Project
Ops Control Center
Founder / Head of Growth & Ops, Dansu
Built an Ops Brain covering fulfilment anomaly detection, stock forecasting, supplier scorecards, QC workflow, and customer pipelines — replacing disconnected manual checks with exception-based visibility.
- Fixed a broken supply chain workflow by mapping the full chain from manufacturer to doorstep, identifying where data gaps let problems go undetected until they became customer complaints.
- Reduced missing orders, cut fulfilment errors, improved product quality consistency, and dropped customer service load — all without adding headcount.
Overview
Early in Dansu's life, operations were invisible. I had no reliable view of where orders sat in the supply chain, when stock was going to run low, which suppliers were creating problems, or whether a parcel was on its way or lost somewhere between warehouse and doorstep. I built the Ops Brain to fix that — an operations control layer that centralised fulfilment tracking, stock forecasting, customer pipelines, and supplier communications into a single view driven by automated alerts.
The Problem
Missing parcels, delays, inconsistent supplier quality, and QC issues were appearing without warning. I found out about problems from customer complaints — not from system alerts. There was no visibility into where failures were occurring across the chain from manufacturer to delivery, and no tooling to surface anomalies before they became customer-facing. Firefighting was consuming time that should have gone toward growth.
What I Did
I started by mapping the full supply chain end to end: manufacturer → packaging → storage → pick-and-pack → courier scan → delivery. At each handoff I identified where data existed, where it didn't, and where delays clustered. Then I built tooling in n8n and Supabase to instrument each stage.
Fulfilment tracking: Automated checks against courier scan data to flag orders that had missed expected scan windows. Anomalies surfaced in a dashboard rather than waiting for customer contact. Exception-based rather than requiring manual daily checks.
Stock forecasting: Modelled sell-through rates by SKU against inventory levels, with automated alerts when stock was projected to hit reorder points. Prevented the stockouts that had previously appeared suddenly mid-campaign.
Supplier scorecards: Structured records of supplier performance across quality, lead time, and communication reliability. Made it easier to identify patterns and manage supplier conversations with evidence rather than memory or gut feel.
QC workflow: Introduced a standardised intake inspection process for new stock arrivals with documented pass/fail criteria. Reduced the rate of quality issues reaching customers by catching them at intake rather than at return.
Packaging redesign: Rebuilt packaging specifications to reduce damage rates in transit and reduce picking errors at the fulfilment stage.
Customer pipeline: Tracked open customer service cases in Supabase with status and resolution tracking so nothing fell through the cracks during high-volume periods. Cases were visible and assignable rather than buried in email threads.
Impact
- Reduced missing and delayed orders significantly through proactive anomaly detection
- Cut fulfilment errors through improved QC intake process and packaging specifications
- Improved product quality consistency across manufacturing runs
- Customer service load dropped as proactive detection replaced reactive complaint handling
- Freed operational attention for growth work — less firefighting, more building
- Full supply-chain visibility made the business easier to hand over as part of the eventual exit