, each with the thing to verify and the official place to check it in India.29 September 2026law firm AI visibilityLaw firm visibility in AI search: a jurisdiction first audit checklist and reproducible citation testOne jurisdiction reviewed against primary regulator material, the regulators named for the others, and the citation test written out in full.29 September 2026hiring a marketing agencyBefore you hire a marketing agency: 30 questions, evidence to request, and a proposal comparison sheetThirty questions in six groups of five, the proof to ask for with each, and fourteen rows to fill in for every proposal you receive.29 September 2026marketing agency pricingMarketing agency pricing in 2026: a dated, auditable survey of published fees and project budgetsThe full survey protocol, the fields to capture, the scope, and an honest count of how many dated public price pages we have reviewed so far.29 September 2026B2B software checklistThe B2B software evaluation checklist: 40 questions buyers can verify before signingForty questions in eight groups of five, each with the document or proof to ask for, and a clear statement that this is a checklist rather than a survey result.29 September 2026B2B software benchmarkB2B software in AI search: a 100 query citation and recommendation benchmark, with raw resultsThe query set, the row schema for raw results, the four outcome levels, and a count of how many of our observations a reader can check from outside.29 September 2026SaaS buyer questionsWhat software buyers ask before choosing SaaS: 100 questions, their frequency, and a reproducible coding methodAll 100 questions enumerated in ten groups of ten, the coding method that would turn them into real frequencies, and why no frequency is claimed here.29 September 2026AI search benchmarkDoes AI search recommend SaaS products? A reproducible 90 day benchmark across ChatGPT, Gemini, Perplexity and Google AI OverviewsThe protocol in full, the observation counts we actually hold with their dates, and a plain statement of the parts we have not run.29 September 2026
See where you stand before you read another page.
Enter your website for a free check. No account, and it asks real buyer questions rather than your brand name.
Already sure you want this? Join the waitlist. We open on 31 October 2026. It costs an email address. No card, and nothing is charged. Or talk to a person if you have a question first. Or start with the guides, which are step by step.