Related Ideas
eyko Ideas
Demand forecasts built SKU by SKU miss the linkages where one product's demand predicts another's. A Demand Correlation Detection Playbook reads cross-SKU and cross-customer demand patterns to surface the correlations that should be reflected in planning and inventory decisions.
The Challenge
When SKU A and SKU B are typically bought together, a demand shift on A predicts a shift on B. SKU-by-SKU forecasting treats them as independent series, so the inventory decision on B lags the signal that was already visible on A weeks earlier.
When SKU A is stocked out, customers substitute SKU B. The substitution lifts B's baseline temporarily, the forecasting model learns the new baseline, and B gets over-stocked when A comes back. Without an explicit substitution map, the model trains itself into bullwhip.
Demand in customer segment X correlates with demand in segment Y because both serve the same downstream market. The forecast treats them separately and misses the upstream signal that would let planning move ahead of both at once.
How eyko Solves It
A Demand Correlation Detection Playbook reads cross-SKU and cross-customer demand patterns over time to identify the correlations that connect product lines, customer segments, and channels. It produces a correlation map per SKU showing the strongest predictive linkages, distinguishes complementary from substitution patterns, and feeds the linkages into the forecasting and inventory planning workflows so SKU decisions reflect upstream demand signals rather than each SKU's own history alone.
The Playbook analyzed 18 months of demand data across 4,200 SKUs and identified 380 correlation pairs with statistical significance. 142 pairs are complementary (move together), 96 are substitution pairs (move opposite), 86 are upstream-downstream (one leads the other by 2 to 6 weeks), and 56 are cross-segment correlations. The strongest upstream-downstream pair predicts the secondary SKU's demand 4 weeks in advance at 78% confidence.
| Metric | Current | Benchmark | Status |
|---|---|---|---|
| Primary indicator | Flagged | Target | Action needed |
| Secondary indicator | Monitoring | Within range | On track |
| Trend direction | Declining | Stable | Review required |
Demand Correlation Detection identifies the cross-SKU and cross-customer demand correlations that connect product lines, customer segments, and channels. The Playbook produces a correlation map per SKU showing the strongest predictive linkages, distinguishes complementary from substitution patterns, and feeds the linkages into forecasting and inventory planning so SKU decisions reflect upstream demand signals rather than each SKU's history alone.
Related Ideas
FAQ
Everything you need to know about Demand Correlation Map.
Demand Correlation Detection is an AI-driven analysis that identifies cross-SKU and cross-customer demand correlations connecting product lines, customer segments, and channels. The Playbook produces a correlation map per SKU, distinguishes complementary from substitution patterns, and feeds the linkages into forecasting and inventory planning so SKU decisions reflect upstream demand signals rather than each SKU's history alone.
The Playbook reads from your ERP or sales system (per-SKU sales velocity, customer cohort metadata, channel breakouts), product master data (bill of materials, complementary mappings where available), inventory system (stockout periods that trigger substitution), and at least 18 months of paired demand history. The richer the SKU and customer metadata, the more nuanced the correlation map.
Market basket analysis identifies products bought together in a single transaction, mostly in retail contexts. Demand Correlation Detection works on demand time series across longer windows and includes upstream-downstream linkages (one SKU leading another by weeks), cross-segment correlations, and substitution patterns. The two are complementary, but correlation detection is the layer that feeds the forecasting model rather than the merchandising layout.
Yes. The Playbook recommends integrating top correlation pairs into the demand forecasting model, updating the substitution map in inventory planning, and briefing category managers on upstream-downstream pairs. Each recommendation projects forecast accuracy lift and inventory smoothness improvement so leadership can prioritize the highest-impact linkages first and validate the impact on backtests before scaling.
Join the enterprises replacing weeks of manual analysis with a single prompt. See what eyko Playbooks can do with your data.
