Real-Time Competitor Intelligence Dashboard

The Problem

In managed IT services, competitive intelligence is the difference between winning and losing six-figure deals. But at most MSPs, competitive research lives in individual reps' heads. One senior rep knows a competitor's pricing model. Another remembers a feature gap from a deal she lost two years ago. None of this is systematized, searchable, or available to the team. Every proposal starts from scratch, and leadership has no aggregate view of the competitive landscape.

What I Built

I designed and deployed a competitive intelligence platform on Cloudflare Workers that tracks 66+ MSP and IT services competitors in real time. The system ingests competitor data from multiple public sources (company websites, Google News RSS feeds, job postings, and review platforms) and structures it into searchable profiles with executive data, service offerings, geographic footprint, and recent activity.

The architecture is deliberately simple and fast. A Cloudflare Worker handles all routing, authentication, and data retrieval. Workers KV stores the entire competitor dataset: base profiles, user-added entries, cached intelligence, notes, and document attachments. The frontend is a single HTML file with inline CSS and JavaScript, served directly by the Worker with competitor data injected at request time. No build step, no npm packages, no framework overhead.

Key Capabilities

The dashboard provides five core views. The Directory gives a searchable, filterable grid of all 66+ competitors with key stats: revenue range, employee count, geographic presence, and service categories. The Intel Feed surfaces recent news, hiring patterns, and market moves across the competitive set using automated Google News RSS ingestion with a noise blocklist that filters irrelevant results (short company names collide with pharma stocks, biotech firms, and unrelated entities). Battlecards provide deal-ready competitive positioning for each competitor: strengths, weaknesses, pricing intelligence, and suggested counter-positioning. The Watchlist lets users star priority competitors for focused monitoring. Analytics surfaces patterns across the full dataset: market concentration, service overlap, geographic competition density.

The system also supports per-competitor notes, document uploads, and live website enrichment that fetches and parses competitor sites to extract current metadata. A companion Python script using Playwright handles deep scraping for competitors with JavaScript-heavy sites, pushing results back to the Worker API.

Technical Decisions

Several architectural choices were driven by the Cloudflare Workers runtime constraints. Workers lack DOMParser, so RSS XML parsing uses regex pattern matching, unconventional but reliable for the structured RSS format. Server-side data injection (competitors, news, watchlist data embedded in the HTML at serve time) eliminates extra API calls on page load, keeping the dashboard responsive. Passcode-based authentication with hashed session cookies provides access control without the complexity of OAuth. The entire system runs on Cloudflare's free tier with near-zero operational cost.

Impact

The platform replaced hours of manual competitive research per deal with a 10-minute briefing workflow. Sales engineers enter a competitive deal and immediately access structured intelligence: current positioning, recent news, known strengths and weaknesses, and patterns from historical win/loss data. Leadership gets a real-time view of competitive landscape shifts without manual reporting. The system processes 200+ data points daily across the competitive set and has been running in production since deployment with minimal maintenance.