How would you optimize the product assortment for an e-commerce category?
This is a product strategy question. Not an "add more SKUs" answer. More selection is not better selection. Find where customers search and don't find, and where the catalogue is full of products nobody buys.
Step 1. Clarifying Questions
- Which category? Assume kitchen appliances on Flipkart.
- What does "optimise" mean: GMV, conversion, margin, satisfaction? Assume conversion and GMV, with margin as a guardrail.
- Marketplace or own inventory? Assume marketplace. We shape assortment through sellers and brands, not by buying stock.
Step 2. Why it matters
Assortment decides whether a search ends in a purchase. If the customer can't find the mixer grinder they want at their price, they buy elsewhere, and may not come back for the next purchase either.
Step 3. Break assortment down
Assortment = Breadth (sub-categories, price bands, brands) × Depth (variants: size, colour, wattage) × Quality (ratings, returns, availability)
Map the category on a grid: use case × price band. Example: mixer grinders × under ₹2,000 / ₹2,000 to 5,000 / above ₹5,000.
Step 4. Demand side: find the gaps
- Zero-result and low-result searches
- Searches where results are shown but nothing is clicked
- Filters that return very few items
- Products customers ask about in Q&A that we do not carry
- Regional needs: wet grinders in the South, roti makers in the North
- Competitor selection in the same grid cells
Step 5. Supply side: find the dead weight
- SKUs with no sale in 90 days
- The same product listed by many sellers, splitting reviews and confusing buyers
- SKUs with high return rates or low ratings
- Top sellers that keep going out of stock
Step 6. Actions
1. Fix: keep the top 20% of SKUs in stock. They drive most of the GMV.
2. Grow: recruit brands and sellers for empty grid cells. Rank gaps by search volume × margin.
3. Prune: delist or de-rank dead and high-return SKUs. Merge duplicates into one listing.
4. Curate: a short, clear set per price band, so customers are not lost in 400 near-identical options.
Prioritise: availability of top sellers first. Cheap and fast. Then fill high-search gaps. Pruning runs in the background.
Step 7. Metrics
- North Star: search-to-purchase conversion in the category
- Zero-result search rate, % of SKUs with a sale in 90 days, GMV per live SKU
- In-stock rate of top SKUs, return rate
- Guardrail: margin per order, seller churn
Trap to call out: optimising GMV per SKU alone pushes you to cut the long tail. The long tail is often why customers trust the store to have everything.
Interview tip: cover both sides, the gaps and the dead weight. Most candidates only talk about adding products.
Community Answers (0)
No answers yet. Be the first to share your approach!