HomeKoalr
The product · 17 capabilities

Everything Koalr measures, and what you do with it.

Koalr is a generative engine optimisation (GEO) platform. It puts real buyer questions to seven AI answer engines on a schedule, records which brands each answer names and which pages it cites, turns the prompts you lose into a ranked list of changes, then drafts and schedules the content. Every capability is defined below: what it is, what it measures, what you do with it, and which plans include it.

Your AI Visibility Score in under 5 minutes. No card required.

The AI answer engines Koalr monitors for your brand

How we do it · the Koalr Framework your content. Strengthen your . Own the . Enjoy the .Every capability on this page sits in one step of it. Here is the whole loop, ring by ring.How we do it

On this page

01

Monitor

Koalr puts your buyers' questions to the engines on a schedule and keeps every answer, so a change is something you can date rather than something you noticed.

AI Visibility Score

A single 0–100 rate: how often the AI engines name your brand across the prompts you track.

Every answer collected for your tracked prompts. The score is the percentage of those answers that name your brand — a rate, never a blend of other metrics, so a fall in it means one thing and can be traced to the prompts that caused it. Your dashboard shows the same number as Prompt Visibility Rate.

Watch one number between runs, then open the prompts underneath it to see which answers changed and which engine changed first.

On every plan, including Starter

Prompt Explorer

The set of real buyer questions Koalr puts to the engines on your behalf, and the record of who each one named in reply.

Tracked prompts written the way a buyer actually asks, each run live on all seven engines: 25 per site on Starter, 40 on Growth, 50 on Business. Prompts are counted per site rather than drawn from one account-wide pool, so a busy brand cannot quietly starve a quiet one. Every run records the brands named, the order they appeared in, and the pages cited.

Edit the starting set or add your own sales questions, then read answer by answer to find the prompts where a competitor is named and you are not. Where an engine exposes them, Koalr also records the background searches it ran before writing its reply — the phrasing it used to research your question, which is the phrasing your page has to match.

On every plan, including Starter

AI Traffic

Which AI crawlers read which of your pages, set against what those same engines went on to cite, and the visits AI assistants then send you.

Crawler visits by bot and by page, joined to the citation record, plus the sessions that arrived from an AI assistant, attributed to the assistant that sent each one and the page it landed on. The state worth finding is crawled-but-never-cited: the engine has read the page and chosen not to use it.

Treat a crawled-but-uncited page as a rewrite rather than a promotion problem — the engine already has it and is passing it over. A page that is cited but never visited is being read on your behalf.

On every plan, including Starter

Seven AI Engines

Every answer Koalr records is collected from all seven major AI answer engines, on every plan.

ChatGPT, Claude, Gemini, Perplexity, Microsoft Copilot, Google AI Overviews and Google AI Mode. Google's two AI surfaces are tracked separately because they answer independently and often name different brands for the same question. Six refresh daily; Claude refreshes weekly as standard, or daily with the optional Claude Daily add-on.

Compare the same prompt across engines. An answer you win on one and lose on another is a different problem from one you lose everywhere.

On every plan, including Starter

Email Digests

An email of what changed since the last one, daily or weekly, or off.

Movement between runs — how your visibility rate moved, and the prompts that moved it most.

Find out the week it turned instead of the quarter it turned, without anyone having to remember to open the dashboard.

On every plan, including Starter

02

Analyse

Koalr reads the record for you: who else was named, how you were described, which pages were quoted, and how your own site reads to a machine.

Share of Voice & Competitor Benchmarking

Your named rivals scored on the same prompts as you, so the comparison is like for like rather than two separate reports.

Share of Voice is your brand's mentions as a percentage of all the brand mentions in the answers to your tracked prompts, one slot per brand per answer. You can track 5 competitors per site on Starter, 15 on Growth and 20 on Business, each read out of the same answers as you. It answers how often, of all the times an engine named someone in reply to your questions, the someone was you.

Open the AI Visibility Leaderboard to see who wins each individual prompt, then take the prompts you lose into Actions rather than working from an aggregate.

On every plan, including StarterHow Share of Voice is counted →

Sentiment

How each engine describes your brand on the occasions it does name you, scored per engine rather than averaged into one figure.

Every mention carrying a clear signal is scored, labelled positive, neutral or negative, and the scores averaged 0–100. Mentions with no signal are excluded from the average instead of dragging it down, and the sentiment score reads Unknown rather than zero when nothing carried a signal. Because engines rarely say anything wholly negative, Koalr also reads the caveats sitting inside otherwise positive answers — the limitation, the drawback, the 'not suited to' — and clusters them into named themes.

Read the themes rather than the score. Each one links to the answers that raised it, so an objection an engine keeps repeating becomes a specific page to write.

On every plan, including StarterHow sentiment is scored →

Sources & Citations

The pages each engine actually drew on to write an answer, listed per engine.

Every cited URL behind every answer, split into pages you own and pages you do not, and typed by what they are — media, forums, roundups, directories. A mention and a citation are separate outcomes: a mention means the engine knows your name, a citation means it drew on your page. You can have either without the other.

Find the off-site pages that keep getting quoted back at your buyers — the directory, the roundup, the forum thread — and work on the ones you can influence. Across 267,963 citations in Koalr's off-page research, 84% pointed at pages the brand did not own.

On every plan, including StarterWhat Koalr does with a citation →

GEO Audit

A crawl of your own pages that scores how well each one is built to be quoted by an AI engine, with a site-wide check of whether AI crawlers can get in at all.

A Site GEO Score from 0 to 100, the average of every scored page. Each page is scored on four of the five CRAFT checks, Recognition, Answers, Format and Trust: Answers counts the most towards the score, then Trust and Format, and Recognition the least. The score is read from your raw HTML and text with no AI calls in the loop, so the same page scores the same way twice, and it means built to be quoted, never will be cited. The fifth check, Crawlability, is site-wide and sits beside the score in the audit's Agent Experience section.

Work the page list rather than the headline score: it names the specific pages holding the average down and what is wrong with each.

On every plan, including Starter

Agent Experience

A section of the GEO Audit that shows whether AI agents can reach your site at all, and whether they can tell who you are.

Two site-wide checks, each a plain count rather than a score. The crawler check reads your robots.txt and shows how many of the AI answer crawlers it lets in, as in 6 of 7 let in, naming any it blocks. The identity check reads your homepage for the business identity block, the one block of structured data that names the business behind the site, and shows how many of its four details are in place, as in 3 of 4 in place. Neither feeds the GEO score: they sit beside it, because a blocked answer crawler keeps you out of that engine’s answers however well the page is written.

Fix a blocked answer crawler first: it is the one failure that makes you invisible on its own, and it shows in no other report. Then complete the identity block. Re-check either from the section once the change is live.

On every plan, including Starter

Kevin

An in-app assistant that answers questions about your own visibility data in conversation.

It reads your metrics, prompts, citations, competitors and actions, and streams the answer with links to the underlying records.

Ask what moved this week instead of assembling the answer from four dashboards yourself.

On every plan, including Starter

GA4 & Search Console

Read-only connections to your Google Analytics and Search Console properties, authorised through Google OAuth.

Your existing search and traffic data, sat next to the AI visibility record for the same pages. Read-only: Koalr never writes to either property.

Report AI visibility in the same review as organic search, rather than as a separate exercise nobody connects to the rest. The Analytics connection is also what feeds AI Traffic, so a visit an assistant sent is counted as one.

On every plan, including Starter

03

Act

Koalr turns the record into work: what to change, in what order, drafted for you to edit, and reported to whoever is paying for it.

Actions

A ranked list of the changes most likely to move your visibility, ordered by measured impact with the effort weighed in.

Each recommendation is scored by the visibility effect its kind of fix has measurably had, with the cheaper fixes and the ones touching more pages lifted within that, and the list is sorted by that score.

Work top-down. A list of everything wrong with a site is not a plan; the ordering is the product.

On every plan, including Starter

Content Studio

The half that writes: it drafts against a specific prompt you are losing and the page currently winning it.

A monthly allowance of content briefs for the whole workspace: 10 on Starter, 20 on Growth and 20 on Business, with a brief pack adding 10 more a month on any of them. Each brief is tied to a named prompt and the competing page, and citations are tracked after publish so you find out whether the piece worked.

Turn a losing prompt into a draft in the same session you found it, then check back for whether the engines started citing it.

On every plan, including Starter

Content Scheduler

Planning and publishing dates for the work the Studio drafts, inside the same tool.

A pipeline of planned and scheduled pieces, so the queue is visible rather than living in someone's calendar.

Schedule from the Studio and skip the handover to a second tool, which is where visibility work usually stalls.

On every plan, including Starter

MCP Server

A Model Context Protocol endpoint at app.koalr.ai that lets your own AI assistant query your Koalr data directly.

A set of read-only tools over your visibility record: workspaces, sites, the visibility summary, prompts, competitors, citations, gaps, actions, AI traffic, content and site health. Read-only by design: an assistant can ask, and cannot change anything.

Point Claude or another MCP client at it and ask about your own visibility inside whatever tool you already work in.

On every plan, including Starter

White-label & Client Reporting

A workspace and login of their own for each client site, with your agency's name and logo on the setup document a client signs off.

A login scoped to one client's site, your logo in place of Koalr's on the client sign-off document, and a pitch workspace for each prospect you have not signed yet.

Hand a client their own login rather than exporting a deck each month, and run a prospect's numbers once, unbilled, before the pitch.

On the Agency and Enterprise plansHow client workspaces work →

What it costs

Depth changes by plan. Access mostly does not.

All seven engines are on every plan. What moves as you go up is how much you can watch: 25 tracked prompts per site on Starter, 40 per site on Growth and 50 per site on Business, with 5, 15 and 20 tracked competitors alongside them. Client logins and agency-branded sign-off documents are on the Agency and Enterprise plans. Full detail is on the pricing page.

  • ChatGPT
  • Claude
  • Gemini
  • Perplexity
  • Copilot
  • AI Overviews
  • AI Mode
Koalr

Starter

25
tracked prompts per site5 tracked competitors

Growth

40
tracked prompts per site15 tracked competitors

Business

50
tracked prompts per site20 tracked competitors
FAQ

What Koalr does, answered.

Still wondering? Run your free scan.

Koalr is a generative engine optimisation (GEO) platform. It measures how AI answer engines describe your brand, and does the work to change it. It puts a set of real buyer questions to ChatGPT, Claude, Gemini, Perplexity, Microsoft Copilot, Google AI Overviews and Google AI Mode on a schedule, records which brands each answer names and which pages it cites, turns the prompts you lose into a ranked list of changes, then drafts and schedules the content those changes call for. Every capability is on every plan.

Free AI visibility scan

Find out what AI says about you.

Scan your site free and get your AI Visibility Score, plus your first ranked fix list, in minutes. No card, no demo call.

One scan, 25 real prompts, 7 AI engines, under five minutes, zero setup.

Share of Voiceyou vs. rivals
0–100AI Visibility Score
Actionsyour next moves, ranked