How do I approach Analytics & Experimentation interview questions?

Analytics & Experimentation questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master analytics & experimentation interviews.

What difficulty level is this interview question?

This is a hard difficulty Analytics & Experimentation question, commonly asked during Technical Screen rounds at Capital One.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at Capital One during technical interviews.

Evaluate launching a vegan burger | Capital One Interview Question

Q: Evaluate launching a vegan burger

Capital One data-scientist technical-screen case: a rival launched a hit vegan burger -- should your fast-food chain add one, and how would you test it? It exercises pricing and willingness-to-pay/elasticity, operational readiness and capacity, cannibalization vs. incrementality, marketing and labeling, success metrics with guardrails, and a matched store-level RCT (difference-in-differences, CUPED, power) feeding an explicit go/no-go rule.

Scenario

You run a fast-food burger chain, and a rival has just launched a hit vegan burger. The question on the table: should you add a vegan burger to your own menu?

Your job is to lay out the decision framework, then design the experiment you would run to make the call.

Clarifying Questions to Ask

Before diving in, scope the problem with the interviewer:

Objective & horizon. Are we optimizing for incremental profit, traffic/market share, or brand positioning — and over what time window (launch quarter vs. steady state)?
Scale & footprint. How many stores, and across how many regions/formats (drive-thru vs. dine-in)? Is this a national chain or a regional test?
Current data. Do we have store-level POS data, item-level margins, and a loyalty program (customer IDs) we can use for incrementality and matching?
Constraints on the test. What's the budget and timeline for a pilot, and is leadership willing to wait for a measured readout or do they want a fast launch?
Risk tolerance. How sensitive is the brand to a food-safety/allergen incident or to cannibalizing high-margin core items?
Competitive pressure. Is the rival's launch eroding our traffic now, or is this a proactive bet?

Constraints & Assumptions

State the assumptions you're anchoring on (and flag that real numbers come from the pre-period data):

Assume a multi-store chain with store-level POS and item-level cost/margin data available, plus a loyalty program for customer-level signals.
Assume you can run a store-level or geo-split pilot for roughly 6–8 weeks plus a baseline period.
Assume a target gross margin consistent with the existing burger line, a defined finance hurdle for launch, and the ability to source a plant-based patty from at least one qualified supplier.
Assume kitchens use shared equipment (so "vegan" vs. "plant-based" labeling and cross-contact matter).
Treat the operational, competitive, and regulatory inputs as real constraints on whether the launch makes money — not as a checklist.

What a Strong Answer Covers Premium

Decision framework — inputs to evaluate

Work through each input as a constraint on whether the launch makes money, not as a box to tick.

1. Pricing and willingness-to-pay

How would you set the price (value-based vs. competitive, target margin)? Estimate willingness-to-pay and own-/cross-price elasticity — e.g. survey methods like Van Westendorp or Gabor–Granger, conjoint analysis, and an in-market price test across matched stores.

2. Operational readiness

Assess prep time and line capacity (time-and-motion study, bottleneck/throughput analysis), crew training and SOPs, cross-contamination / food-safety controls, and supplier risk (sourcing, fill-rate SLAs, safety stock).

3. Capacity constraints

Model station utilization (grill, fryer, assembly) at peak with the new SKU; identify bottlenecks and mitigations before scaling.

4. Competitive landscape and differentiation

Size the addressable demand and segments (vegans/vegetarians, flexitarians, health- and sustainability-minded), map competitor offerings, and define how you would differentiate.

5. Demand forecasting — cannibalization vs. true incrementality

Forecast demand and separate net-new traffic from substitution away from your existing beef/chicken burgers and sides.

6. Marketing implications

Target segments, messaging (taste-first vs. sustainability), channels, and launch creative/offer variants to test.

7. Regulatory and labeling

Allergen labeling (soy, gluten, pea protein), nutrition disclosure, "vegan" vs. "plant-based" claims, and shared-equipment / cross-contact disclaimers.

Experiment design — how you'd make the call

8. Success metrics

Define a primary metric (e.g. incremental gross profit / contribution margin per store-week) plus guardrails (throughput/service time, order accuracy, NPS/CSAT, stockouts, waste, food-safety incidents).

9. Experiment design

Specify a geo-split or store-level randomized test: matched test vs. control stores, randomization, duration, KPIs, success thresholds, the estimator (difference-in-differences with variance reduction), and a sample-size / power calculation.

10. Go / no-go decision rule

State the explicit rule that turns the experiment readout into a launch, iterate, or stop decision, and name the key risks and mitigations.

Follow-up Questions

Be ready for the interviewer to push deeper:

Scarce stores. If you can only get a handful of pilot stores, how do you still reach adequate power — and when would you reach for a Bayesian analysis or synthetic control?
Novelty fade. The first two weeks look great, then sales sag. How do you tell genuine decay-to-steady-state apart from a real demand problem?
Conflicting guardrails. The primary metric clears the hurdle but peak service time degrades just past the bound. Launch, iterate, or stop — and why?
Scaling. How does your measurement plan change when you roll out to 100x the stores, where marketing and supply spillovers across regions become real?