Data analytics cases in retail and consumer goods interviews test whether you can translate messy consumer data into actionable business decisions. Based on our analysis of consulting interview trends, analytics-driven retail cases have increased roughly 40% since 2022 — reflecting the industry’s $15–20 billion annual investment in data capabilities across MBB and Big Four client portfolios.
Why Retail Analytics Cases Are Different
Retail generates more granular transaction data than almost any other industry — a mid-size grocery chain processes 50–100 million SKU-level transactions weekly. The challenge interviewers set isn’t “do you understand data science” but rather “can you identify which data question, answered well, would unlock the most business value?”
The decision framework for retail analytics cases follows a distinct logic:
flowchart TD
A[Business Problem Identified] --> B{Data Available?}
B -->|Yes| C[Define Key Metric]
B -->|No| D[Estimate or Proxy]
C --> E[Segment & Analyze]
D --> E
E --> F{Actionable Insight?}
F -->|Yes| G[Quantify Impact]
F -->|No| H[Refine Segmentation]
H --> E
G --> I[Recommend & Prioritize]
In our experience coaching candidates through retail analytics cases, the most common failure mode is jumping to sophisticated techniques (machine learning, propensity modeling) before clarifying the business problem. Interviewers reward candidates who start with “what decision will this analysis inform?” rather than “what algorithm should we use?”
Five Core Analytics Case Types in Retail
| Case Type | Typical Prompt | Key Data You’ll Receive | What Interviewers Assess |
|---|---|---|---|
| Customer Segmentation | “Which customer cohorts should we invest in?” | Transaction history, demographics, channel usage | MECE segmentation, value-based prioritization |
| Demand Forecasting | “How much inventory should we hold for Q4?” | Historical sales, seasonality indices, external factors | Statistical reasoning, business judgment on uncertainty |
| Pricing Analytics | “Should we raise prices on our premium line?” | Price elasticity data, competitor pricing, margin splits | Sensitivity analysis, second-order effects |
| Promotional Effectiveness | “Are our promotions profitable?” | Lift data, baseline sales, cannibalization metrics | Incrementality thinking, ROI calculation |
| Location Analytics | “Where should we open our next 20 stores?” | Demographic data, traffic patterns, competitor density | Scoring methodology, trade-off articulation |
Customer Segmentation: The Most Tested Pattern
Customer segmentation cases appear in approximately 30% of retail analytics interviews at McKinsey and BCG. The interviewers’ goal is to see whether you can move beyond basic demographic cuts (age, income) into behavioral segmentation that drives differentiated strategy.
A proven structure for retail customer segmentation:
mindmap
root((Customer Segmentation))
Behavioral
Purchase frequency
Basket size
Channel preference
Category affinity
Value-Based
Current spend
Growth potential
Cost to serve
Retention risk
Attitudinal
Price sensitivity
Brand loyalty
Convenience priority
Sustainability preference
The 80/20 trap: In our work with retail cases, candidates often cite the “top 20% of customers drive 80% of revenue” rule — then stop. Interviewers want you to go further. The actionable question is: within the remaining 80% of customers, which sub-segment has the highest migration potential to higher-value tiers, and what would trigger that migration?
Worked Example: Grocery Loyalty Program
A grocery retailer’s loyalty program has 5 million members. Average spend is $85/week but varies enormously. The CEO asks: where should we invest to grow same-store sales by 3%?
Step 1 — Segment by value and trajectory:
| Segment | % of Members | Avg Weekly Spend | Trend (YoY) | Strategic Priority |
|---|---|---|---|---|
| Champions | 8% | $210 | +2% | Retain — high value, stable |
| Rising Stars | 12% | $130 | +15% | Accelerate — highest ROI opportunity |
| Steady Core | 45% | $75 | 0% | Nudge — small uplift at scale |
| Declining | 20% | $60 | -12% | Diagnose — win-back or release |
| Dormant | 15% | $15 | -30% | Deprioritize — low expected value |
Step 2 — Size the prize: Rising Stars (12% × 5M = 600K members) spending $130/week with +15% growth trajectory. Accelerating their migration by just 5 percentage points = $600K × $6.50/week additional spend = ~$200M annualized revenue.
Step 3 — Identify intervention: What drives Rising Stars’ growth? Analysis reveals 70% adopted online ordering in the past year. Recommendation: expand click-and-collect capacity in top 50 stores serving Rising Star concentrations.
Demand Forecasting: Combining Quantitative and Qualitative
Demand forecasting cases test your ability to structure uncertainty. Interviewers rarely expect you to build a model on the spot — they want to see that you understand what inputs matter, what drives forecast error, and how business decisions should adapt to uncertainty ranges.
Key principles interviewers expect:
- Decompose demand drivers: Base demand × seasonality factor × trend × promotional uplift × external shocks
- Acknowledge asymmetric risk: In retail, stockout cost (lost sales + customer defection) typically exceeds overstock cost (markdowns + carrying cost) by 2–4x for high-velocity items
- Segment forecast accuracy: A-items (top 20% of SKUs by revenue) should forecast at 85–95% accuracy; C-items (long tail) may only achieve 50–60% — and that’s acceptable
Red flag interviewers watch for: Candidates who present a single-point forecast without discussing confidence intervals or scenario ranges. Always say: “My central estimate is X, but the realistic range is Y to Z, which means our decision should be robust across that range.”
Promotional Effectiveness: The Incrementality Question
Roughly 25–30% of grocery revenue flows through promotional pricing, yet based on industry analysis, 20–30% of trade promotion spending generates zero incremental profit. This makes promo effectiveness a high-stakes analytics case.
The core analytical framework:
| Metric | Definition | Good Benchmark | Poor Benchmark |
|---|---|---|---|
| Incremental Lift | Sales above baseline during promo | >30% | <10% |
| Post-Promo Dip | Sales decrease after promo ends | <15% of lift | >50% of lift |
| Cannibalization Rate | Sales stolen from adjacent products | <20% | >40% |
| Pantry Loading | Customers stockpiling, shifting future demand | <25% of lift | >60% of lift |
| Net ROI | (Incremental margin - promo cost) / promo cost | >15% | Negative |
The mistake most candidates make: Calculating promo “success” by comparing promo-period sales to the prior period. This ignores baseline growth, cannibalization, and pantry loading. Interviewers reward candidates who immediately ask: “Are we measuring against a counterfactual baseline — what would have happened without the promotion?”
Location Analytics: Scoring and Trade-Offs
When a retailer asks “where should we open next?” the analytical challenge isn’t finding good locations — it’s ranking them and managing cannibalization with existing stores.
A typical scoring framework interviewers accept:
| Factor | Weight | Data Source | Consideration |
|---|---|---|---|
| Demographic fit | 25% | Census data, spending patterns | Does the trade area match target customer profile? |
| Traffic density | 20% | Mobile location data, foot traffic | Raw potential for awareness and visits |
| Competitive intensity | 20% | Competitor mapping, market saturation | Blue ocean vs. head-to-head battle |
| Cannibalization risk | 15% | Distance to existing stores, overlap | Net new vs. redistributed revenue |
| Real estate economics | 20% | Rent/sqft, lease terms, buildout cost | Payback period sensitivity |
The critical follow-up question interviewers love: “How would you validate this model before committing $50M in capital?” Strong answers mention A/B testing with pop-up locations, analog store analysis (finding existing stores in similar trade areas), and sensitivity testing on the highest-weighted factors.
Key Takeaways
- Retail analytics cases test business judgment, not technical sophistication — always start with “what decision does this inform?”
- Customer segmentation is the most common pattern (~30% of analytics cases); go beyond demographics into behavioral and value-based cuts
- For demand forecasting, always present ranges rather than point estimates, and acknowledge asymmetric stockout vs. overstock costs
- Promotional effectiveness requires incrementality thinking — measure against counterfactual baselines, not prior periods
- Location analytics is a scoring and trade-off exercise; show interviewers you can weight factors, then validate before committing capital
- In every analytics case, quantify the business impact in revenue or margin terms — an insight without a dollar figure attached rarely drives a recommendation
Apply These Frameworks
Practice retail analytics thinking with real scenarios in our retail industry cases and consumer goods cases. For the underlying analytical structures, review the profitability framework guide and our pricing strategy guide. When you’re ready to test your approach under time pressure, try a timed session with our AI Mock Interview — analytics cases reward structured delivery as much as structured thinking.