AI Visibility and GEO: The 22 August 2026 Six-Engine Shortlist

Fifteen of the 79 AI-visibility software products counted in the CitedIndex index on 22 August 2026 were tagged for all six named answer engines and categorised for both monitoring and optimization. The 22 August edition recorded eight of those fifteen as advertising a stated-length free trial: Rankshift, Omnia, Qwairy, Goodie AI, RankScale, Writesonic, Mentionable and AIclicks. The rest of this page is how the filter gets to fifteen, and what it throws away.

The original 22 August 2026 revision said it re-derived these figures against the live index, replacing an 18 August edition with 92 entries rather than 95. A retained warehouse snapshot, collector run 120415 observed on 21 August 2026 at 21:59 UTC, independently preserves the same 95-entry / 79-tool population and reproduces the category and six-engine tag counts below. It is corroborating historical evidence, not a claim that this was the original author’s query run. The trial and price observations remain the dated article’s records; that warehouse population alone does not verify them.

What “does both” actually narrows down

The useful distinction here is the recorded category overlap, not proof that either job is performed well. The 22 August edition counted 95 published entries, of which 16 sit in our citation-services category: managed programmes somebody else runs for you rather than software you operate. That leaves 79 software tools, and those 79 are the denominator for every number below.

Of those 79 tools, 76 are categorised for AI visibility monitoring, 47 are categorised for GEO and content optimization, and 44 sit in both categories at once. The three figures reconcile exactly: 76 plus 47 minus 44 is 79, and no tool in the index sits outside both categories.

So 44 of 79 were filed under both categories in that taxonomy snapshot; classification is not a demonstrated optimization outcome. Asking a vendor whether it measures and optimises is not a shortlist question, because almost everyone answers yes. The filter that actually bites is engine coverage.

The filter that bites is Microsoft Copilot

We tag six named answer engines. Among the 77 tools whose coverage we could determine, ChatGPT is claimed by 76, Perplexity by 69, Gemini by 65, Google AI Overviews by 57, Claude by 52, and Microsoft Copilot by 31. Copilot is the one that thins a shortlist: it is the only engine of the six that fewer than half these tools track.

Only 19 of the 79 tools carry all six. Intersect that with the scope filter and you get the number worth having: fifteen tools carried all six engine tags and both monitoring and optimization categories.

The three tools we added since 18 August make the point better than the totals do. Indexly arrived carrying five of the six and misses only Copilot. Distribb carries three. Neither joins the nineteen, which is why the six-engine count has not moved while the corpus grew by three.

Four six-engine tools drop out on scope rather than on quality. Profound, AI Sightline, LLM Pulse and LLMrefs all carry the full six-engine set, but we file them under monitoring and rank tracking rather than optimization. If measurement is all you need, they belong on your list and this page’s filter is the wrong one for you.

The eight trials recorded on 22 August 2026

The trial findings are withdrawn as verified research. We cannot reconstruct the source-to-row historical inputs for the eight-with-trials / seven-without-trials split or the longest-trial finding. The following tables preserve what the 22 August 2026 edition reported, including prices and plan names; neither the archived article nor the recovered taxonomy snapshot establishes that those offers existed on that date. They are not current offers or proof that signup was exercised. A source correction on 5 September found that RankScale’s pricing page now leads with Pro, not the $20 Essentials tier recorded here. The current page does not establish when the plan changed.

Tool	Entry price	Trial	Notable
Rankshift	$82/mo Starter	30 days	9-engine tracking paired with AI-crawler log evidence (rankshift.ai)
Omnia	€79/mo Growth	14 days	Geo-located browser checks, agent drafts content to close gaps (useomnia.com)
Qwairy	€65/mo Starter, billed annually	14 days	EU-hosted, ships an MCP server on the entry tier (qwairy.co)
Goodie AI	$399/mo Explorer	7 days	Tracks AI shopping assistants such as Amazon Rufus (higoodie.com)
RankScale	$20/mo Essentials	7 days	Cheapest entry point of the eight, at roughly 120 credits a month (rankscale.ai)
Writesonic	$79/mo Starter, billed annually	7 days	Content generation with a visibility suite alongside it (writesonic.com)
Mentionable	€79/mo Growth	4 days	All engines on the entry tier, native GA4 integration (mentionable.ai)
AIclicks	$59/mo Starter	3 days	3 of 6 engines on the entry tier (aiclicks.io)

In the recorded table, the lowest dollar-denominated entry price and the longest trial belonged to different products. $20 is about one twentieth of $399, not one fifth as this article previously said; the amounts also buy different allowances and engine coverage. Rankshift’s recorded 30 days is ten times AIclicks’ 3 days. A short trial may still be enough to inspect output and exports, but judging change over time requires knowing the product’s scan schedule and allowing enough repeated observations. We did not measure a category-wide minimum scan cycle.

Historical 22 August plan observation, not re-verified for current purchase: The eight also split by currency in a way a buyer notices at renewal. Five price in dollars, spanning $20 to $399 a month at the entry tier, and three price in euros in a much tighter band of €65 to €79. That is eight data points, which is too few to say anything about the market, but it is enough to say you should check which currency you are being quoted in.

The seven without a trial recorded in that edition

The other seven of the fifteen cleared the same historical tag and scope filters but had no free trial recorded: Brandlight, Cognizo, Conductor, EdenRank, Evertune, Promptwatch and Rankability.

Historical 22 August plan observation, not re-verified for current purchase: No trial does not always mean no self-serve. Rankability publishes four tiers you can buy directly, from $99 to $799 a month, with no trial and no free tier attached to any of them. EdenRank runs the opposite arrangement, a free-forever tier on three engines with no trial on any paid plan.

Historical 22 August plan observation, not re-verified for current purchase: Among the rest, the distinction that matters is whether a number is published at all. Cognizo lists $499 a month for its Platform tier and Evertune lists $800 a month for Pro, and both still route the signup through a demo. Brandlight discloses no pricing on any tier, and Conductor sells custom enterprise contracts only. If your purchase needs a figure before a meeting, that split matters more than the engine count does.

Six engines does not mean six engines on the cheapest plan

Historical 22 August plan observation, not re-verified for current purchase: Our engine tags describe the product, not the tier, and for several of the fifteen the gap between those two is the whole buying decision. Four of the eight trial tools name fewer than six engines on their cheapest plan. Writesonic’s $79 Starter names ChatGPT, Gemini and Google AI Overviews, with the other three on Enterprise. Goodie AI’s $399 Explorer names three engines and puts Copilot on Pro, which is quoted after a demo. AIclicks’ $59 Starter tracks three of six, and Omnia’s €79 Growth tier gives you four default engines.

Historical 22 August plan observation, not re-verified for current purchase: The other four publish their engine count at the entry price and do not move it: Mentionable names all seven of its engines on the €79 Growth plan, RankScale sells access to 17-plus engines on the $20 Essentials tier, Qwairy gives Starter ten AI providers, and Rankshift tracks nine engines from its free trial upward. Outside the trial group the same split holds, and Cognizo is the sharpest case: its $499 Platform tier covers five of the six, with Claude reserved for a custom Enterprise contract. If six-engine coverage is the reason you are shortlisting a tool, read the entry tier’s feature list before you read the headline.

What we could not determine

The retained snapshot has three of the 79 tools without any of the six engine tags. The original edition interpreted them as three different situations; those explanations were not re-tested here. GEOrank is a determined zero: it is a page-diagnostic workbench rather than an engine monitor, so nothing is missing from its record. AI Visibility Report Group sells one-time audits without naming which engines it reads, so its coverage is genuinely undetermined. Elmo published on 20 August and names its engines publicly, but we have not recorded them in our tags yet, so we are counting it as undetermined rather than guessing.

On those original editorial interpretations, engine coverage was treated as determined for 77 of 79. The snapshot itself verifies tags, not the judgment distinguishing true zero from unknown. Of those 77, ChatGPT is tracked by 76 and Copilot by 31. An undetermined field is not a zero, and rolling the two together would understate every share on this page. All three untagged tools are excluded from the nineteen and from the fifteen.

What can be reproduced, and what cannot

The historical category and engine counts are reproducible from archived CitedIndex observation 120415: published listing records, their category and engine-tag assignments, and the labels used in that same observation. Match records by observation and listing identity. The live directory cannot reproduce an old population merely by applying the same filters today. Price, trial and entry-tier details require their own historical vendor-page evidence; the retained article revision is a record of what we published, not proof that each vendor offered it.

Recovered taxonomy measurements

These findings describe stored directory tags in the separately observed 21 August snapshot, not tested product capabilities or the original author’s execution run:

• 95 published listings are preserved in warehouse collector run 120415.
• 16 published listings carry the citation-services category in warehouse collector run 120415.
• 79 published listings remain after excluding citation-services in warehouse collector run 120415.
• 76 of those software listings carry the visibility-monitoring category in warehouse collector run 120415.
• 47 of those software listings carry the geo-optimization category in warehouse collector run 120415.
• 44 of those software listings carry both categories in warehouse collector run 120415.
• 19 of those software listings carry all six named engine tags in warehouse collector run 120415.
• 15 of those software listings carry both categories and all six named engine tags in warehouse collector run 120415.
• 4 of the all-six-engine listings fall outside the both-category intersection in warehouse collector run 120415.
• 76 of those software listings carry the ChatGPT tag in warehouse collector run 120415.
• 31 of those software listings carry the Copilot tag in warehouse collector run 120415.
• 3 of those software listings carry none of the six named engine tags in warehouse collector run 120415. The AI visibility monitoring and GEO and content optimization categories are the two halves whose overlap gives the 44. The six-engine collection and the free-trial collection apply the same two filters as browsable pages, though each collection also applies its own editorial checks, so its membership is not always identical to the raw tag filter used here.

The caveat we would flag hardest is that engine coverage is the fastest-moving fact in this category. Vendors add and quietly drop engines between pricing-page revisions, and we have twice had to correct a tool’s Copilot claim after re-reading the vendor’s live page. Read the six-engine set above as historical index tags corroborated by the 21 August snapshot, not independently tested engine behavior or a current recommendation. Check the vendor’s own pricing page and the exact engine surface before you sign.

If you want the whole category ranked rather than this intersection of it, our best AI visibility tools page compares all of them, and how to choose an AI visibility tool is the checklist version of this decision.

5 September 2026 editorial correction: recovered a separately timestamped historical taxonomy run, corrected the $20/$399 comparison, and removed current-trial and scan-cycle assurances. The original publication date and 22 August census date are unchanged. No trial signup, product execution, new census or full vendor re-verification was performed.