NC 05: Sell-through rate
Track: No-code sandbox (sandbox.altfndata.com) Prerequisite: NC_04.
The task
Calculate a brand's sell-through (clearance) rate, the share of offered lots that actually sold, and compare it across a couple of brands or vendors.
Starter steps
- Open the SQL editor.
- Paste and run this query, which counts sold lots and total offered lots for a single designer, then divides:
SELECT
designer,
COUNT(*) FILTER (WHERE status = 'sold') AS sold_count,
COUNT(*) AS offered_count,
CAST(COUNT(*) FILTER (WHERE status = 'sold') AS DOUBLE) / COUNT(*) AS sell_through_rate
FROM all_watches_data
WHERE designer = 'Patek Philippe'
GROUP BY designer;
- If the sandbox's SQL dialect does not support
FILTER, use the equivalentSUM(CASE WHEN status = 'sold' THEN 1 ELSE 0 END)form instead, the intent is identical: count sold lots over all offered lots. - Re-run with a different designer to compare sell-through rates side by side.
- Try grouping by
vendorinstead ofdesignerto see which auction houses or marketplaces clear a higher share of what they list.
Expected result
A single row per designer (or vendor) with a sold_count, an offered_count, and a sell_through_rate between 0 and 1. A high-demand brand or a strong single-owner sale should show a sell-through rate closer to 1.0 (nearly everything offered found a buyer), while a broader or more mixed set of lots typically clears at a lower rate.
Stretch challenge
Combine designer and vendor in the same GROUP BY to find the single vendor-designer pairing in the data with the highest sell-through rate among pairings with at least, say, 20 offered lots (add a HAVING COUNT(*) >= 20 clause).