Frequently Asked Questions

Most of the questions on this page came up the same way: someone signed in, started using Trakkr, hit a moment of "wait, how does X actually work?", and emailed us. We've collected the answers in one place so you don't have to send the email.

The sections below mirror the way most people learn the product: the concepts first, then the metrics behind them, then the day-to-day questions about prompts, competitors, citations, plans, and integrations. Use the table of contents on the right to jump to your question, or skim top to bottom on a quiet afternoon; it's a decent tour of how the whole thing fits together.

Getting Started

What is AI visibility and why does it matter?

AI visibility measures how often and how prominently your brand appears when people ask AI assistants (like ChatGPT, Claude, or Perplexity) for recommendations. As more consumers use AI for product research and purchasing decisions, being visible in AI responses directly impacts your discoverability and revenue.

Learn more: Core Concepts explains the mental model behind AI visibility - what gets measured, why, and how to think about your scores.

How is this different from SEO?

Traditional SEO focuses on ranking in Google's list of links. AI visibility is about being included in the AI's synthesized answer - there are no "10 blue links" to rank in. The two are related (good SEO often helps AI visibility) but they're fundamentally different:

Learn more: Introduction to AI Visibility

Which AI models does Trakkr track?

Trakkr tracks up to 8 AI models: ChatGPT, Claude, Gemini, Grok, Copilot, and Meta AI are included on all plans (6 models). Growth and Scale add Perplexity and DeepSeek (8 models total). Each model has different training data and biases, so tracking all of them gives you the complete picture.

Learn more: Core Concepts | Dashboard Guide

How quickly can I see results?

You'll see your first visibility scores within minutes of running research. Improving your scores takes longer:

• Perplexity & AI Overviews: Days to weeks (they search the live web)
• ChatGPT & Claude: Weeks to months (depends on training data updates)
• Gemini: Variable (mix of training data and search)

The key is consistency. Changes that improve your authority across multiple sources tend to compound over time.

Learn more: Core Concepts

Understanding Your Scores

How is the visibility score calculated?

Visibility is a weighted score based on your position when mentioned. Higher positions earn more points:

• Position 1: 10 points
• Position 2: 9 points
• Position 3: 8 points
• And so on down to position 10

Your visibility score is the percentage of maximum possible points earned across all your prompts. A brand appearing at #1 in 50% of prompts would score around 70% visibility.

Learn more: Metrics for the full formula | Core Concepts for the mental model

Why does my score differ between models?

Each AI model has different training data, cutoff dates, retrieval systems, and inherent biases. It's completely normal to score 80% on Claude but 40% on ChatGPT. This is actually valuable information - it tells you where to focus improvement efforts.

Common patterns:

• Perplexity favors recent, well-structured content (it searches live)
• ChatGPT relies heavily on training data from authoritative sources
• Claude tends to be more conservative and fact-focused
• AI Overviews pulls from Google's search index

Learn more: Core Concepts | Dashboard Guide

What's the difference between visibility and presence?

Visibility measures how prominently you appear (position-weighted). Presence is simpler - it's the percentage of prompts where you're mentioned at all, regardless of position.

A brand with 100% presence but 50% visibility appears everywhere but usually in lower positions. A brand with 50% presence but 90% visibility appears less often but dominates when it does.

How accurate are the visibility scores?

Our scores reflect real AI model responses to your prompts. We query actual AI APIs (not simulations) and analyze genuine responses. AI models can produce slightly different outputs for the same prompt, so there's natural variation - but the trends and patterns are reliable.

Why do the visibility score and competitor rank seem to disagree?

They measure different things, so they can look out of sync without either being wrong:

• Visibility is an absolute number - the percentage of maximum position-weighted points your brand earns across all your tracked prompts. 50% means you captured half the possible points.
• Rank is your position in an ordered list of brands sorted by visibility. Your rank depends on what everyone else scored.

Two scenarios that catch people out:

1. High score, lower rank. You can score 80% visibility and still rank #4 if three competitors score above 80%. The score reflects how well you're doing in absolute terms; the rank reflects how well you're doing relative to a specific peer set.
2. Filtered views shift everything. When you filter by a single model, the score recomputes against just that model's responses - and so does the rank. A brand that's strong on Perplexity but weak on ChatGPT can swap ranks dramatically when you flip the filter.

If a number ever looks off, click into the prompt-level breakdown. It shows every model's response and every brand's position, which makes the math concrete.

Prompts

What are prompts and why do they matter?

Prompts are the questions you track - the queries your potential customers are asking AI. They're the foundation of everything in Trakkr. Your visibility score is calculated across your prompts, so choosing the right ones is crucial.

Learn more: Prompts Overview

How many prompts should I track?

Quality over quantity. Start with 5-20 highly relevant prompts covering:

• Discovery questions: "Best [category] for [use case]"
• Comparison queries: "[Your brand] vs [competitor]"
• Reputation queries: "Is [brand] good for [purpose]?"
• Category queries: "Top [products] in [industry]"

You can always add more as you learn what matters. The Ideas tool at the bottom of the Prompts page suggests new ones based on your brand, tags, or any keyword you give it.

What makes a good prompt?

Good prompts are:

• Natural questions (how real people actually ask)
• Specific enough to be relevant to your business
• Questions your customers genuinely ask
• A mix of branded and unbranded queries

Avoid prompts that are too generic ("best software") or too specific to ever be asked ("best project management tool for 47-person marketing agencies in Austin").

Learn more: Prompts Overview

Can I organize my prompts?

Yes! Use tags to organize prompts by intent, funnel stage, product line, or any other dimension. Tags also let you analyze performance by segment - see which types of queries you're winning vs. losing.

Where can I see the raw AI responses for each prompt?

Two places, depending on what you need:

• Prompts page → click any prompt opens the detail drawer. You'll see the trend, the phrasings AI used to talk about you (and competitors), the sources it cited, and recent activity.
• Competitors page → Prompts mode → expand any row shows the full, unedited model response with your brand and competitor names highlighted inline. This is the place to read exactly what ChatGPT or Claude said and understand why you were (or weren't) mentioned.

You can also pull raw responses programmatically: the MCP server (any paid plan) lets your AI assistant query them conversationally, and the REST API (Scale plan) is useful for piping into spreadsheets, BI tools, or your own analysis.

Brands & Markets

Does Trakkr track prompts in languages other than English?

Yes. Trakkr supports 50+ languages and regional variants. When you set a brand's language to Dutch, German, Brazilian Portuguese, or any other supported language, AI models are prompted in that language and responses are analyzed natively. No translation step in between.

Set the language in Settings → Brand → Language. The catalog includes:

• English variants: US, UK, Australia, Canada, India, Singapore, Ireland, New Zealand, South Africa
• German variants: Germany, Austria, Switzerland
• French variants: France, Canada, Belgium, Switzerland
• Spanish variants: Spain, Mexico, Argentina, Colombia, Chile
• Portuguese: Portugal, Brazil
• Other European: Italian, Dutch (NL + Flemish), Swedish, Danish, Norwegian, Finnish, Polish, Czech, Slovak, Hungarian, Romanian, Greek, Croatian, Slovenian
• Asia: Japanese, Korean, Chinese (Simplified + Traditional), Hindi, Thai, Vietnamese, Indonesian
• Middle East & Africa: Arabic, Hebrew, Turkish

If your language isn't listed, email [email protected] - adding new ones is straightforward.

Can I target a specific country or region for my brand?

Yes. Every brand has a primary market (a country, with optional region or city), and you can add additional markets to track the same brand across multiple geographies. Each market runs its own research, so you can compare AI visibility in Australia vs the UK vs Singapore - for the same brand - and see where AI knows you and where it doesn't.

Set markets in Settings → Brand → Markets. Trakkr currently supports 50+ countries including:

• Americas: US, Canada, Mexico, Brazil, Argentina, Chile, Colombia
• Europe: UK, Ireland, Germany, France, Spain, Italy, Netherlands, Belgium, Switzerland, Austria, the Nordics, Portugal, Poland, Czech Republic, Hungary, Romania, Greece
• APAC: Australia, New Zealand, Singapore, Malaysia, Hong Kong, Taiwan, Japan, South Korea, India, Indonesia, Thailand, Philippines, Vietnam
• MENA: UAE, Saudi Arabia, Israel, Egypt, Turkey
• Africa: South Africa, Nigeria

Pro tip for agencies: pick the primary market based on where your client actually sells, not where they're headquartered. AI behaviour is meaningfully different from market to market - especially for Perplexity and AI Overviews, which weight local sources.

Competitors

Why can't I manually add competitors?

This is a common question! In Trakkr, competitors are discovered automatically when you run research. Here's why:

Trakkr's job is to show you who AI actually recommends alongside you - not who you think your competitors are. When you track prompts, the research identifies which brands appear in responses. These are your real AI competitors, which often differ from your traditional market competitors.

This approach surfaces competitors you might not have considered and shows you the actual competitive landscape in AI responses.

How do I see my competitors?

After running research, go to Competitors to see:

• Which brands appear alongside yours
• Head-to-head win rates
• Model-by-model breakdowns
• Competitive threats and opportunities

Learn more: Competitors

Why is a competitor scoring higher than me?

Check your Citations for clues. Higher-visibility competitors often have:

• More mentions on authoritative sources AI models trust
• Better coverage on review sites, comparison articles, and industry publications
• Clearer, more consistent brand positioning
• Longer history of being discussed online

Learn more: Citations Overview | Outreach

Can I edit or remove competitors?

You can't add competitors manually; as explained above, the list is discovered from real AI answers, not configured. But you do have full curation control over what's discovered:

• Hide invalid competitors. Click the row menu (the three dots) on any competitor and select Hide. Useful for brands that show up in responses but aren't really competitors - a generic platform AI mentions in passing, a product from a totally different category, or an obvious false positive. Hidden competitors disappear from your dashboard, rankings, and matrix. You can unhide them later from the same menu.
• Merge variants. Open Manage Groups (top-right drawer on the Competitors page) to group "On" and "On Running," "HubSpot" and "Hubspot," or any other duplicate phrasings into a single tracked brand. Trakkr also auto-suggests groups it detects, which you can accept or dismiss.

If a brand you consider a competitor isn't showing up, the fix isn't to add it manually; it's to track prompts where you'd expect it to appear. Once AI names it in a response, it'll surface in your competitor set automatically.

Citations & Sources

What's a citation?

A citation is a source that AI models reference or draw from when generating responses. Trakkr identifies these by analyzing model responses - some models explicitly cite sources (like Perplexity), while others we infer from response content.

Citations show you which websites, articles, and platforms influence what AI says about your brand.

Learn more: Citations Overview

How do I get cited by AI?

Focus on getting mentioned on sources that AI trusts:

• Industry publications and news sites
• Review platforms (G2, Capterra, Trustpilot)
• Comparison articles and "best of" lists
• Wikipedia and knowledge bases
• Reddit and community discussions

Learn more: Citations Overview | Outreach

Improving Your Visibility

How do I improve my AI visibility?

There are three main levers:

1. Get cited on authoritative sources - AI learns from the web. Get mentioned on sites it trusts.
2. Optimize your own content - Make your site easy for AI crawlers to understand (structured data, clear positioning, comprehensive content).
3. Build entity authority - Ensure AI understands what your brand is and what it does consistently.

Learn more: Core Concepts | Content Studio

How long do changes take to affect visibility?

It depends on the model:

For fastest results, focus on Perplexity and AI Overviews first. For lasting impact, build authority that persists across training cycles.

Learn more: Core Concepts

What is AI Pages and how does it help?

AI Pages is Trakkr's real-time AI crawler optimization. It's a lightweight integration that intercepts AI crawler requests and serves optimized versions of your pages, better structured data, cleaner content, faster responses. AI Pages works with Cloudflare, Vercel, Netlify, Next.js, AWS CloudFront, WordPress, Node.js, Nginx, and more.

Think of it as "serving your best self" to AI crawlers without changing your actual website.

Learn more: AI Pages Overview | AI Pages Setup

Features

What is Perception analysis?

Perception shows how AI models perceive your brand across 20 dimensions - trust, quality, value, innovation, and more. It's not just whether AI mentions you, but what it says about you.

Use Perception to identify narrative gaps, find positioning opportunities, and track how your brand story evolves in AI responses.

What are Workflows?

Workflows are automated alerts and actions. Set up "when X happens, do Y" rules - like getting a Slack notification when visibility drops, or an email when a competitor gains ground.

Learn more: Workflows | Integrations

What is the Agent?

The Agent is Trakkr's AI assistant. It knows your brand, your data, and your history. Ask it questions, get insights, and take actions through natural conversation.

What do I need to give my developer to connect my site?

In most cases just a URL and an OAuth click - no IP allowlisting or firewall changes needed.

WordPress. Send your dev your site URL and ask them to make sure /wp-json (the REST API) is reachable. Trakkr connects via OAuth or an Application Password generated by an existing admin user (an "admin REST API user" — same thing), and you'll authenticate inside Trakkr - not WordPress. If your site uses a security plugin or WAF like Wordfence, iThemes Security, or Sucuri, make sure REST API access isn't blocked for authenticated users. (The optional Trakkr WP plugin, used for crawler-visit tracking, adds /wp-json/trakkr/v1/crawler-visits on your site — whitelist that path too if /wp-json/ is locked down.)

Vercel. Install the Trakkr integration from the Vercel marketplace and grant access to your project. Edge middleware deploys with your next build - no manual config.

GitHub. Install the Trakkr GitHub App on the repo. You choose which repos to grant access to; Trakkr only sees what you authorize.

AI Pages (any platform). Trakkr generates the integration code for your stack - Cloudflare Worker, Vercel middleware, Netlify Edge function, Next.js middleware, AWS CloudFront Lambda@Edge, WordPress mu-plugin, Node.js, Nginx, or a Cloudflare DNS proxy for platforms like Shopify and Squarespace. You copy-paste, your dev ships. See AI Pages installation for platform-specific snippets.

No IP allowlist required. All connections happen over standard HTTPS from our managed infrastructure. If your security team asks for a specific IP range to allow, the honest answer is: you don't need one. Treat Trakkr like any third-party SaaS that calls your API.

Plans & Billing

What plans are available?

Trakkr offers three plans:

Learn more: Billing & Plans | Pricing

How do I upgrade my plan?

Go to Settings → Billing and click Upgrade. Select your new plan and complete checkout. Upgrades are prorated - you only pay the difference for the remaining billing period.

What happens if I cancel?

You keep access until the end of your billing period. After that, you're moved to the Free plan (1 brand, 5 prompts). Your data is retained for 30 days in case you resubscribe.

Do inactive prompts count toward my limit?

No! Only active prompts count. Deactivate prompts you're not currently tracking to save quota without losing historical data.

Can I add more brands or prompts?

Yes, with add-ons. Growth and Scale plans can purchase extra brands ($39/brand/month) and prompt packs (+50, +100, or +200 prompts per brand). Manage add-ons at Settings → Billing.

What if I need more than the Scale plan offers?

Scale includes 10 brands and 50 prompts per brand, which fits most agencies and multi-brand teams. If you need more, you have two paths:

Add-ons (self-serve, instant):

• Extra brands at $39/brand/mo - add as many as you need
• Prompt packs at $39-99/brand/mo for +50, +100, or +200 extra prompts per brand
• Article credit packs at $49-149/mo for additional AI-generated content

Enterprise (custom, for larger needs):

• 50+ brands or fully custom limits
• SSO (SAML / OIDC) and SCIM provisioning
• Dedicated success manager and onboarding
• Invoice billing with custom payment terms
• SLA guarantees
• Custom legal terms / DPAs

Email [email protected] with a rough sketch of what you need - team size, brand count, integration requirements - and we'll come back with pricing. The Enterprise tab in Settings is another way in.

Technical & Security

How do you access AI models?

We use official APIs from each provider (OpenAI, Anthropic, Google, etc.). We're making real requests, not simulating responses.

Is my data private?

Yes. Your prompts, brands, and results are private to your account. We don't share individual customer data. See our Privacy Policy for details.

Do you train AI on my data?

No. Your data is used solely to provide Trakkr's services. We don't use customer data to train AI models.

What about security and compliance?

We take security seriously. See our Security page for details on our practices, certifications, and policies. For security concerns, email [email protected].

Do you have an API?

Yes! The MCP server is available on every paid plan (Growth and Scale), so you can connect Trakkr to ChatGPT, Claude, Cursor, and other AI assistants. The REST API is included on the Scale plan. See the API documentation for endpoints, authentication, and examples. Manage your developer access in Settings → Developer.

How do I change the email on my account?

Go to Settings → Profile, click the email row, and enter your new address. Supabase (our auth provider) sends a confirmation link to the new email - click it to finalize the change. Your brands, history, team membership, billing, and integrations all carry over unchanged.

If you signed up with Google, you'll need to keep using Google sign-in with the original Google account. To switch to a different login identity (or merge two accounts you accidentally created), email [email protected] and we'll handle the migration manually.

Getting Help

How do I contact support?

• Chat: Click the chat icon in the bottom-right corner of the app
• Email: [email protected]
• Docs: You're already here!

Scale plan customers get priority support with faster response times.

Can I get a demo?

Yes! Book a demo to see Trakkr in action and get your questions answered.

Where can I learn more?

• Documentation - Feature guides and how-tos
• API Reference - For developers building integrations
• Changelog - Latest product updates

Still have questions? Contact us - we're happy to help!