llemmy is a GEO (Generative Engine Optimization) intelligence platform. It tracks how AI engines - ChatGPT, Claude, Perplexity, Google Gemini, Google AI Overviews, Google AI Mode and self-hosted Ollama - cite and describe your brand, then turns that into visibility scores, share of voice, sentiment, cited-domain reports and content-gap recommendations. This page answers the most common questions about GEO and about using llemmy. For a free, no-signup snapshot of how AI describes any domain, run the free GEO audit. To start tracking, open the app.
Generative Engine Optimization (GEO) is the practice of improving how often and how accurately AI engines cite, mention and describe your brand when they answer user questions. Where traditional SEO optimizes for ranking in a list of blue links, GEO optimizes for being the source an AI model quotes inside a generated answer. It spans engines like ChatGPT, Claude, Perplexity, Google Gemini and Google AI Overviews.
SEO targets search engine results pages - getting a page to rank in Google's ten blue links. GEO targets the answer itself - getting your brand named, cited and described correctly inside an AI-generated response. The two overlap, because AI engines often draw on the same crawled web content, but GEO adds new signals: which domains the model cites, how it characterizes you, and whether it recommends you over competitors.
A growing share of buyer research starts in AI assistants rather than a search box, so if AI engines omit or misdescribe your brand, you lose pipeline before a click ever happens. GEO matters because these answers are often the first - and sometimes only - impression a prospect gets. Tracking and improving your AI visibility protects demand that traditional analytics can't see.
Yes, within limits. You can't directly edit a model's output, but you can influence it by improving the source content models draw on - clear, factual, well-structured pages; consistent entity information; authoritative third-party coverage; and content that answers the real questions buyers ask. llemmy's job is to measure the result so you know which efforts move the needle.
Share of voice in GEO is the proportion of AI answers about your category in which your brand is mentioned, compared with competitors. If ten tracked prompts about your space produce answers and you appear in six while a competitor appears in eight, your share of voice is lower than theirs. It's the clearest single measure of who is winning the AI conversation in your market - see the Competitor Intel feature.
Create a llemmy project for your brand, add the prompts you want to monitor (or import suggestions), select the AI engines to run them against, and llemmy executes them and records every response. From there you get visibility, share-of-voice, sentiment and cited-domain reports in the dashboard. Open the app to create your first project.
No. llemmy runs prompts against AI engines and analyzes the responses, so it works without any code on your site or a tag in your pages. Optional integrations like Google Search Console and Google Analytics 4 connect via OAuth, not a script. You can be tracking your brand within minutes of signing up.
Yes. The free GEO audit at llemmy.com/geo-audit grades any domain and shows how ChatGPT, Claude, Perplexity and Google AI describe it - no signup required to see your score. It's the fastest way to understand your current AI visibility. For ongoing tracking, competitor benchmarking and scheduling, you then create a project in the app.
A project is an isolated workspace for one brand or website, holding its own prompts, responses, settings, competitors and integrations. Agencies and multi-brand teams create one project per client so data never mixes. You switch between projects from the project switcher in the dashboard.
You'll see your first responses as soon as a run finishes, typically within a minute of triggering it. Trends like visibility and share of voice become meaningful once you've accumulated a few daily runs, so most teams have a useful baseline within a week. A daily refresh keeps the trend lines current automatically.
Your AI visibility score measures how often your brand appears in AI answers across your tracked prompts and engines. A higher score means your brand is mentioned in a larger share of relevant AI responses. It's the headline metric for whether AI engines “know” and surface your brand - see the AI Visibility feature for detail.
The AI citation gap is the set of prompts and topics where competitors are cited by AI engines but your brand is not - the visibility you're missing. llemmy surfaces these gaps so you can see which questions you are losing and which content is most likely to win them back. Closing the citation gap is the core day-to-day GEO workflow.
llemmy analyzes the language each AI engine uses about your brand and classifies it as positive, neutral or negative. This tells you not just whether you're mentioned, but whether you're described favorably. Tracking sentiment over time flags reputation shifts - for example, an engine starting to associate your brand with a recurring complaint.
Cited domains are the websites AI engines reference when they generate answers about your category. They reveal which sources models trust - review sites, competitors' blogs, industry publications - so you know where to earn coverage or placement. If a third-party domain repeatedly drives your competitor's mentions, that's a concrete GEO target.
The GEO Score is llemmy's 0 to 100 grade of your brand's visibility in AI answers, combining visibility, position and sentiment into one number. It gives teams and clients a single, trackable headline that goes up as your GEO efforts work. Learn how it's calculated on the GEO Score feature page.
Prompts that name your brand directly (branded prompts) almost always mention you, so counting them in your core visibility score would inflate it. llemmy keeps branded prompts out of the core visibility metric and surfaces them separately for sentiment tracking. This keeps your visibility score an honest measure of unprompted, category-level discovery.
llemmy tracks ChatGPT (OpenAI), Claude (Anthropic), Perplexity, Google Gemini, Google AI Overviews, Google AI Mode and self-hosted Ollama models. This is a broader engine set than most GEO tools cover. You choose which engines to run each prompt against per project.
llemmy sends each of your tracked prompts to the selected AI engines and records the full response, then parses it for brand mentions, sentiment and cited sources. Google AI Overviews and AI Mode are read from the search results page. Every response is stored so you can read the exact wording and audit how answers change over time.
By default, platform-managed models power every plan, so you don't need any API keys to start tracking. On Enterprise you can bring your own provider keys (OpenAI, Anthropic, Gemini, Perplexity) to route runs through your own accounts; those keys are encrypted at rest with AES-256-GCM and are never shown again after you save them.
Yes. llemmy supports Ollama for self-hosted models, which is useful if you need to keep prompts and data on your own infrastructure. Self-hosting is a deliberate differentiator for enterprises with data-residency requirements. Local-model tracking requires a self-hosted llemmy deployment, since it needs access to your Ollama instance.
AI engines update frequently, and the same prompt can produce different answers from one day to the next as models and their sources change. That's why llemmy stores every response with a timestamp and plots trends rather than relying on a single snapshot. Scheduling a daily run is the best way to catch shifts as they happen.
Track the real questions your buyers ask AI engines about your category - “best [product] for [use case]”, “alternatives to [competitor]”, “is [your brand] any good” - not just branded queries. The strongest GEO signal comes from non-branded, category-level prompts where you're competing to be mentioned at all. llemmy can suggest prompts and competitors to get you started.
The number of prompts depends on your plan, and the limit applies per project so multi-brand accounts aren't penalized. llemmy does not impose the hard per-prompt quotas common to other GEO tools. See current limits on the pricing page.
llemmy runs a global daily refresh that re-executes your tracked prompts across all projects so your visibility, share-of-voice and sentiment trends stay current without manual effort. You can also trigger any prompt, or a full project run, on demand at any time. The daily refresh is what powers the trend lines in your dashboard.
Yes. You can run a single prompt, or an entire project, immediately from the dashboard or the API and see the new responses right away. On-demand runs are useful after publishing new content or when you want to confirm a change moved your visibility. Scheduled and on-demand runs both feed the same trend history.
Yes. llemmy offers a public API (v2) authenticated with API keys, covering projects, prompts, responses and triggering prompt runs. It's designed for agencies and teams that want to pull GEO data into their own dashboards or automate reporting. You create an API key in the app under API Keys.
Yes. llemmy connects to Google Search Console (GSC) and Google Analytics 4 (GA4) via Google OAuth, with tokens stored per project. GSC brings in keyword and page performance; GA4 brings in traffic and AI-referral attribution. Together they let you compare traditional search demand against your AI visibility. Integrations are available on paid plans - see the pricing page.
You add competitor brands to a project, and llemmy measures how often each competitor is mentioned across the same AI answers as you - your share of voice. It then highlights prompts where competitors are cited and you're not, which is your citation gap. llemmy can also suggest likely competitors to track. See Competitor Intel.
The Brand Profile is where you tell llemmy who your brand is - industry, description and competitors - so its analysis and recommendations are accurate. A complete profile sharpens sentiment classification, competitor matching and content-gap suggestions. It takes a few minutes and meaningfully improves the quality of every downstream report.
Yes. By combining your citation gaps, cited-domain analysis and (when connected) GSC/GA4 data, llemmy surfaces the topics and questions where creating or improving content is most likely to win AI mentions. The goal is to point you at the highest-leverage GEO work rather than a generic checklist. Deep AI-generated analysis is metered by plan.
llemmy offers a free tier plus paid plans (Starter, Growth, Pro and Enterprise) priced in USD, scaling on projects, prompts, runs and advanced features. The free tier lets you try core tracking; paid plans add integrations, higher limits and metered AI analysis. See current pricing on the pricing page.
Yes - agencies are a core llemmy use case. Multi-project workspaces keep each client's data isolated, the free GEO audit works as a prospecting tool, white-label reports carry your own brand, and the API supports custom client reporting. Learn more on the agencies page.
The free tier lets you create a project and track prompts across AI engines so you can see your baseline AI visibility. Advanced capabilities - GSC/GA4 integrations and deep AI-generated analysis - are reserved for paid plans. It's a no-cost way to confirm llemmy fits before upgrading.
Billing runs through Stripe, with secure hosted checkout and a self-serve billing portal to upgrade, downgrade or manage your subscription. Plan changes take effect through the portal, and Enterprise is handled via sales contact. You can manage everything from the billing section in the app.
Yes. llemmy supports self-hosting for teams with data-residency or ownership requirements, and the public API lets you export projects, prompts and responses. Data ownership is a deliberate design principle - you're not locked into the cloud. Contact sales to discuss self-hosted and enterprise options.
Run a free, no-signup GEO audit to see your own numbers, or get in touch - we're happy to help.