Shelly guide

Industry packs: beyond retail

Each pack is a set of skills with its own demo data, findings and decision proposals. Swap in real data through connectors.

At a glance

Electronics: sell-through, weeks of cover, aged-stock ladder with a below-cost markdown check, ASP erosion, accessory attach rate
Wholesale: customer contribution after cost-to-serve, AR ageing and DSO, fill rate and OTIF
Warehousing: ABC (velocity) × XYZ (variability) slotting, lines per hour with coaching flags, weeks until 95% full
Production: OEE = availability × performance × quality, scrap cost, schedule adherence

Run a pack: python -m shelly pack electronics (or wholesale, warehousing, production).
Findings become actions; major ones become decisions for confirmation.
Full reports are in docs/example/pack_*.md.

Weeks of cover

on hand ÷ average weekly units (4 weeks)

Cost-to-serve

orders × cost per order ($18 default)

DSO

receivables ÷ 90-day sales incl. GST × 90

OEE

(run time ÷ planned) × (units ÷ (run time × ideal rate)) × (good ÷ units)

In electronics, stock loses value every week it sits, so aged stock and weeks of cover matter more than turnover.
In wholesale, the biggest customers aren't always the most profitable once cost-to-serve is included.
In production, fix the biggest OEE loss first (availability, speed or quality).

Pack demo data is synthetic. Cost assumptions (cost per order, unit cost) are placeholders to replace.

Tell me about warehousingAny electronics markdowns?What is our OEE?

Type or say these in the chat on the main page.