Written by supply chain practitioners who have spent their careers running operations and the technology behind them. No vendor spin and no textbook recitals. Just metrics, trade-offs and tools as they behave in the real world, not in the demo.
Start here
DIFOT & delivery performance
What a good score actually is, how to calculate it honestly, and how teams game it once the target has a bonus attached.
Forecast accuracy & bias
MAPE, WAPE, bias and forecast value add, plus the incentive problems underneath the numbers that no algorithm will fix.
AI in the supply chain
Agentic AI, AI forecasting, control towers and digital twins, reality checked by people who have lived the implementations.
Latest articles
- How to Actually Reduce Forecast Bias (It’s Not the Algorithm)Most forecast bias is an incentive problem wearing a modeling costume. A practitioner method: find it by product and customer, map it to who benefits, and separate the forecast from the target before you touch the model.
- Tariff Volatility: The Inventory Decisions That Actually Matter in 2026Buffer inventory is a loan against a guess. The carrying-cost math on stockpiling, when nearshoring actually pays back, and a lane-by-lane framework for deciding between stockpile, nearshore, or pass the tariff through.
- MAPE vs WMAPE vs WAPE: When Each Forecast Metric Lies to YouMAPE and WAPE can tell opposite stories about the same portfolio. The worked example, the two lag-and-predictability cuts that make forecast KPIs actionable, and why bias needs tracking alongside either one.
- DIFOT vs OTIF: Same Metric, Different GamesDIFOT and OTIF share a formula. The measurement rules behind them can swing the same performance by ten points. The five decisions that decide your score, and where customer-experience scorecards are taking this.
- Agentic AI in Supply Chain: A Practitioner’s Reality CheckGartner calls agentic AI a top 2026 supply chain trend. Only 10% of leaders trust it unsupervised. Practitioners who have lived the implementations explain the gap: data gatekeeping, Excel nostalgia, and process maps that say Bob knows what to do.
- Why S&OP Fails (and What AI Actually Fixes)Most S&OP failures are political, not analytical. A practitioner breakdown of the six failure modes, a time-budgeted agenda that fixes them, and an honest look at what AI does and does not change.
Who we are
We are operators, planners and systems people. Most of the industry’s bad advice comes from the gap between those jobs. The people who know the tech have never run a warehouse, and the people who have run a warehouse rarely know the tech. We have done both, and this site exists to say the things certification bodies and software vendors can’t.