Track 10 free prompts in ChatGPT · No credit card needed

Oly, AI that finds you in AI, tracks every answer, watches your rivals, audits your site, plans your content, helps you rank higher.

Oly asks the questions your buyers are asking and reads the answers, then shows you where you appear, where you don’t, who shows up instead, and which sources helped shape the result and how you can get ranked higher faster.

Oly helps you become the biggest brand AI recommends in your category
ChatGPTGoogle AI OverviewGoogle AI ModeGeminiCopilotPerplexity
Trusted by teams at
  • Finoya.ai
  • elk
  • Digital Crew
  • Femino
Search has changed

You rank on Google. AI may still recommend someone else.

People are asking ChatGPT, Gemini and Perplexity what to buy, which company to use and which brand to trust. The answer is a shortlist, and it is the new search result.

If competitors are on it and you are not, your rankings only tell part of the story. keenoly tells the rest: who is named, why, and what the engines read to decide.

01

People ask AI

Whole questions, in six different engines, not keywords in one.

“What are the best running shoes for wide feet?”
02

Each engine picks a shortlist

A few brands are named, compared and recommended, and every engine does it differently.

1 HOKA2 Allbirds3 Brooks
03

It can be measured

Where you appear, where rivals win, which sources shaped it, and whether it moved since last run.

32%Discovery visibility↑ 6 pts vs last run
What keenoly tracks

Everything between the prompt and the recommendation.

Sixteen views of your AI visibility and the data around it, each answering one question, none of them blended into a single score.

Non-branded visibility

Discovery prompts never name you. How often do the answers name you anyway, and where do you rank against the brands they name instead?

Branded perception

Prompts that name your brand. Not whether you were mentioned, but whether the answer told the reader to choose you, and how it described you.

Competitive win rate

Head-to-head prompts against named rivals: won, lost, tied or conditional, and the decision factors the engines weighed.

Rankings by engine

One leaderboard per engine. Gemini and Perplexity rarely agree, and a single blended number would hide it.

Prompts and related AI prompts

Every question you track, plus the searches the engines ran to answer them and the questions they suggested next.

Sources and citations

The pages engines cite or consult, scored for influence, and which of them cite your competitors without naming you.

Findings

Each finding names the page or prompt, how important it is and why, and whether it is self-serve, worth an expert review, or one for our managed GEO team.

ChatGPT ads and shopping

Who is paying to appear beside ChatGPT's answers to your prompts, and which product cards it shows, by brand and by merchant.

Places and media

The locations, videos and images engines put beside their answers. Being named in the text and absent from the pictures is its own kind of invisible.

Technical and content audits

A GEO readiness score for your site with evidence on every finding, and a read on how much of your copy sounds AI-written.

Personas

Visibility by the buyer segment each prompt belongs to, weighted by share, so a brand cannot look healthy while invisible to the people who buy.

Reports and run history

Weekly and monthly reports with share links that work without an account, CSV exports, and every run kept exactly as it was scored.

Pages

Every page on your site, with how often AI answers cite it beside its Google clicks, conversions and health. See which pages earn answers and which only earn rankings.

Google Analytics and Search Console

Connect GA4 and Search Console to see the visits AI assistants send you, the queries you rank for, and the pages strong in Google but missing from AI answers.

Results after the fix

Mark a finding done and keenoly measures what changed: citations in AI answers, Google clicks and visits, against the site as a whole.

Executive briefing deck

A PPTX for your senior team in plain sentences, with speaker notes: the headline, the trend, who AI recommends instead, what moved, search and traffic, and the next steps.

Every answer, kept

Not a score. The actual answers, per engine, per prompt.

Each run stores what every engine said to every prompt: the text, the brands it named, the sentiment, the sources it cited and the searches it ran. Filter by engine, open any answer in place, and see exactly what a buyer saw.

Run your first scan
A run, shown for a sample brand: one line per engine under each prompt, the full answer opened in place.
Meet Oly

Ask your data a question. Get an answer, with a link.

Oly is the assistant in the corner of every page. It answers from your own tracked runs: why a number moved, which competitor is winning a prompt, what an engine actually said, which source to go after next. Every figure comes from your data, never estimated, and every answer ends with the page to open.

It also knows when an engine is locked on the free scan, when a run was partial, and when a drop is one bad run rather than a trend.

VisibilityRankingsCompetitorsAnswersSourcesActionsAds and shoppingAuditsPersonasCredits

10 questions on the free scan. Unlimited on every paid plan.

OlyAsk about Northwind Running
Sample
Which engines name us least, and why?
Perplexity and AI Overview, at 27% and 29% of discovery answers on the Oct 7 run. AI Mode names you most, at 44%. Most of the gap is the wide-fit prompts: there Perplexity cites a Runner's World guide and a Reddit thread that recommend Stridewell and never mention you.Discovery visibility by engineAI Mode44%ChatGPT41%Gemini38%Copilot37%AI Overview29%Perplexity27%Open Rankings by engine
Can your team take on the wide-fit pages?
Preparing a request for the GEO team
Send this to the keenoly GEO team?
  • Managed GEO: the keenoly GEO team does the work for you.
  • 2 findings: Strong in search, absent from AI answers; Competitor cited where you rank.
  • Someone replies by email, usually within one working day. Nothing on the site changes until you agree a plan.
How it works

From your website to AI visibility data in about a minute.

01

Add your brand

Your website and your market. keenoly reads the site and works out what buyers ask AI about your category.

02

Check the prompts

Six discovery, two branded, two comparison prompts, written for you. Edit any, add your own, name competitors, assign personas.

03

Every engine answers

ChatGPT answers first, in about a minute, while you watch. The other five engines answer the same prompts in the background.

04

Read, ask, act

The Overview, the answers, the sources and the actions. Ask Oly what changed. On a plan, it runs again on your schedule.

Prompt methodology

Three kinds of question, measured three ways.

A prompt that names your brand will always get your brand in the answer, so keenoly never mixes it with one that does not. Discovery prompts measure whether you are found. Branded prompts measure how you are described. Comparison prompts measure whether you win. Related AI prompts are the searches the engines ran on their own.

DiscoveryBrandedComparisonRelated AI promptsPersonasYour own prompts
Pricing

Start free. Track more as you grow.

Every paid plan includes all six engines, unlimited projects, prompts and competitors, scheduled tracking, technical and content audits, Google Analytics and Search Console, and unlimited Oly. Credits decide how much tracking you can run.

Free

$0
Free visibility scan
10 free ChatGPT promptsNo credit card requiredChatGPT only
  • 1 project
  • 10 prompts, every answer readable
  • One time visibility scan
  • Competitor insights
  • GA4 and Search Console: a one-time 60-day snapshot
  • Oly chat: 10 questions
  • 1 blog post written by Oly, to try it
Start free

Lite

$49/ month
Billed monthly
2,500 credits$0.0196 / creditCredits reset each billing cycle
  • Unlimited projects
  • Daily, weekly or fortnightly tracking
  • About 25 prompts across 3 engines, tracked daily
  • Competitor tracking
  • Citations and sentiment
  • Technical and content audits
  • GA4 and Search Console included: 12 months, updated weekly
  • Oly chat, unlimited
  • Executive briefing deck (PPTX)
  • 5 blog posts a month, written by Oly and sent to WordPress
Choose Lite

Elite

$299/ month
Billed monthly
25,000 credits$0.0120 / creditCredits reset each billing cycle
  • Unlimited projects
  • Daily, weekly or fortnightly tracking
  • Everything in Growth
  • About 130 prompts across all 6 engines, or several projects
  • GA4 and Search Console included: 12 months, updated daily
  • Oly chat, unlimited
  • Executive briefing deck (PPTX)
  • 20 blog posts a month, written by Oly and sent to WordPress
Choose Elite
1 credit = 1 AI answer
Free AI visibility scan

Curious where your brand ranks?

Ten prompts your buyers ask, answered by ChatGPT, every answer readable, in about a minute. Ten questions to Oly included. More engines and tracking over time unlock on a plan.

Start free No credit card required · Results in about a minute
Your free scanComplete
Discovery visibility32%
Ranked #2 of 8 brands
10 prompts answeredChatGPT, every answer readable
+
5 more engines answeredGemini named you in 7 of 10. Unlock to read them
Locked
+
Track over timeDaily, weekly or fortnightly on a plan
Locked
FAQ

Clear answers before your first scan.

Start with ChatGPT for free. Upgrade when you want the other five engines and tracking that keeps running.

Find out how to rank higher in AI.

Your first 10 ChatGPT prompts are free.

Start for freeNo credit card required

keenOly: the facts

A plain summary of what keenOly is, what it measures, what each plan includes and what it costs, written for people and for AI assistants. Prices and credits come from the same source as the pricing on the keenOly homepage.

What keenOly is and who it is for

keenOly measures whether and how AI engines mention, describe, compare and recommend a brand, and which sources they cite to do it. It asks each engine the questions buyers ask, keeps every answer, and shows where the brand appears, where it does not, and which companies are named instead.

It is for brands and their marketing teams, and for agencies that track several clients: an agency workspace groups projects by client and can send reports under its own name.

Oly is the assistant inside keenOly. Oly answers questions about the account’s own data and can carry out a few actions after the person confirms.

Software and managed GEO services

keenOly is software the customer uses themselves, with a GEO team behind it. Each finding the product raises says who should handle it:

  • Self serve: the steps are given and the customer makes the change.
  • GEO Services: a person on the keenOly GEO team checks the change before it is made.
  • Managed: the keenOly GEO team does the work.

Supported AI engines

keenOly tracks six engines: ChatGPT, Google AI Mode, Google AI Overview, Microsoft Copilot, Google Gemini and Perplexity.

On the free scan all six answer the same prompts, and the ChatGPT answers are readable; the other five unlock on a paid plan or for 50 earned credits. On a paid plan a project can track any of the six, and the plan’s credits decide how many prompts and engines it runs.

Discovery, branded and comparison prompts

A prompt that names the brand will always get the brand in the answer, so keenOly keeps the three kinds of prompt apart and measures each one differently.

  • Discovery: names neither the brand nor a competitor. It measures whether the brand is found: visibility (the share of answers that name it), share of voice and rank.
  • Branded: names the brand. It measures how the brand is described: sentiment, and whether the answer recommends it.
  • Comparison: names the brand and a tracked competitor. It measures whether the brand wins: won, lost, tied or conditional, and the factors the engine weighed.
  • Related AI prompts: the searches the engines ran on their own while answering.

What the product does today

  • Answers: every answer from every engine, with the brands it named, the sentiment, the sources it cited and the searches it ran.
  • Rankings by engine: one leaderboard per engine, never blended.
  • Companies Cited: the competitors being tracked, and every other company the engines name.
  • Sources and citations: the pages engines cite or consult, and which of them cite competitors without naming the brand.
  • Pages: each page of the site with how often AI answers cite it, beside its Google clicks and visits.
  • Actions: findings grouped as citation, technical and content work, each with its importance and the steps to take; a finding marked done is measured afterwards.
  • Technical audit and content auditor: how readable the site is to AI engines, and how much of its copy reads as machine written. Paid plans can crawl the whole site.
  • Content on Autopilot: Oly suggests posts for the questions where the brand is missing, writes them in the brand’s voice, checks them, and can save them to WordPress as drafts or schedule them; posts are then tracked in Google and in AI answers.
  • Google Analytics 4 and Search Console: visits from AI assistants, search queries, and pages strong in Google but missing from AI answers.
  • Shopping, places and media: the product cards, places, videos and images engines show beside their answers, where they show them.
  • Reports: weekly and monthly reports, share links that work without an account, CSV exports and, on paid plans, an executive briefing deck.
  • Teams: owners, admins and members, with members limited to chosen projects if needed.
  • Integrations: Google Analytics 4, Google Search Console and WordPress.

How onboarding and analysis work

  • Add the brand: the website and the market (country and language). keenOly reads the site and drafts a summary, the category and likely competitors.
  • Check the prompts: keenOly writes discovery, branded and comparison prompts; the customer edits them, adds their own and names competitors.
  • Run: the engines answer the prompts. On the free scan the ChatGPT answers arrive first.
  • Read and act: the overview, the answers, the sources and the actions; Oly explains what changed. On a paid plan the project runs again daily, weekly or fortnightly.

What the free scan includes

No card is needed.

  • 10 prompts: 6 discovery, 2 branded and 2 comparison when competitors are named.
  • All six engines answer; the ChatGPT answers are readable, and the other five unlock on a paid plan or for 50 earned credits.
  • 10 questions to Oly.
  • A technical audit and a content audit, starting from the homepage; the whole site crawl is on paid plans.
  • A one time Google Analytics and Search Console snapshot of the last 60 days.
  • 1 post written by Oly.
  • One free scan per workspace, per person and per website.

Plans, prices and credits

Plans, prices and key limits
PlanMonthlyAnnualCreditsKey limits
Free$0Not offered10 free ChatGPT promptsOne free scan: 10 prompts, with the ChatGPT answers readable; 10 questions to Oly; Audits read the homepage; Google Analytics and Search Console: a one time snapshot of the last 60 days; 1 post written by Oly, then 250 credits a post
Lite$49 / month$490 / year2,500 credits a month; 30,000 credits a yearDaily, weekly or fortnightly tracking; Google Analytics and Search Console: 12 months, updated weekly; Whole site crawl for the technical and content audits; Up to 5 active findings shown, and the top 50 pages; 5 posts written by Oly a month; Executive briefing deck
Growth$149 / month$1,490 / year10,000 credits a month; 120,000 credits a yearDaily, weekly or fortnightly tracking; Google Analytics and Search Console: 12 months, updated daily; Whole site crawl for the technical and content audits; Every finding, with history; 10 posts written by Oly a month; Executive briefing deck
Elite$299 / month$2,990 / year25,000 credits a month; 300,000 credits a yearDaily, weekly or fortnightly tracking; Google Analytics and Search Console: 12 months, updated daily; Whole site crawl for the technical and content audits; Every finding, with history; 20 posts written by Oly a month; Executive briefing deck
  • Prices are in US dollars. Annual plans cost the same as ten months.
  • One credit is one successful answer from one engine. A run costs the number of prompts multiplied by the number of engines; failed answers are not charged. The free scan is paid for by keenOly.
  • Monthly credits reset each billing cycle and do not roll over. Annual plans receive their annual credits upfront.
  • Every paid plan includes unlimited projects, prompts, competitors and team members, and any of the six engines. The credits decide how much tracking runs.
  • There are no add on credit packs. Credits can also be earned by linking to keenOly, posting about it or referring a business; earned credits last six months.
  • For more than Elite covers, email hello@keenoly.com.

What Oly does

Oly answers questions about the account’s own tracked runs, such as why a number moved, which competitor is winning a prompt, what an engine said or which source to go after next, and links to the page with the figure. Oly works from the account’s data and says when a run was partial or an engine is locked.

Oly can also propose actions: activating prompts, tracking a competitor, drafting a content brief and sending a request to the keenOly GEO team. Nothing happens until the person confirms.

  • Free plan: 10 questions in total.
  • Paid plans: no question limit while the plan is active, with a fair use limit of 60 questions an hour.

How to request managed GEO support

In the product, press Request GEO Services in the header, or Request GEO Services on a finding, and describe what is needed; the request goes to the keenOly GEO team and someone replies by email. Oly can prepare the same request in a chat. Without an account, email hello@keenoly.com.

What the measurements can and cannot tell you

  • Engines answer the same question differently, so each engine is reported on its own and never averaged together.
  • Answers can change from one run to the next. Each run is a sample of what the engines said at that moment.
  • Results are for the market the project tracks (country and language); answers elsewhere can differ.
  • Visibility counts whether the brand is named, not where in the answer it appears.
  • Changes are compared only between runs with the same engines and prompts. A first run has nothing to compare against.
  • Failed answers are left out of every figure. Few answers on a prompt make its figures less certain.
  • Shopping cards and places appear only for some questions and markets.
  • keenOly measures and suggests. It does not guarantee rankings, recommendations, traffic or revenue, and a change after a fix is reported as a change, not as proof of cause.

Contact and links

  • Sales, questions, demos and GEO services: hello@keenoly.com
  • Support: support@keenoly.com