How we reduced inventory costs by $180K annually for a manufacturing company with demand forecasting and inventory optimization.
A mid-sized manufacturing company producing industrial parts was experiencing the worst of both worlds: frequent stockouts on high-demand items while simultaneously having $200K+ tied up in excess inventory that wasn't moving. Inventory decisions were based on gut feeling and "we've always ordered this much."
The operations team had no visibility into actual demand patterns, seasonal trends, or supplier lead times. They were constantly firefighting—expediting orders to cover stockouts while watching warehouse space fill with slow-moving parts. Cash flow was suffering and customer satisfaction was declining.
We built a demand forecasting and inventory optimization system that analyzed historical sales data, seasonality patterns, and supplier lead times to automatically calculate optimal reorder points and order quantities for each SKU across all warehouse locations.
Consolidated 3 years of sales data, inventory levels, purchase orders, and supplier lead times from their ERP system (NetSuite) into PostgreSQL. Cleaned inconsistent SKU naming and categorization.
Built time-series forecasting models using Python (statsmodels) that identified seasonal patterns, trends, and demand volatility for each SKU. Generated 90-day demand forecasts updated weekly.
Created dbt models to calculate optimal reorder points, safety stock levels, and economic order quantities based on forecasted demand, supplier lead times, and desired service levels.
Built Looker dashboards showing: inventory levels vs. reorder points by warehouse, demand forecasts vs. actuals, supplier performance metrics, and slow-moving inventory identification. Integrated Excel export for purchasing team.
Within 5 weeks, the company went from flying blind to having data-driven inventory management. Over the next 6 months, they reduced excess inventory by 45% (freeing up $180K in working capital) while simultaneously reducing stockouts by 60%.
This system has been transformational. We went from constantly running out of critical parts while drowning in slow-moving inventory to having exactly what we need when we need it. We've freed up $180K in working capital that was sitting on shelves, and our stockout rate is down 60%. The forecasting model even caught seasonal trends we didn't know existed. This paid for itself in 3 months.
Included forecasting model, optimization engine, dashboards, Excel integration, documentation, and 2 months of support.