In development · Early access for Gulf brands and agencies
Know where AI answers rank you. In Arabic and in English.
Track how ChatGPT, Gemini, Google's AI answers, Perplexity, Copilot and Claude name, rank and cite your company. Market by market, language by language, with the margin of error on every number.
From MxD Digital, for brands and agencies in the UAE and Saudi Arabia.
Theme view · Lumera Dental
عيادة أسنان في دبي
Named in 20 of 20 questions · Mention rate 42%
2.4 (2.2–2.7)No clear changeزراعة الأسنان دبي
Named in 20 of 20 questions · Mention rate 59%
1.8 (1.5–2.0)No clear changeتبييض الأسنان في دبي مارينا
Named in 20 of 20 questions · Mention rate 42%
2.3 (2.0–2.5)Moved up
Why it matters
Buyers ask AI for a shortlist. You need to know whether you're on it.
Ask an assistant for the best clinic, bank or agency in Dubai and it names a handful of companies, in an order. That list shifts from one run to the next, differs between engines and differs again between Arabic and English. One screenshot tells you very little.
- 01
Answers change between runs
Ask the same question twice and the companies named often change. Only repeated runs give a picture you can act on.
- 02
Arabic is its own market
People in the Gulf search in Arabic and in English, and engines answer each differently. We measure both and never mix them.
- 03
Engines disagree
ChatGPT, Gemini and Google's AI Overviews draw on different sources and name different companies. Each gets its own reading.
Sample workspace
Track themes the way your customers search.
A theme is what a customer is looking for, such as “dental implants Dubai” or “زراعة الأسنان دبي”. Each is asked as 20 questions every Saturday, three times on every engine. Explore a fictional dental group: switch market, language and engine, open a theme, and see who else is named and which sources the engines cite.
Lumera Dental
Customers see ranks, competitors and cited sources. Answer text stays internal.
Loading the sample workspace…
What you learn
Four questions, answered every week.
Where do we stand?
Your average position and how often you are named, per theme, engine, market and language. Each figure has its range, and an arrow appears only when a change is real.
How it works in the productWho is named instead of us?
Every company the answers put forward for a theme, tracked or not, ordered in tiers so that statistical ties are shown as ties.
How it works in the productWhich sources shape the answers?
The domains engines cite, split into yours, competitors' and third parties', with the directories and publications that cite competitors but not you.
How it works in the productCan engines read our site?
An audit of crawler access, indexing and server-rendered content: the technical prerequisites the evidence supports, each with a link to the engine's own documentation.
How it works in the product
How it works
From a domain to results you can defend.
- 1
Start with your domain
About 90 secondsEnter your website and a work email. Quick research reads your public pages and proposes competitors, markets and themes, with a progress view that shows each real step.
- 2
Approve your themes
Your decisionEdit five or six recommended themes for each market and language on one screen. Gulf workspaces see Arabic first. Nothing runs until you confirm.
- 3
Questions, sent as written
About a minuteEach theme becomes 15 to 25 questions, phrased the way customers ask. They reach every engine exactly as written: no added instructions, personas or formatting.
- 4
Collected every Saturday
WeeklyThree runs per question on every engine, in each market's local time. A first look, shown as counts, arrives within minutes of confirming your themes.
- 5
Ranks with their ranges
Every weekTheme and question ranks, mention rates, competitors and cited sources, each labelled with how much the data can support.
- On paid plans, deeper organisation research runs in the background and suggests further themes for you to accept or reject.
Collection feed· Sample replay
What a collection day looks like: engine, pair, question, companies found. Never answer text.
Measurement you can defend
Every number carries its range, its label and its source.
AI answers are variable by nature. We treat them as a sample to be measured honestly, not a leaderboard to be screenshotted.
Never blended
Engines, markets, languages and collection methods are reported separately. No single score hides the differences.
Averages, not one-off positions
Positions are averaged over repeated runs and shown with their range. A single run's position is never presented as your rank.
Labelled confidence
Results are marked Exploratory, Directional or Decision-grade, with the reasons attached, using labels modelled on the IAB's 2026 guidance for measuring AI visibility.
Only real changes
A movement is flagged only when it is statistically significant and larger than the noise we measure between identical runs.
Provenance on every answer
Each answer records whether it came from the consumer interface or an official API. The two are never combined: API answers can differ substantially from what people see.
Shown with a method
A public methodology explains how every figure is produced, including what we cannot measure yet.
Average position
2.4(2.2–2.7)
95% range
Named in 14 of 20 questions
ExploratoryLower positions are better. Ranges narrow as questions and weeks accumulate.
For agencies
Run AEO for every client from one place.
An agency home that shows every client at once, pitch audits for prospects, and white-label reports in Arabic and English under your brand.
- Mission control across clients
- Pitch audits for prospects
- White-label portal and reports
- View-only client access
Where we are
In development, and precise about it.
We are validating data coverage, measurement accuracy and legal questions before the build is committed. Here is what is in the launch scope, what follows it and what depends on that validation.
Measurement
- Theme and question ranks with ranges, mention rate, share of voiceLaunch scope
- Arabic and English in the UAE, reported separatelyLaunch scope
- Saudi Arabia, Arabic and EnglishDepends on validation resultsConditional
- UK, Australia, New Zealand and US, EnglishConfigurable; launch scope not yet decidedUnder decision
Engines
- ChatGPT, Google AI Overviews, AI Mode, Gemini, Perplexity, CopilotSubject to validation per engineLaunch scope
- Claude, through the official API, labelled as suchLaunch scope
- Grok, optionalLaunch scope
Analysis
- Competitor landscape per theme, cited sources and source gapsLaunch scope
- Technical eligibility auditLaunch scope
- GA4, Search Console and Bing AI Performance imports, shown as a floorLaunch scope
- Personal AI agents (OpenAI dots and others), via labelled stand-insSimulated
Agencies
- Mission control, pitch audits, white-label portal and reportsLaunch scope
Platform
- REST API (read and write), webhooks and MCP read toolsLaunch scope
- Evidence-tiered recommendations and before/after experimentsAfter launch
- Server-log analysis of AI crawler and agent trafficAfter launch
Be first to see where you stand.
Tell us your domain and the markets you care about. We'll be in touch as places open in the pilot and early access. No payment details and no commitment.