Explore insights on Generative Engine Optimization (GEO), AI visibility, and how modern brands position themselves for discovery by AI assistants like ChatGPT, Claude, Gemini, Perplexity & more.
We re-asked every question our own reviews recorded as unverifiable, on the vendor's own pages. 6 of 98 are answered today, and 0 of 14 Enterprise prices.
Maher El Ouahabi · · 16 min read Every vendor in our verified catalogue, read on the page its own entry points at, against a frozen engine list. 11 of 77 name none where they sell.
Maher El Ouahabi · · 14 min read We asked two AI surfaces what LLM SEO is. They cited fourteen pages giving the same eight tips, none with a measurement. Five do not survive our register.
Maher El Ouahabi · · 41 min read Across three strata of cited domains, 48 of the 76 refusals of AI training were a CDN default, and 53 of the 81 signals a person typed permit it.
Maher El Ouahabi · · 26 min read AI Overview returned nothing for 11 buying questions in Spain and Germany. AI Mode answered all 36 cells in three markets: the surface, not the market.
Maher El Ouahabi · · 25 min read We asked an assistant the price of 30 AI visibility tools and checked every answer against the vendor's own page. 13 of 20 right, and 4 invented prices.
Maher El Ouahabi · · 21 min read Fourteen buying questions asked twice in August and again six days later. The full cited set overlaps 50 per cent. The durable core holds 73 per cent.
Maher El Ouahabi · · 18 min read Ask an AI assistant for alternatives to a vendor and the vendor itself usually is not there. In one answer, neither was the category.
Maher El Ouahabi · · 24 min read A sweep of 77 AI visibility vendors found one disallowing AI crawlers. The block is a managed robots.txt that stops training, not retrieval.
Maher El Ouahabi · · 12 min read Across 48 vendor observations in 7 software categories, answers named a vendor 27 times and cited that vendor's own page 4 times.
Maher El Ouahabi · · 15 min read Reddit is cited in 8 of 8 small-business cells and 1 of 5 enterprise ones. In 4 of 4 paired cells, only the buyer changed.
Maher El Ouahabi · · 13 min read BrightEdge publishes no price. Thirteen alternatives, eleven of which do, with the engines each price buys. We are in the table.
Maher El Ouahabi · · 12 min read Crossed on 77 AI visibility vendors, doing one convention does not predict the other, a 5 point gap. Repeated on 2,147 cited domains in three strata, it does.
Maher El Ouahabi · · 20 min read Sixteen Peec AI alternatives with the entry price read from each vendor's own page and the engines that price buys. We sell here, and we are in the table.
Maher El Ouahabi · · 12 min read A week of our own edge logs by agent and path. Assistants spend a median 77 percent of fetches on articles, crawlers 4 percent. One day erases the split.
Maher El Ouahabi · · 12 min read Schema markup is the standing advice for AI visibility. Of 46 readable pages that AI answers actually cited, 93% carry JSON-LD and 41% declare what the page is.
Maher El Ouahabi · · 13 min read Every active vendor in our catalogue asked for markdown two ways. The category ships llms.txt at four times the rate, and no incumbent serves it.
Maher El Ouahabi · · 13 min read Eleven Profound alternatives with entry prices read from each vendor's own page and the engines each one buys. We sell here, and we are in the table.
Maher El Ouahabi · · 13 min read One buying question, asked twice minutes apart, in four markets. The sources that hold repeated 100% of the time in the United States and 50% in France.
Maher El Ouahabi · · 14 min read Twenty buying questions in four markets, one reading each, no retries. Seventeen answered every run. All three that did not are the same category.
Maher El Ouahabi · · 19 min read Publishing on a domain you do not own is sold as AI visibility. We read the robots.txt of every host that advice names. Two AI firms block each other.
Maher El Ouahabi · · 20 min read 11 of 15 comparable categories return identical sources whichever way you ask. 4 return almost nothing in common, and only a control tells you why.
Maher El Ouahabi · · 34 min read We fetched every URL nine AI Overview answers cited and read each host's robots.txt. None was dead, and 11 cited publishers disallow AI crawlers.
Maher El Ouahabi · · 14 min read We recorded the exact URL of every citation in ten AI Overviews across two markets. Domain counting hides 18 of 91 pages, and no cited page was a homepage.
Maher El Ouahabi · · 16 min read We asked 69 AI visibility vendors and 122 domains we have seen cited for their llms.txt on one day. The vendors ship it far more often than the cited sites.
Maher El Ouahabi · · 17 min read We fetched robots.txt from all 69 vendors in our catalogue. Seven declare Content Signals, nobody has adopted the fourth field, and one says no to training.
Maher El Ouahabi · · 16 min read GeoGenie charges $100 a month for any two of five engines. The $500 tier still gives three. Verified prices, the arithmetic, and where it is the right call.
Maher El Ouahabi · · 11 min read Changing the answer surface costs more than changing the day: median source overlap 0.40 against 0.63 across two days, on identical questions.
Maher El Ouahabi · · 20 min read Across 29 category and market cells, 119 durable slots go to non-vendors. 79 per cent of those domains appear in exactly one cell. YouTube is in 24.
Maher El Ouahabi · · 9 min read Twelve domains cited in seven of eight runs and one in one. Cited-in-every-run is not measuring stability, it is measuring whether you caught the odd run.
Maher El Ouahabi · · 12 min read Thirteen categories re-asked after two days. Median source overlap 0.63, and the best and worst rows both moved when read again the same afternoon.
Maher El Ouahabi · · 15 min read Pooled across six markets the vendor share climbs 4, 19, 22, 22, 38 per cent by category. Two same-language pairs each show the second market far lower.
Maher El Ouahabi · · 21 min read Across 120 published measurements in four markets, the stable core stays a narrow band while cited breadth varies ninefold. The percentage moves anyway.
Maher El Ouahabi · · 12 min read Six questions read deeply, then the same intent asked three ways. Only one wording returns a source set you could hand a buyer, and it is the shallowest.
Maher El Ouahabi · · 51 min read Everyone defines the difference the same way. Three measurements change the decision: when an answer exists, who holds its citations, and which page wins.
Maher El Ouahabi · · 11 min read Six brands asked what they cost and what their category's best is, same surface, same day. Six of six cited their own site on price. One of six on category.
Maher El Ouahabi · · 14 min read We tested seven programmatic comparison pages one URL at a time. The four that name their own publisher rank for nothing at all.
Maher El Ouahabi · · 12 min read Three checks, sixty-six vendors, zero failures. A clean sweep is not good news about the category, it is bad news about the checks. Here is what they miss.
Maher El Ouahabi · · 11 min read Limy starts at $149 a month for three engines. The $449 plan still buys three, it just lets you choose which. The arithmetic and what we could not verify.
Maher El Ouahabi · · 12 min read Thirty local service questions across three markets returned no AI Overview at all. Education and hospitality behave differently, and we found where.
Maher El Ouahabi · · 23 min read Across twelve categories in three markets, one domain spans them all and almost everything else is category specific. The gatekeeper list does not exist.
Maher El Ouahabi · · 15 min read Two conventions for talking to AI systems. The category adopted the one with no demonstrated reader and skipped the one with a specification.
Maher El Ouahabi · · 10 min read Four category buying questions, the organic top 20 of each. Seventeen ranked lists are published by a company selling one of the tools, at position one.
Maher El Ouahabi · · 22 min read Ten videos, eight channels, no vendor. Widened to 25 videos in 4 markets: 20 channels, 1 vendor, and no video or channel holding a slot in two of them.
Maher El Ouahabi · · 23 min read Nineteen buying questions, four markets, three AI surfaces, 46 runs. Of 408 cited sources, 58% appeared exactly once and about a third held across every run.
Maher El Ouahabi · · 20 min read Cloudflare's Bot Fight Mode challenges crawlers that are not on its verified list. PerplexityBot is not on it, and the usual test with curl cannot detect this.
Maher El Ouahabi · · 10 min read Rewording a hotel question in Spain left zero stable sources. Rewording a university question in the same country left four, and none of them is a university.
Maher El Ouahabi · · 13 min read One intent, four wordings, two markets. In Spain the stable share held between 30% and 48% while zero domains survived all four phrasings.
Maher El Ouahabi · · 12 min read Hotels and flights asked in four markets on three AI surfaces. The gap between the two questions runs from 3 points in Germany to 53 in the United States.
Maher El Ouahabi · · 11 min read We asked eleven buying questions three times each on AI Overview. Fewer than three in ten durably cited domains sell the product being asked about.
Maher El Ouahabi · · 18 min read Seven sectors asked who the best marketing agency is, on three AI surfaces with the cache bypassed. 133 domains came back, 92% in one sector only.
Maher El Ouahabi · · 11 min read AIclicks publishes monthly LLM response counts, which almost no vendor does. Two tiers check out to the digit. The $59 tier is 67% high.
Maher El Ouahabi · · 13 min read Same category, same market, same day. Asked which platform to buy, ChatGPT cited eight domains. Asked what the thing is, it cited zero.
Maher El Ouahabi · · 24 min read LLMrefs sells a single $79 plan covering 500 prompts and ten engines. Its own page gives two different country counts for the same product.
Maher El Ouahabi · · 11 min read Search Atlas defines a credit as one query on one platform once. The entry AI tier costs 2.85 times per credit what the tier above it does.
Maher El Ouahabi · · 14 min read Alhena sells an ecommerce AI concierge with visibility monitoring attached. Its table says All major engines and the count is in a collapsed FAQ.
Maher El Ouahabi · · 10 min read BrightEdge publishes no price on its own site. Its AWS Marketplace listing says $1,000,000 for 12 months. Five other public sources give five other numbers.
Maher El Ouahabi · · 12 min read Conductor sells enterprise AEO. Its pricing page has no prices, its entry tier has no AI credits, and no AI engine is named anywhere on it.
Maher El Ouahabi · · 15 min read Same question, same market, same moment, four surfaces. Forty-one of fifty-one cited sources appeared on exactly one of them. Google's own three barely overlap.
Maher El Ouahabi · · 10 min read Goodie's only priced plan is $399 for 3 engines. Its top card says 11 models, the table below lists 12, and all three tiers promise daily collection.
Maher El Ouahabi · · 10 min read Gumshoe costs $99 or $299 a month. Its comparison table scores it 11 models against rivals' 4 to 6, then its pricing table sells 6 and 4.
Maher El Ouahabi · · 10 min read 42 AI Overview measurements across six categories: stability ran from every source repeating to none. Re-measured a day later, one category flipped.
Maher El Ouahabi · · 15 min read HubSpot AEO is live in beta at $50 a month for 25 prompts on three engines, and HubSpot's own page credits the XFunnel acquisition as its origin.
Maher El Ouahabi · · 10 min read Knowatoa costs $59 or $199 a month. Its own FAQ refers twice to a question limit that appears nowhere on the page, and never names its seven services.
Maher El Ouahabi · · 10 min read Nightwatch publishes prompts and total AI responses on every tier. Divide them and every plan gives exactly thirty answers per prompt per month.
Maher El Ouahabi · · 10 min read Promptmonitor sells six AI engines for $29, the cheapest we verified. Its three published numbers disagree, and a second company shares its name.
Maher El Ouahabi · · 11 min read Rank Prompt sells up to 1,000 brands and 2,000 credits a month. Divide them and each brand gets two credits, which is two thirds of one full prompt.
Maher El Ouahabi · · 12 min read Rankability is back to three tiers at $199, $399 and $799 a month, metered on daily prompts. Fourth structure in twenty days, every reading recorded.
Maher El Ouahabi · · 13 min read ScalePost measures AI citations from publisher CDN logs and calls prompt sampling a guess. It is right about sampling and wrong that a fetch is a citation.
Maher El Ouahabi · · 12 min read SE Ranking's plans list five LLMs. The AI Search add-on that costs 72% more names four, and Gemini is the one missing. Prices verified 7 August 2026.
Maher El Ouahabi · · 12 min read Fifteen AI visibility tools that Google's AI never cited draw 151,387 US searches a month between them. The three it named that we can measure draw 7,864.
Maher El Ouahabi · · 14 min read ArcAI tracks nine AI engines, the widest list we have verified. All three of its packages say Ask for a Quote, on top of a platform starting at $2,500.
Maher El Ouahabi · · 10 min read Serpstat says 100+ AI models in its hero, 140+ in its FAQ, and names four. Free tier, then $39. It sells checks on one plan and credits on another.
Maher El Ouahabi · · 12 min read Similarweb's AI Search Intelligence is $99 a month for 150 prompts. It tracks four engines for visibility and four more for referral traffic, and says so.
Maher El Ouahabi · · 10 min read We read every source behind three AI answers. One category was answered entirely by independent reviewers, another mostly by vendors ranking themselves.
Maher El Ouahabi · · 9 min read Waikay costs $69.95 for 120 prompts. Its plan names ChatGPT/Google/Claude/Sonar, and our own measurement found Google's surfaces cite disjoint sources.
Maher El Ouahabi · · 10 min read We published 41% from a sample of seven. Eighteen vendors published both numbers and ten had no gap at all. Recounted properly, entry plans cover 65%.
Maher El Ouahabi · · 21 min read One question, four AI surfaces, 20 domains cited and not one shared by three of them. Twice, the agency named first was the publisher of the cited ranking.
Maher El Ouahabi · · 13 min read The same buying question in three markets cited 35 domains. Two appeared in all three. Of the 36 tools in our verified catalogue, five were cited at all.
Maher El Ouahabi · · 10 min read We counted the top 20 for GEO agency searches in four markets. In the US, thirteen of the rankings were written by an agency that named itself number one.
Maher El Ouahabi · · 20 min read HubSpot announced it was acquiring XFunnel on 31 October 2025. The standalone site still sells, with a free audit, an unpriced Enterprise tier and no middle.
Maher El Ouahabi · · 11 min read Across 36 runs, three markets and nine categories, YouTube appeared in eleven measurements and in every single run of eight. Reddit is pickier.
Maher El Ouahabi · · 10 min read Brand Radar is an add-on, not a plan. It is also the only tool here measuring Reddit, YouTube and TikTok, where our own runs found the citations.
Maher El Ouahabi · · 12 min read Two GEO tools showed 27,100 and 9,900 searches a month. Disaggregated they are 627 and 53. And disaggregation is not enough: three more names are not theirs.
Maher El Ouahabi · · 18 min read AEO is optimizing to be the answer, not a link. What separates it from GEO and SEO, what the controlled evidence supports, and what it does not.
Maher El Ouahabi · · 11 min read We asked three GEO buying questions. One cited six sources, two cited nothing at all. On those, no amount of content wins you a citation.
Maher El Ouahabi · · 12 min read Athena publishes its unit: one credit is one AI response. Run the arithmetic and $295 a month buys about fifteen prompts daily across eight engines.
Maher El Ouahabi · · 10 min read Bluefish raised $68M and sells to the Fortune 500. It covers agentic commerce, which few other tools here do. It also publishes no price at all.
Maher El Ouahabi · · 11 min read Brandlight raised $30M and its homepage claims a global ranking with no analyst, report, date or link behind it. What it covers, and what we verified.
Maher El Ouahabi · · 11 min read Doubao is the second-largest AI app in the world by users and almost no Western tool measures it. What that means if you sell into China.
Maher El Ouahabi · · 10 min read Cloudflare says it measures AI visibility from network signals rather than sampling. That solves one hard problem and cannot solve its own second metric.
Maher El Ouahabi · · 10 min read Cognizo's $499 plan bundles 150 prompts with eight AI-written articles. Its $899 plan states an article count and no prompt count at all.
Maher El Ouahabi · · 10 min read Otterly starts at $29 and names seven engines on its homepage. Four are in the plan and three are paid add-ons. What that costs, and what we verified.
Maher El Ouahabi · · 11 min read Profound raised $155M at a $1B valuation and is the category leader. Its entry plan tracks one engine. What that costs, and when it is the wrong buy.
Maher El Ouahabi · · 11 min read Quattr publishes more customer results than any vendor we have reviewed, with McAfee and Marriott named. Not one states a baseline. What that is worth.
Maher El Ouahabi · · 11 min read Rankscale starts at $20 and names ten engines on every plan, including Mistral. It bills in credits rather than prompts. What that actually buys.
Maher El Ouahabi · · 12 min read For eight days in June, 163 Google queries hitting our site carried an identical instruction preamble. A monitoring tool was leaking its own method.
Maher El Ouahabi · · 18 min read We asked which tool tracks ChatGPT visibility in the US, Spain and France. Not one tool was recommended consistently in more than one market.
Maher El Ouahabi · · 16 min read Semrush sells its AI Visibility Toolkit standalone at $99 a month per domain, not bundled. What it covers, how it compares, and a correction we owed.
Maher El Ouahabi · · 10 min read Every tool sells share of voice and none publishes how many runs are behind it. How it is calculated, and how many runs it takes.
Maher El Ouahabi · · 11 min read Five of our articles say only Evertune publishes its sampling. Siftly publishes enough to derive it, and reports confidence intervals. We were wrong.
Maher El Ouahabi · · 10 min read Superlines publishes its prices, and the arithmetic says its €379 tier is the worst value on the page. Ten engines, nineteen named customers, no sampling.
Maher El Ouahabi · · 12 min read Surfer's €49 plan has no AI tracker. Its €99 plan tracks ChatGPT weekly. Its €158 standalone tracker beats the €182 bundle. Every price needs a year.
Maher El Ouahabi · · 11 min read We asked AI for the best universities in Madrid and Paris. It named them every time and cited student housing, tutors and relocation agents instead.
Maher El Ouahabi · · 14 min read Wellows publishes its prices, which is rare here, but they are per domain. What that costs an agency, which engines it covers, and what we verified.
Maher El Ouahabi · · 9 min read Six runs on Google AI Overview asking which tool tracks ChatGPT visibility. Not one of among the best-funded vendors appeared. Here is what did, and why.
Maher El Ouahabi · · 14 min read We checked the 27 comparison articles Google returns for AI visibility tools. Twenty-two were published by a company selling one. Most put themselves first.
Maher El Ouahabi · · 13 min read Writesonic's tracker page says ten platforms. Its pricing page puts three on every plan you can buy and the other seven behind a custom quote.
Maher El Ouahabi · · 11 min read AirOps has raised about $118M and is among the best-funded player here. It is a content platform with visibility attached, which decides whether it fits you.
Maher El Ouahabi · · 14 min read Evertune samples each prompt up to 100 times per model, and Pro is $800 a month for 100,000 prompts across 11 models. What that buys, and what it does not.
Maher El Ouahabi · · 14 min read How an agency sells GEO without overpromising: the method, the deliverable, the pricing, and the four client objections you will meet.
Maher El Ouahabi · · 12 min read Peec AI raised $29.1M in under a year and published no price until August 2026. What it tracks, who it is for, and when a cheaper tool is the better call.
Maher El Ouahabi · · 13 min read Scrunch was acquired by Sitecore for $225M in June 2026. What that changes for buyers, what it now costs, and when a cheaper tool is the better call.
Maher El Ouahabi · · 14 min read Three European brands in three sectors collapsed in GPT-5.3 on the same day, while US brands held. A three-week EchoWi study of a new LLM failure mode.
Maher El Ouahabi · · 34 min read Step-by-step guide to making AI chatbots recommend your brand. Five strategies drawn from the research that survives scrutiny, and the limits of what it shows.
Maher El Ouahabi · · 38 min read 81 AI visibility tools, 80 you can still buy, with every price read from the vendor's own page in August 2026. One shut down.
Maher El Ouahabi · · 25 min read Buyers are skipping Google and asking AI for recommendations instead. The brands that AI mentions win. Here's how the shift works.
Maher El Ouahabi · · 12 min read GEO is how brands get cited by ChatGPT, Perplexity and AI Overviews. What the research proves, what it does not, and how to optimize a page.
Maher El Ouahabi · · 16 min read Cookies on this site
We would like to use analytics cookies to understand how our pages are used. Nothing is loaded until you choose, and the site works the same either way. Read the cookie policy