Promptmonitor Overview

An overview of what Promptmonitor does, the problem it solves, and how it helps brands understand and improve their visibility in AI-generated answers

Promptmonitor is a SaaS platform that helps businesses track and improve their visibility in AI-generated answers. When users ask ChatGPT, Claude, Perplexity, or Gemini questions like "What's the best CRM software?", Promptmonitor monitors whether your brand appears in those AI responses.

The Problem Promptmonitor Solves

Traditional Search Engine Optimization (SEO) tracks Google search rankings. However, many users now get answers directly from AI assistants without clicking on search results. When an AI recommends a competitor instead of your brand, you may lose potential customers, and traditional analytics tools don’t show when or why this happens. Promptmonitor fills that visibility gap.

Core Capabilities

  1. AI Visibility Tracking

Users define prompts (questions) relevant to their business (e.g., "best project management tools").

  • Promptmonitor regularly queries multiple LLMs (GPT-4, Claude, Gemini, Perplexity, etc.) using these prompts

  • The platform analyzes responses to detect brand mentions, competitor mentions, and ranking position

  1. Visibility Score

The Visibility Score is a calculated metric (0-100%) that represents how often and prominently a brand appears in AI-generated answers:

  • Presence Rate (80% weight): How often the brand is mentioned

  • Cross-Model Consistency (20% weight): Whether the brand appears across different AI models

  1. Brand Mentions Analysis

    • Extracts every instance where a brand is mentioned in AI responses

    • Classifies sentiment (positive, negative, or neutral)

    • Shows the context surrounding each mention

    • Tracks brand mentions over time

  2. Competitor Tracking

    • Identifies which competitors appear in the same AI-generated responses

    • Ranks competitors by their visibility scores

    • Shows competitive positioning changes over time

  3. Source Attribution

    • Identifies which URLs/sources the AI models cite when mentioning brands

    • Helps teams understand which content influences AI recommendations

  1. Web Analytics (Bot & Human Traffic)

    • Tracks AI crawlers visiting your website (e.g., GPTBot, ClaudeBot, PerplexityBot)

    • Tracks human visitors separately

    • Shows which pages AI bots crawl most frequently

Technical Architecture

  • Three-Repo System

Promptmonitor uses a modular architecture made up of three core repositories, each responsible for a specific part of the platform.

Repository

Function

promptmonitor

Next.js web app → dashboard UI, authentication, public website

pmserver

Backend worker → executes LLM API calls, processes responses, extracts mentions

mayo

Analytics ingestion → collects pageview and bot visit data

  • Data Storage

Promptmonitor uses multiple storage and processing tools to handle different types of data efficiently:

  • Supabase (PostgreSQL): Stores core application data such as users, teams, projects, prompts, responses, mentions, and competitors

  • Tinybird: Handles time-series analytics for web traffic, including pageviews and AI bot visits

  • Redis + BullMQ: Manages background jobs and task queues for asynchronous processing

Summary

Promptmonitor answers the question, "Is my brand visible when people ask AI assistants about my industry?" Then, it provides data on AI visibility, competitor positioning, and the sources of content that influence AI recommendations.