CoreDash.app Reviews and Details

Questions & Answers

1. What makes CoreDash.app unique?

   CoreDash is a Real User Monitoring platform with element level attribution for every Core Web Vital and a built in MCP server that lets Claude, Cursor, Windsurf, and other AI agents query your field data through the Model Context Protocol. Ask why LCP is slow on mobile in Germany and the agent reads your real user data, pinpoints the bottleneck phase, and names the responsible element. Most RUM tools show averages or sparklines. CoreDash shows you the exact CSS selector, the exact phase (TTFB, LOADDELAY, LOADTIME, RENDERDELAY for LCP; INPUTDELAY, PROCESSING, PRESENTATION for INP), and the exact Long Animation Frame causing INP regressions. Data is stored in the EU and the script runs without consent banners.

2. Why should a person choose CoreDash.app over its competitors?

   Three reasons. First, depth of attribution. Every metric drills down to the exact element and phase causing the slowdown, not just a number on a dashboard. Second, the MCP server. AI assistants like Claude and Cursor can query your real user data directly through the Model Context Protocol, so debugging a CWV regression becomes a plain English conversation with an agent that actually reads the data. Third, price. Paid plans start at $19 per month. SpeedCurve and New Relic Browser run into the hundreds or thousands per month for similar coverage. Privacy is also built in. Data is stored in the EU, no consent banner is needed, and visitors stay anonymous by default.

3. How would you describe the primary audience of CoreDash.app?

   Frontend engineers, performance engineers, and technical SEO teams who own Core Web Vitals at sites that need real evidence from real users rather than synthetic lab numbers. Heavy use among teams shipping ecommerce, news, publishing, and content sites where CWV moves search rankings and conversion. Also popular with web performance consultants and agencies who need to show clients exactly which element is hurting their scores. The MCP integration draws an additional audience of AI native developers who run their performance investigations from inside Claude, Cursor, or Windsurf instead of clicking through dashboards.

4. What's the story behind CoreDash.app?

   CoreDash was built by Arjen Karel, an independent Core Web Vitals and PageSpeed consultant based in the Netherlands with 17 years of web performance experience. He has optimized over 925,000 mobile URLs and audits performance for eBay, Zalando, Nestlé, Adevinta, DPG Media, and Harvard. Existing RUM tools were either too expensive, too shallow on attribution, or hid the field data behind dashboards that did not surface the actual cause of a regression. He built CoreDash as the tool he wanted to use himself, then opened it to other developers when they kept asking for access. The MCP server was added in 2025 once it became clear that performance debugging through AI agents was about to become the new normal.

5. Which are the primary technologies used for building CoreDash.app?

   A lightweight browser script using the Performance Observer API to capture LCP, INP, CLS, FCP, and TTFB with full attribution. The script also records element selectors for LCP and CLS, and Long Animation Frames (LoAF) so INP root causes can be traced to the exact long task and script. Beacons are sent to a high throughput backend pipeline with EU data storage. The dashboard is a modern web application. The MCP server speaks the Model Context Protocol so any compliant AI agent (Claude, Cursor, Windsurf, ChatGPT) can query field data programmatically. The CWV Insights MCP and a Core Web Vitals Visualizer Chrome extension are open source. The core platform is closed source.

6. Who are some of the biggest customers of CoreDash.app?

   CoreDash is built and used by Arjen Karel for performance consulting work, and the same tool powers audits and ongoing Core Web Vitals monitoring across major sites:

   eBay Zalando Nestlé Adevinta DPG Media Harvard

   Not all of these are direct paying customers of the platform yet (the SaaS product is recent), but the engine has been running against their sites under Arjen's consulting work and informs every product decision.