ProjectPulseAI

Project managers and controls teams spend hours every week manually analyzing schedules, calculating EVM metrics in spreadsheets, and trying to spot risks buried in thousands of tasks. By the time problems surface in status meetings, they've already caused delays and cost overruns. The real pain points: - Schedule analysis is manual and slow - Exporting to Excel, building pivot tables, calculating CPI/SPI by hand - Risks hide until it's too late - Critical path issues, resource conflicts, and logic errors go unnoticed in complex schedules - Portfolio visibility is fragmented - Executives can't see which projects are truly healthy vs. which are masking problems - EVM expertise is scarce - Many teams know they should track earned value but don't have dedicated analysts -- 1. Started with schedule parsing - The first challenge was reliably importing P6 XER, MS Project XML, and Excel schedules. Each format has quirks. We built parsers that handle real-world messy data, not just textbook examples. 2. Added automated EVM calculations - Instead of expecting users to enter BCWS/BCWP/ACWP manually, we calculate earned value metrics directly from schedule data—percent complete, baseline vs. actual dates, costs from resource assignments. 3. Built AI analysis on top - Once we had structured schedule data, we could feed it to AI for pattern recognition. The AI identifies issues like missing logic, unrealistic durations, and resource overloads that would take an expert hours to find manually. 4. Expanded to portfolio view - Individual project health wasn't enough. We added organization-wide dashboards showing aggregated metrics, cross-project resource allocation, and trend analysis over time. 5. Made resources first-class citizens - The latest evolution: automatically importing resources from P6 schedules and building staffing forecasts to answer "do we have the right people at the right time?" The through-line: reduce the time from "data exists" to "insight delivered" from days to seconds.