An inventory optimization case asks how much stock a company should hold, and where, to avoid stockouts without tying up excess cash in warehouses. The core method is diagnosing whether the problem is too much inventory, too little, or the wrong mix, then rebuilding the stocking policy around service-level targets, demand variability, and the trade-off between holding cost and stockout cost — not just recommending ‘order less’ or ‘order more.’
Inventory optimization cases sound like a math exercise, and candidates often treat them that way — plug numbers into an EOQ formula, get an answer, move on. That instinct misses what the interviewer is actually testing. A client rarely asks “what’s our optimal order quantity” in isolation; they ask because working capital is trapped, a warehouse is overflowing, or a product line keeps stocking out during peak demand. The formula is a small part of the case. The larger skill is diagnosing which stocking problem you’re actually looking at before you touch a calculation.
Why Inventory Cases Reward Diagnosis Over Formulas
Excess inventory and insufficient inventory look similar from a distance — both show up as “inventory is a problem” in the prompt — but they call for opposite fixes. A candidate who jumps straight to calculating an optimal reorder quantity without first establishing whether the client is over-stocked or under-stocked risks solving the wrong problem correctly.
Based on our analysis of operations and cost-reduction case prompts, inventory questions appear as a standalone case type roughly 10% of the time, and as a supporting sub-question inside profitability or working-capital cases far more often — inventory carrying cost is frequently the hidden lever behind a margin-erosion prompt that doesn’t mention “inventory” at all.
flowchart LR
A[Inventory Problem Identified] --> B{Symptom?}
B -->|Cash Trapped, Warehouses Full| C[Overstock Diagnosis]
B -->|Stockouts, Lost Sales| D[Understock Diagnosis]
B -->|Both, by SKU| E[Mix and Segmentation Diagnosis]
C --> F[Rebuild Stocking Policy]
D --> F
E --> F
F --> G[Recommend + Quantify Impact]
Step 1: Diagnose Which Inventory Problem You Actually Have
Before any calculation, establish the symptom. Ask for inventory turnover trend, stockout frequency by product line, and how the client currently sets order quantities. Three archetypes cover almost every inventory case:
| Archetype | Signal in the Prompt | What’s Actually Wrong |
|---|---|---|
| Overstock / working capital | “Cash is trapped,” “warehouses are full,” margin erosion from markdowns | Order quantities or safety stock set too high relative to actual demand variability |
| Understock / stockouts | “Losing sales,” “customers can’t find product,” seasonal demand spikes | Reorder points set too low, or lead time longer than the policy assumes |
| Mix / segmentation | “Some SKUs sell out while others sit,” inventory looks fine in aggregate | A single stocking policy applied uniformly across SKUs with very different demand patterns |
The mix archetype is the one most candidates miss, because aggregate inventory metrics can look healthy while the underlying SKU-level picture is broken — a retailer can carry “normal” total inventory while its top sellers stock out constantly and its long tail collects dust.
Step 2: Segment Demand Before Touching a Formula
Not every SKU deserves the same stocking policy. Apply a simple two-cut segmentation before recommending anything: sales volume (high vs. low) and demand variability (stable vs. volatile).
flowchart TD
A[SKU Segmentation] --> B[High Volume, Stable Demand]
A --> C[High Volume, Volatile Demand]
A --> D[Low Volume, Stable Demand]
A --> E[Low Volume, Volatile Demand]
B --> F[Tight reorder points, low safety stock]
C --> G[Higher safety stock, frequent review]
D --> H[Larger batch orders, infrequent review]
E --> I[Candidate for discontinuation or make-to-order]
High-volume, stable-demand items (think: a grocery chain’s best-selling staples) can run lean — demand is predictable enough that a small safety buffer covers most variability. High-volume, volatile-demand items (seasonal apparel, promotional electronics) need a larger buffer because a stockout on a popular item costs real revenue. Low-volume, volatile-demand items are frequently the ones driving the overstock problem: they’re ordered in the same batch sizes as everything else even though a stockout barely matters and a leftover unit sits on a shelf for months.
Step 3: Build the Trade-Off, Not Just the Number
Once you know which segment you’re solving for, the core trade-off is holding cost against stockout cost — and the interviewer wants to see you frame this as a trade-off before you calculate anything.
Holding cost includes warehouse space, capital tied up in unsold goods, insurance, obsolescence risk, and markdown risk for perishable or seasonal items. Stockout cost includes lost sale margin, and — for repeat-purchase businesses — a harder-to-quantify customer loyalty cost when a shopper who can’t find a product switches to a competitor.
The standard quantitative building blocks:
- Economic Order Quantity (EOQ): balances ordering cost against holding cost to find the order size that minimizes total cost. Useful for framing, but treat it as a starting point — interviewers care more about whether you can name the inputs (demand rate, ordering cost, holding cost per unit) and reason about which lever moves the answer than whether you compute the exact formula from memory.
- Safety stock: the buffer held above expected demand during lead time, sized to hit a target service level given demand variability. Higher demand variability or longer, less reliable lead times both push safety stock up.
- Reorder point: expected demand during lead time plus safety stock — the inventory level that triggers a new order.
You don’t need to derive these from scratch in the room. What separates strong answers is connecting the formula to the business question: “if lead time variability is the real driver of our stockouts, the fix is a safety-stock adjustment or a supplier-reliability conversation — not a bigger blanket order across every SKU.”
Worked Example: A Consumer Electronics Retailer
Prompt: A consumer electronics retailer’s inventory turnover has dropped from 8x to 5x over two years, tying up an additional $40M in working capital, while customer complaints about stockouts on their three best-selling products have doubled.
Diagnosis: This is the mix archetype — aggregate inventory is up (turnover falling), but the complaint is about stockouts on specific high-volume items. A uniform stocking policy is very likely producing both symptoms at once: over-ordering slow movers while under-ordering the fast movers actually driving revenue.
Segmentation: Pull SKU-level sales velocity and variability. Likely finding: the three best-sellers are high-volume, moderately volatile (new product cycles, promotional spikes), while a long tail of older SKUs is low-volume and slow-moving but still ordered in similar batch sizes.
Recommendation: Raise safety stock and review frequency specifically on the top sellers to close the stockout gap; apply tighter reorder quantities — or a clearance/discontinuation plan — on the slow-moving tail to unlock working capital. Quantify: if the long tail represents, say, $50M of the $40M+ working-capital increase, a partial liquidation funds the safety-stock increase on the best-sellers without net new cash — turning a “we need more capital” case into a “we need to reallocate what we have” recommendation, which is the stronger answer interviewers are listening for.
How to Practice This Case Type
Reading a framework is not the same as producing one live under time pressure — the mix-archetype trap in the worked example above is easy to spot on paper and easy to miss when you’re the one holding the marker. CasesCoach’s AI mock interview runs this exact case type against you and scores the attempt across five dimensions: Structure, Analysis, Communication, Business judgment, and Math.
Two modes fit different stages of prep. Guided Mode nudges you if you jump to a formula before segmenting demand — useful while you’re still internalizing the diagnosis-before-calculation habit this case type demands. Expert Mode removes the hints and tightens pacing, closer to what an actual second-round interviewer does when you propose “just order more” without first asking whether the client is overstocked or understocked. If you want a lower-pressure way to test your framework against a real prompt first, Chat with Case walks the same background → framework → data → analysis → conclusion sequence through a real case from the library, at your own pace, before you take it into a timed mock.
The free tier includes 3 full cases and a free AI Mock Interview attempt, so you can see how the scoring works before committing. Pro unlocks the full 835+ case library and multiple additional AI Mock attempts.
Key Takeaways
- Diagnose before you calculate: overstock, understock, and mix problems look similar in the prompt but call for opposite fixes
- Aggregate inventory metrics can hide a segmentation problem — a company can look “fine” on turnover while its best-sellers stock out and its slow movers pile up
- EOQ, safety stock, and reorder point are the core quantitative tools, but interviewers weight your ability to connect them to the business trade-off over exact formula recall
- The strongest recommendations reallocate existing inventory investment rather than asking for more capital
- Practice this case type with an AI mock interview scored across Structure, Analysis, Communication, Business, and Math — or start with Chat with Case to walk through a real case at your own pace before timing yourself
Ready to test your inventory framework against a real case? Sign up free for 3 complete cases and a free AI Mock Interview, or explore the operations case framework for the broader supply-chain toolkit this case type sits inside.