Cornerstone Research Case Interview 2026: Practice Cases
A Cornerstone case interview is a litigation problem, not a strategy case: the Cornerstone cases you get in the Cornerstone Research interview ask whether a stock drop, a patent breach or a price rise really caused the harm someone claims, and how much. The Cornerstone practice cases on the firm's own site (Zilo and Duff) show that style, and every Cornerstone Research case interview answer starts from the claim, then the data that would prove or disprove it. If you are comparing Analysis Group vs Cornerstone Research, the biggest difference is the case: Cornerstone gives one in every round.
Start a free case interview for Cornerstone Research

Don't know how to case yet? Learn how to case
How does the Cornerstone Research interview process work?
You apply with a cover letter, resume and transcript, then sit two rounds: two virtual 30-minute interviews, then four 45-minute interviews in a multi-hour final round. That is the process on Cornerstone's interview process page as of November 3, 2026, and the Cornerstone Research analyst interview is identical for Analyst and Summer Analyst candidates. Click a stage to see what Cornerstone tests there and the practice that matches it.
From application to Cornerstone Research offer
Send all three documents
A cover letter with your office preferences, a resume and a transcript. Applications with missing materials are not considered.
Cornerstone recruits through country portals for the US, the UK and Belgium, and US-based students are not considered for UK or Belgium openings, according to its how to apply page. On-campus events and submission deadlines sit on the firm's analyst recruiting calendar rather than on one fixed date, so check your school's listing.
Causation-style practice cases at a glance
How a Cornerstone case works: it is a litigation problem that asks whether a drop, a breach or a price rise really caused the harm someone claims, and how much, so practise cases about what caused a decline and what the data would show.
Start free now: Fruit Drink Turnaround is a BCG-style turnaround case adapted from the Wharton casebook: a pomegranate drink is losing market share and earning less than its main competitor. It has 6 questions, 1 with math, 1 exhibit, and a median of 12 minutes across 25 completed voice runs. An AI interviewer runs it out loud, shows each exhibit when you reach it and scores the debrief. Your first full case is free when you create an account.
| Case (style, industry) | Quant difficulty | Qualitative difficulty | The math you run | Median time |
|---|---|---|---|---|
Fruit Drink Turnaround (BCG, consumer packaged goods). 6 (1 with math), 1 exhibit. | easy | easy | Cost per pack gap, driver split | 12 min, 25 voice runs |
High Q Plastics (Bain, automotive). 7 (2 with math), 3 exhibits. | hard | medium | Plant profit scenario, truck share revenue | Fewer than 10 runs |
Thomas the Tank Rail Revenue Decline (McKinsey, rail). 7 (1 with math), 1 exhibit. | easy | medium | Ticket volume and class mix shift | Fewer than 10 runs |
Race to the Bottom Airline Profitability (McKinsey, airlines). 7 (2 with math), 2 exhibits. | medium | medium | Profit decline, revenue per flight | 20 min, 13 voice runs |
All four are adapted from consulting club casebooks (Wharton and Kellogg). The library has no case in Cornerstone Research's own style, so these are matched by case type to what Cornerstone Research asks. None is Cornerstone Research's own material. More are in the case library, 92 cases in all. Every debrief scores seven categories: structure (25% of the score), quantitative analysis (20%), business judgment (15%), and problem understanding, exhibit reading, communication and synthesis (10% each), with feedback on every question. Median times come from completed runs between June 25 and September 22, 2026, and count runs, not people.
What does the Cornerstone Research case interview look like?
Each interview combines your background, behavioral and situational questions, one case to review and discuss, and time for your questions, and each round brings a new case. Cornerstone's Analyst and Summer Analyst interview page says the interviews test your thought process, comprehension and how clearly you explain ideas, and its interview process page adds your ability to take and apply feedback in real time (checked November 3, 2026). Candidates describe the case as interviewer-led, and Cornerstone's own tips point the same way: listen to your interviewers, who are there to guide you and keep you on track.
First round
virtual
- Two interviews of 30 minutes each
- Background, behavioral questions, a case, Q&A
Final round
multi-hour
- Four interviews of 45 minutes each
- A new case, plus lunch or a chat with staff
In an office, or virtually
The case
every round
- A litigation problem to review and discuss
- Firm examples: a stock drop, a patent claim
What happens in the Cornerstone Research first round interview?
Two 30-minute virtual interviews, each with the same four parts. Candidates report questions such as "why economic consulting rather than management consulting?", a walk-through of a research or data project, and a short case built around a dispute. Cornerstone's interview page quotes a Washington manager who tells candidates to give the behavioral part as much weight as the case, or more, so prepare stories on collaboration, managing a conflict and mentoring someone.
What happens in the Cornerstone final round interview (Superday)?
Four 45-minute interviews with a different case, plus a lunch or informal chat with Cornerstone staff; candidates often call this day the Superday. Candidates on Glassdoor describe final-round cases with exhibits such as stock charts or regression output, and follow-ups that change one assumption ("what if you shortened the estimation window?"). Cornerstone's tips for both rounds: break the question into smaller parts, take notes, ask questions, say your reasoning out loud, and take a moment before you answer. AI tools are not allowed during interviews.
A Cornerstone case rewards the math behind a damages figure as much as the structure. If percentages, growth rates and per-share arithmetic slow you down, train them first:

Learn the case math method
Setup, units, and shortcut math, linked to timed scored reps.
How is a Cornerstone case different from an MBB case?
A Cornerstone case asks you to prove or disprove an economic claim in a legal dispute; an MBB case asks for a business recommendation. The toolkit overlaps, but what counts as a good answer does not.
The most common mistake is forcing a profitability tree or market entry framework onto a damages question. When Cornerstone asks how you would assess damages from alleged price-fixing, the answer defines the but-for world, finds a comparison group and isolates the overcharge, not revenue and cost buckets.
What are the Cornerstone Research practice cases?
Cornerstone publishes two practice cases, the closest thing to Cornerstone sample cases: Zilo, a securities case, and Duff, an intellectual property case. These Cornerstone case examples are on its analyst case examples page (checked November 3, 2026), and the firm says they prepare you for similar scenarios in the interview.
How do you work through the Duff practice case?
Test customer substitution before calculating lost profits. In Cornerstone's published Duff solution (checked November 5, 2026), ACME sells directly to large institutions; Duff reaches individual consumers through retail chains. That difference undermines the claim that Duff's customers would otherwise have bought from ACME. The team also finds that a newer, cheaper technology explains ACME's decline, then examines costs before concluding that little profit was lost.
Rehearse the case in this order:
- Request customer evidence. Compare buyer types and distribution channels. Explain what overlap would support substitution.
- Test another cause. Use industry publications and analyst reports to investigate competing technology rather than attributing every lost sale to infringement.
- Translate sales into profit. Identify which costs change with the sales you attribute to the conduct. Explain how fixed costs affect the comparison.
Illustrative extension, using invented numbers: suppose the evidence supports $2M of lost sales and $1.2M of avoided variable costs. If fixed costs are unchanged, incremental lost profit is $2M - $1.2M = $0.8M. These are practice assumptions, not Duff's reported damages. Change either the supported sales loss or the avoided costs and recompute before defending your answer.
How do you solve a Cornerstone case step by step?
Follow the four questions Cornerstone's own solutions ask: what the claim assumes, what else could explain the loss, what the but-for world looks like, and what is left after costs.
- Find the assumption in the claim. Zilo's plaintiffs blame every dollar of the $40 fall on the misconduct, which assumes nothing else moved the stock.
- List the other explanations. Economy-wide news, industry news, other firm news; for ACME, a sector downturn or a new competitor.
- Build the but-for world from clean data. Fit the relationship on a period the misconduct does not touch (Zilo's pre-class-period prices) or compare customers (Duff against ACME).
- Size what is left, as profit. Lost revenue is not lost profit: subtract variable costs and account for fixed costs before giving a number.
Candidate reports on Glassdoor and Wall Street Oasis (WSO) add two more case families that follow the same steps.
Antitrust and market definition
Example: "A tech company controls 75% of search advertising. The DOJ alleges monopoly. Define the relevant market and assess whether 75% share proves market power."
Ask whether a small, lasting price rise (5 to 10%) would be profitable or would push buyers elsewhere, compare narrow and broad market definitions (search ads against all digital ads), and assess barriers to entry. Concentration matters too: the 2023 Merger Guidelines treat a market above 1,800 on the Herfindahl-Hirschman Index as highly concentrated. Market share alone never proves monopoly power.
Damages and valuation disputes
Example: "A retailer claims a supplier breach caused $15M in lost sales. Assess this figure."
Build the but-for scenario (what sales would have been without the breach) from the firm's pre-breach trend or comparable businesses, control for seasonality, the economy and new competitors, then subtract actual sales. Valuation disputes add data interpretation and DCF logic: two parties disagree on what an asset or business is worth.
No library case reproduces a court dispute, but a valuation case trains the same math: projecting a revenue stream, costing it and turning it into one defensible number. City Bike Share asks what a bidder should pay for 15 years of a city's bike share revenue; it is original practice, not Cornerstone material.
Worked example: valuing a city bike share concession
The prompt: In this McKinsey-style case, Austin wants to sell the operating rights to its 5,000-station bike share system for 15 years, and RideNext Mobility needs a reasonable bid. How do you estimate the concession's value?
Ask first: what discount rate or required return applies, what the contract allows on pricing and extra revenue, and who pays for equipment at the end of the term.
Structure: revenue (stations, rides, fare, 365 days), costs (one-time setup and recurring), 15 years of cash flow adjusted for time, then a bid set below economic value.
Recommendation: the $150M is undiscounted, so it overstates what the concession is worth. Bid below the $26.8M base value, for example $20M, which leaves about $6.8M of headroom, and say that the 10% rate is your assumption. The $17M a year of surge pricing, advertising and subscription revenue is gross and unproven, so keep it out of the bid. A litigation team does the same thing: it shows the number, then states what it assumes.
Practice for Cornerstone Research in a real case interview
TransportationHard
Worked example: pricing hospital software
The prompt: In this McKinsey-style case, MedSuite Technologies has built a cloud health records platform for the roughly 5,800 US hospitals and needs an annual price that maximizes first-year revenue.
Ask first: whether first-year revenue means contract value, cash collected or revenue earned in the launch year, how conversion changes with price and hospital size, and what sales capacity the company has.
Structure: demand at each price by hospital size, competitor positioning, then launch-year sales, comparing one flat price with a size-based menu.
At the case's stipulated 10% penetration, $150K gives 440 customers and $66M, against 300 customers and $60M at $200K. A size-based menu (800 large hospitals at $350K, 1,400 medium at $200K, 2,200 small at $150K) totals $890M of annual potential if demand holds. The price changes who is in the market, the same question you ask when you define a market by who leaves after a small price rise.
Recommendation: launch at $150K, the highest listed flat price on annual potential, and treat it as provisional. The figures are not a launch-year forecast, so check renewal timing, conversion by price and delivery capacity, and test simple tiers because of the $890M upside.
Practice a real healthcare market entry case
HealthcareMedium
Worked example: valuing a smart radiator acquisition
The prompt: In this Bain-style case, Radiator Co. wants to buy a European smart radiator valve startup to add wireless control. Which startup should it buy, and what is a fair deal value?
Ask first: which features are mandatory (app control, scheduling, air-quality control), what the stated $4M synergy measures and when it arrives, and what debt and cash sit between firm value and the price paid.
Structure: product and technical fit, financials, synergies, execution risk.
Recommendation: make B the lead diligence target because it balances app control, scheduling, scale and channels, and compare the smaller startup C as the alternative. Put standalone value at $9M to $9.3M. Do not call $13M a fair price: the $4M is lifetime revenue, not discounted net cash, so it overstates the synergy. Confirm the margin, timing and integration cost first. It is the same discipline as separating a revenue figure from a profit figure in a damages model.
Practice a real consumer electronics M&A case
Consumer ElectronicsMedium
What regression knowledge does Cornerstone expect?
Enough to explain a regression, not to run one: Cornerstone's Zilo solution uses a regression on market and industry returns, then an event study. You do not need an economics degree: a Cornerstone manager says on the interview process page that its cases are designed to be accessible to a wide variety of experiences and academic backgrounds. The pages list no statistics requirement, so treat this as reasoning you should be able to say out loud.
Regression reasoning for Cornerstone cases
- 01
Step 1: Define the outcome. What are you measuring? Stock returns, sales, market prices.
- 02
Step 2: Pick the explanatory variables. What else moves the outcome? Market returns, industry trends, seasons.
- 03
Step 3: Choose controls. Economy-wide shocks, industry trends, unrelated firm news. Leaving a relevant one out biases the result.
- 04
Step 4: Check significance. Is the effect statistically different from zero, and how wide is the confidence interval?
- 05
Step 5: Test robustness. Does the result hold with a different time window, controls or specification?
Three ideas come up again and again. Omitted variable bias: in Zilo, leaving out the industry index would blame the firm for an industry-wide fall. Statistical against economic significance: a detectable $0.02 effect on a $500 product changes nothing. Reverse causation: price may drive volume as much as volume drives price.
How do you solve a Cornerstone event study case?
Predict the stock's return from the market, subtract it from the actual return, and multiply only the remaining, firm-specific drop by the shares. Here is a worked example we wrote in the Zilo style.
Prompt: TechCo disclosed on March 15 that it overstated revenue by $200M. The stock fell from $85 to $62. TechCo has 150M shares outstanding. Plaintiffs claim $3.45B in damages ($23 × 150M shares). You are retained by the defense.
Build the market model: Regress TechCo daily returns on the S&P 500 and a software index over 250 pre-event trading days. Result: TechCo return = 0.2% + 1.3 × (S&P) + 0.8 × (software index).
Predict the return on March 15: The S&P fell 1.2% and the software index fell 2.5%. Predicted return = 0.2% - 1.56% - 2.0% = -3.36%.
Calculate the abnormal return: Actual return = ($62 - $85) / $85 = -27.06%. Abnormal return = -27.06% - (-3.36%) = -23.7%. Market and industry moves explain 3.36 points of the fall.
Compute damages: Damages per share = $85 × 23.7% = $20.15 (not $23). Total = $20.15 × 150M = $3.02B (not the plaintiffs' $3.45B). Controlling for the market cuts the claim by $430M.
Key insight: "Total decline × shares" ignores what the market did that day. Every event study case tests whether you separate firm-specific news from market noise. Drill the three calculations:
What mistakes hurt Cornerstone candidates?
Four mistakes come up most: MBB frameworks, no counterfactual, confusing statistical with economic significance, and choosing a method before asking about the data.
How should you prepare for a Cornerstone Research interview?
Cornerstone interview prep takes three to four weeks: economics and statistics basics, both Cornerstone practice cases, then mock interviews that mix behavioral questions and a case.
- Economics foundations: Review supply and demand, market structure, price discrimination and basic econometrics (what a coefficient means, controls, omitted variable bias). Read Cornerstone's two case examples until you can explain each step without notes.
- Case math: Work 2 or 3 antitrust and 2 or 3 event study scenarios, practice reading regression output, and review the case math guide. Use Quick Math when percentages or units are slow.
- Mock interviews: Run 4 to 6 mocks that start with two behavioral questions and end with a case, the shape of every Cornerstone interview. Prepare the collaboration, conflict and mentoring stories Cornerstone names, and use the behavioral interview guide to tighten them.
Which Cornerstone Research interview questions should you rehearse?
Prepare evidence of collaboration, conflict management and mentorship alongside your case work. A Cornerstone manager names those three dimensions on the firm's interview process page and says to prepare the behavioral part as much as the case, or more; the same page says the interview assesses communication and your ability to receive and apply real-time feedback, and that AI is not permitted during interviews (checked November 5, 2026). The prompts below are our practice questions based on those criteria, not a leaked interview bank.
For each answer, separate what you did from what the team did. In a case rehearsal, ask your partner to challenge an assumption, then practice revising your answer aloud. This makes coachability something you demonstrate rather than claim.
What should your Cornerstone Research cover letter and resume show?
Cornerstone asks for three documents: a cover letter with your office preferences, a resume that gives your expected degree completion date, and an official or unofficial transcript. Its how to apply page says applications with missing materials are not considered.
Make both documents show research you can defend: the question, the data, the check you ran and where the evidence stops.
- Before: Analyzed a large dataset.
- After: Tested the project question across two data cuts, documented the limitation, and summarized the finding for the research lead.
The cover letter explains why that evidence fits litigation work and names the offices you want. The resume step in the process above grades your resume before you send it.
Analysis Group vs Cornerstone Research: how do the interviews compare?
Cornerstone gives a case in every round; Analysis Group says formal case studies are not part of its process. Both are economic consulting firms working mostly for law firms, so the content of the questions overlaps.
Both checked November 3, 2026. The Analysis Group interview guide covers its behavioral questions in detail. For other economic consulting firms, see the NERA, Compass Lexecon and Charles River Associates guides, and where economic consulting fits among the types of consulting firms.
What does a Cornerstone Research analyst do?
Analysts are generalists who perform the quantitative analyses behind a case at every stage, across practice areas. Cornerstone's analyst page describes the role that way and recruits analysts from bachelor's and master's programs graduating in 2027 (checked November 3, 2026).
Cornerstone's cases support academic and industry experts who testify, so the analysis has to survive an opposing expert. Pay depends on office, degree and start year; use current offer data and Glassdoor's Cornerstone salary pages as a directional benchmark.
Three ways to practice for Cornerstone Research
- Start a free case
Run a full case. 92 cases in voice or guided text, scored on seven dimensions.
- Try a free drill
Drill one skill. 550+ drills across 6 skill types.
- Find your test
Practice your firm's test. Solve, Casey, CCA, SOVA and TestGorilla, each free to try.
Sources
- Cornerstone Research: The Interview Process. cornerstone.com/careers/analyst/interview-process (checked November 3, 2026)
- Cornerstone Research: Analyst and Summer Analyst Interview Process. cornerstone.com/careers/analyst/analyst-summer-analyst-interview-process (checked November 3, 2026)
- Cornerstone Research: Analyst Case Examples (Zilo and Duff). cornerstone.com/careers/analyst/analyst-case-examples (checked November 3, 2026)
- Cornerstone Research: How to Apply, the Analyst Role. cornerstone.com/careers/analyst/how-to-apply (checked November 3, 2026)
- Cornerstone Research: Analyst careers. cornerstone.com/careers/analyst (checked November 3, 2026)
- Analysis Group: Interview Tips and Tricks (PDF, 2026). analysisgroup.com (checked November 3, 2026)
- U.S. Department of Justice and Federal Trade Commission: 2023 Merger Guidelines. justice.gov/atr/2023-merger-guidelines (checked November 3, 2026)
- Glassdoor: Cornerstone Research Interview Questions. glassdoor.com (checked June 17, 2026)
- Wall Street Oasis: Cornerstone Research Interview Questions. wallstreetoasis.com/company/cornerstone-research/interview (checked June 17, 2026)
- Glassdoor: Cornerstone Research Salaries. glassdoor.com (checked June 17, 2026)
- Vault: Cornerstone Research profile, 2026 rankings (#2 for economic consulting, #6 on the Vault Consulting 50 North America). vault.com (checked November 5, 2026)
Frequently asked questions
Let your favorite AI make the case
Ask ChatGPT, Claude, or Perplexity whether Road to Offer is worth it. We're confident you'll like the answer.
No affiliation with ChatGPT, Claude, or Perplexity
Resources and related guides
Practice for Cornerstone Research in a real case interviewPractice
- Browse all free resourcesResource hub
- McKinsey vs BCG: 10 Real Differences (2026 Comparison)Firm Specific · May 1, 2026
- Accenture Assessment Test 2026: Format, Cappfinity & PrepFirm Specific · Jun 27, 2026
- Accenture Case Interview: Workbook, Case Study (2026)Firm Specific · Mar 15, 2026




