Production Visibility & KPI Platform
Took a 200+ employee plant from zero production visibility to real-time dashboards and structured KPI reporting.
The Problem
The company president was manually exporting data from the ERP, feeding it through AI tools, and building pivot tables by hand. That process took 30+ minutes per product type, one customer at a time. That was the best visibility leadership had into what their 200+ employee plant was producing.
On the floor, it was worse. Shift supervisors tracked goals, actuals, downtime, and utilization in Excel spreadsheets saved to a shared drive. Two years of that had produced roughly 2,000 files, nearly half full of stale or duplicated data from copy-paste errors. None of it tied back to the ERP.
Leadership had identified 30 KPIs they needed to run the business: production rates, backlog, quality, waste, capacity. Not one existed in any reporting system. Different managers were requesting variations of the same data from different people, getting different answers, and making decisions based on whichever number showed up last.
Meanwhile, a third-party consultant was advising on six-figure equipment purchases, and leadership had no way to check those recommendations against their own numbers. The plant couldn't answer the most basic question in manufacturing: what did we actually produce today?
What We Built
Data Recovery & Reconciliation
First job was rescuing two years of production history from ~2,000 inconsistent spreadsheets. We built an ingestion pipeline that normalized formats, flagged errors the manual process had been quietly propagating, and loaded everything into a structured database. Tedious, necessary work. Nothing else could move forward until the data was trustworthy.
The floor reports and the ERP didn't speak the same language: different naming conventions for work centers, different granularity for time periods, different definitions of what counted as "production." We built a mapping layer to reconcile the two systems, then surfaced mismatches in both directions so leadership finally had checks and balances across their data sources. Operations used the discrepancies to tighten up shift reporting processes that had been drifting unnoticed.
Structured Data Capture
Replaced the manual Excel tracking process with a purpose-built entry form tied directly to the ERP. Shift supervisors capture the same production metrics they always tracked (goals, actuals, downtime, utilization) but now with validation, structure, and a direct link to the system of record. No more copy-pasting between spreadsheets on a shared drive. Data goes in clean, once, and flows straight into the reporting layer.
Production Metrics Platform
The core of the project. We built a metrics engine directly on top of the ERP data: throughput by work center, shift, and checkpoint; scheduled vs. actual production time; queue state and late tracking at the order level. Refreshes every 15 minutes. Sub-second drill-down from plant-wide totals to individual orders. Not a reporting layer bolted onto spreadsheets; an actual system of record for production performance.
KPI Feasibility & Phased Rollout
Mapped all 30 requested KPIs against the data actually available across the ERP and the new reporting systems. Seven were immediately buildable. Ten were possible but needed new data collection. Thirteen required investigation into whether the underlying data was even being captured. That mapping alone eliminated months of circular requests for reports that couldn't exist yet. Delivered the first tier within weeks, then worked through the remaining tiers in priority order so leadership saw progress fast instead of waiting on a single monolithic delivery.
Executive Reporting Suite
Built a full reporting dashboard with six interconnected reports covering summary output, machine performance trends, utilization, downtime analysis, first pass yield, and maintenance logging. Every report drills down from plant-wide totals to individual order detail, with sub-second response times that often beat pulling the same data directly from the ERP.
What used to take the president half an hour per product type (for a single customer) now covers every product line and customer in the plant automatically, refreshing throughout the day.
Delivered in SSRS to meet the business where they were. The data layer underneath was built to be fully portable. When their team moves to Power BI, the entire metrics foundation lifts and shifts without rework.