Skip to main content

​
Understanding Scores & Metrics

The GEO Score is a composite score from 0 to 100 built from 4 weighted sub-scores:
Sub-scoreWeightWhat it measures
Brand Mention Visibility35%Percentage of AI responses that mention your brand
Share of Voice25%Your mentions vs. all competitor mentions
Source Citation Visibility20%Percentage of responses citing your domain as a source
Sentiment20%Tone of how AI describes your brand (0–100)
On top of the base score, Qwairy adds bonus points (up to 23) for:
  • Progression — score is improving over time (up to 10 pts)
  • Quality — high sentiment (up to 5 pts)
  • Multi-provider presence — visible across multiple AI models (up to 5 pts)
  • Top position — frequently ranked first in lists (up to 3 pts)
The final score is capped at 100.
  • Mention — your brand name appears in the AI-generated text (e.g., “Qwairy is a GEO platform”). The AI talks about you, but may not link to your site.
  • Citation (or source citation) — the AI references a specific URL from your domain as a source. This means the model not only knows you exist, but trusts your content enough to cite it.
A high mention rate with a low citation rate means AI talks about your brand but relies on third-party sources to back it up. The goal is to improve both.You can see the breakdown in Cockpit > Overview (Brand Mention Visibility vs. Source Citation Visibility).
Share of Voice (SoV) is the percentage of total brand mentions that belong to you, compared to your direct competitors detected in AI responses. Only SELF and DIRECT competitors are included — INDIRECT mentions are excluded.Formula: Your brand mentions (SELF) ÷ Total mentions (SELF + DIRECT) × 100For example, if AI mentions your brand and 3 direct competitors, and you appear in 25% of those mentions, your SoV is 25%.Track it in Cockpit > Overview and compare it against competitors in Cockpit > Compare.
Each AI provider (ChatGPT, Claude, Gemini, Perplexity, etc.) uses different training data, retrieval methods, and ranking algorithms. This means:
  • A provider with web search (Perplexity, AI Overview) may cite your recent content
  • A provider relying on training data (ChatGPT, Claude) may reflect older information
  • Some providers weight certain source types (Reddit, Wikipedia, review sites) more than others
Use the Cockpit > GEO Matrix to see your score per provider, and the Cockpit > Compare > By Provider breakdown to benchmark against competitors on each provider.
The GEO Matrix is a heatmap that maps your brand’s AI visibility across two dimensions: rows (topics, tags, funnel stages, or individual prompts) and columns (AI providers).Each cell displays a score from 0 to 100, color-coded from red (invisible) to green (excellent). This lets you instantly spot which topic × provider combinations need work.Click any cell to see the detailed score breakdown, top competitors, cited sources, and trend sparkline.Navigate to Cockpit > GEO Matrix to access it.

​
Credits & Pricing

The formula is:
Total credits = Number of prompts × Sum of credits per selected model
Core models cost 1 credit each: ChatGPT, Perplexity, Google AI Overview, Google AI Mode, Microsoft Copilot, Grok.Premium models cost between 2 and 5 credits each:
CreditsModels
2Claude Haiku 4.5, GPT-5 nano, Perplexity Sonar, DeepSeek Chat, DeepSeek Reasoner
3Claude Sonnet 4/4.5/4.6, GPT-5 mini, Perplexity Sonar Pro, Grok 3 Mini, Mistral Small
4Gemini 2.5 Flash, Gemini 3 Flash, Mistral Medium, Mistral Large
5Claude Opus 4.5/4.6, Gemini 2.5 Pro, Gemini 3 Pro, Grok 3, Grok 4
Example: 30 prompts with ChatGPT (1) + Claude Sonnet 4 (3) + Perplexity (1) = 30 × 5 = 150 credits.The exact cost is displayed before you confirm any analysis run.
  1. Go to Team Management > Billing or visit: https://www.qwairy.co/dashboard/team/billing
  2. Click Upgrade Plan or Buy Credits
  3. Select your desired plan or credit package
  4. Complete the payment through Stripe
Credits are shared across all team members and all brands in your workspace.
PlanPriceCredits/monthBrands
Starter€59/month or €590/year1,0003
Growth€189/month or €1,890/year4,00010
Business€449/month or €4,490/year10,00050
EnterpriseCustom50,000+Unlimited
All plans include unlimited team members. See the full comparison at qwairy.co/pricing.

​
Monitoring & Data

  1. Go to Workspace > Monitoring
  2. Find the Automatic Schedule section
  3. Select your preferred frequency:
    • None — manual runs only
    • Daily — every day
    • Weekly — pick a day of the week
    • Monthly — pick a day of the month
  4. Click Save Schedule
Higher frequency consumes more credits. Weekly is recommended for most brands.
Yes. In Workspace > Monitoring, you can filter by:
  • Topic — only run prompts from a specific topic
  • Tag — only run prompts with a specific tag
  • Model — choose which AI models to include
The credit cost updates in real time as you adjust the filters.
Most results appear within a few minutes after launching an analysis. For large prompt sets (40+ prompts) or premium models, it may take up to 10–15 minutes.Look for the “Generating” indicator next to your brand name — it disappears once all responses are processed.
If your responses are stuck or not generating, check the following:
  • Insufficient credits — check your credit balance in Team Management > Billing
  • Generation in progress — look for the “Generating” indicator next to your brand
  • Provider issues — some AI providers may be temporarily unavailable
  • Rate limits — wait a few minutes if you’ve triggered many generations
If the issue persists, contact us via the chat on Qwairy.

​
Competitors

Competitors are automatically extracted from AI-generated responses. When an AI model mentions other brands alongside yours in a response, Qwairy identifies them and adds them to your competitor list.You don’t need to configure anything — Qwairy discovers competitors as it analyzes responses. New competitors can appear after each monitoring run.Manage your competitor list in Monitor > Competitor Mentions, where you can mark competitors as favorites, set their relationship (direct, indirect), or hide irrelevant ones.
Go to Cockpit > Compare and select up to 10 competitors from the dropdown. The tab shows:
  • Metrics Table — side-by-side comparison of 10 KPIs
  • Positioning Map — scatter plot on any 2 metrics
  • Overlap Matrix — heatmap of how often brands co-appear
  • Evolution Chart — trends over time per competitor
  • Breakdown — comparison by provider, topic, tag, or funnel stage
See the Compare documentation for details.

​
Workspace & Brand Management

  1. Click the brand switcher in the top navigation
  2. Select Add Brand
  3. Enter your website URL
  4. Follow the 5-step onboarding wizard:
    • Review auto-detected brand info
    • Review topics
    • Select prompts to monitor
    • Configure monitoring frequency and AI models
    • Launch your brand
The number of brands you can add depends on your plan (Starter: 3, Growth: 10, Business: 50).
  1. Go to Workspace > Settings for the brand you want to archive
  2. Scroll to the Danger Zone section
  3. Click Archive Brand
  4. Type the brand name to confirm
Archiving a brand permanently removes all prompts, responses, analytics, and settings. This action cannot be undone.
  1. Navigate to Workspace > Prompts
  2. Find the prompt you want to delete in the list
  3. Click the three-dot menu (…) next to the prompt Delete prompt menu
  4. Select “Delete Prompt” from the dropdown
  5. Confirm your action
You can also select multiple prompts using checkboxes and use Bulk Delete.
Deleting a prompt removes all associated responses and analytics. This action is permanent.
  1. Go to Workspace > Topics
  2. Find the topic you want to remove Topic list with delete option
  3. Click the trash bin icon in the Actions column
  4. Confirm the deletion
Deleting a topic removes it from all associated prompts and from GEO Matrix / Compare breakdowns. Prompts themselves are not deleted.

​
Exports & Integrations

You can export data in two ways:1. From Workspace > ExportsTrigger an export from the dashboard, then download it as CSV or XLSX from the Exports page. Available export types include: responses, competitors, sources, prompts, sentiment, shopping, social, and more.2. Via Looker StudioConnect Qwairy to Google Looker Studio for automated, refreshable reports. See the Looker Studio documentation.3. Via the APIUse the REST API to pull data programmatically. See the API Reference.
  1. Go to Measure > Google Search Console
  2. Click Connect Google Search Console
  3. Sign in with your Google account
  4. Select the property you want to connect
  5. Grant Qwairy read-only access to your data
Once connected, Qwairy syncs your keyword data and cross-references it with AI visibility. You can see which queries rank on Google but are invisible in AI responses.
  1. Go to Workspace > API Keys
  2. Click Create API Key and give it a descriptive name
  3. Copy the key immediately (it’s shown only once)
  4. Use it in the Authorization header:
curl -H "Authorization: Bearer YOUR_API_KEY" \
  https://api.qwairy.co/v1/brands
See the API Reference for all available endpoints.
The Model Context Protocol (MCP) lets AI assistants like Claude access your Qwairy data in real time. You can ask questions like “How is my brand performing on Perplexity?” and get live answers from your monitoring data.See the MCP documentation to set it up.

​
Team & Billing

  1. Go to Team Management > Members or visit: https://www.qwairy.co/dashboard/team/members
  2. Click Invite Member
  3. Enter the email address and select a role
Access levels:
RolePermissions
OwnerFull access including ownership transfer, cannot be removed
ManagerFull access including billing, adding brands, team settings, viewer management
MemberFull access except billing management
ViewerRead-only access to assigned brands only
All invoices are available in Team Management > Billing or at: https://www.qwairy.co/dashboard/team/billingClick on any past invoice to download or view the PDF.

​
Troubleshooting

This can happen for several reasons:Empty performance dashboard1. Generation is still in progressAfter onboarding, Qwairy needs a few minutes to generate responses. Look for the “Generating” indicator next to your brand name.Generating indicator2. Topics don’t match your brandIf the auto-generated topics don’t align with your actual business, AI responses won’t mention your brand. Go to Workspace > Topics to review and adjust them.3. Your brand is genuinely absent from AI responsesThis is the most common case — AI models don’t mention your brand for the monitored prompts. This is actually the starting point for a GEO strategy: use Content Opportunities and Content Studio to create content that gets your brand into AI responses.If you believe there’s a bug, contact us via the chat on Qwairy.
A provider showing 0% visibility means that none of the AI responses from that provider mention your brand for your monitored prompts. Common reasons:
  • Provider relies on different sources — ChatGPT and Claude use training data, while Perplexity uses live web search. Your brand may be visible on one but not the other.
  • Topic mismatch — the provider may not cover your niche well.
  • Content gap — the sources the provider trusts don’t mention your brand.
Check the Cockpit > GEO Matrix to see which provider × topic combinations are 0, then use Monitor > Citation Sources to understand which sources the provider relies on.
No. Qwairy does not crawl or scrape your website. It analyzes what AI models (ChatGPT, Perplexity, Claude, Gemini, etc.) say about your brand by sending monitored prompts and analyzing the responses.The only connection to your site is through Google Search Console and Bing Webmaster Tools integrations — and these require your explicit authorization with read-only access.