Most candidates treat a Deloitte live case like a shorter BCG case and stop when they land a recommendation. That is where they leave points on the table. Deloitte's own recruiting pages publish a five-step approach that ends on next steps and expected results, not on the decision alone, and the firm hosts a free practice library that mirrors its client mix rather than a generic consumer P&L.
After this piece you should be able to map Deloitte's published rubric onto what the interviewer is listening for, pick the right practice cases from the firm's tool, and close a case the way the sheet actually scores it.
What Deloitte says in public
Start with the primary sources, not campus folklore. Deloitte's case and scenario interview tips state that cases test problem-solving, analytical ability, and strategic and logical thinking, and that you are not expected to produce the "right" answer. The stated goal is to convey logic under ambiguity. The same page lists five named steps: understand the issue, identify underlying assumptions, summarise findings, state recommendations, and outline next steps and expected outcomes or impacts.
The firm also ships practice tools. The tips page above points entry-level consultative candidates to an interactive case-interview practice tool (short business situations, about 15 to 20 minutes, with questions you can answer and check) and Audit & Assurance candidates to a separate scenario tool. The US recruiting hub at usrecruiting.deloitte.com/case-prep-tool is the durable entry point to that consulting prep material.
None of that is a confidential scoring sheet leaked online. It is Deloitte telling you, in plain language, the shape of a good answer. Treat it as the public half of the rubric.
The fifth step is the differentiator
In a typical first-round case slot of 30 to 45 minutes, I hear clean trees and competent math from most people who reach Deloitte. Where scores separate is the close. Ending on "I recommend defending the segment" without a number, a risk, and a dated next test is finishing after step four on a five-step sheet.
| Step (Deloitte's wording) | What the interviewer is grading | Weak close | Strong close |
|---|---|---|---|
| Understand the issue | Clarifying questions without stalling | Restates the prompt | Names the decision and success metric |
| Identify assumptions | Explicit constraints (budget freeze, statute, capacity) | Silent assumptions | States two to three assumptions out loud |
| Summarise findings | So-whats, not data dumps | Lists every exhibit | Names the one driver that moves the answer |
| State recommendations | A decision the client could act on | Vague "improve operations" | One clear go / no-go or priority |
| Next steps and expected impact | A plan with a result and a check | "Monitor progress" | Dated action, expected delta, risk |
Ned's rule. If your last sentence does not contain a date, a number, and a risk, you have not finished a Deloitte case. The recommendation is step four; the sheet still has one box open.
That rule is an interviewer heuristic, not a Deloitte policy quote. Use it as a rehearsal filter: record yourself closing three cases and score the last fifteen seconds against it.
Why the practice library matters more than another profit tree
Deloitte is a large professional-services network, not a pure strategy boutique. For the fiscal year ending 31 May 2025 it reported aggregate global revenue of US$70.5 billion and a workforce of more than 470,000 people. Consulting sits inside Strategy, Risk & Transactions and Technology & Transformation; those lines grew 5.5% and 4.7% in local currency in FY2025 per the same release. The practice cases on Deloitte's tool are fictional, but they are designed to feel like that book of business: federal agencies, technology architecture, cyber, forensic and valuation work, human capital, and a handful of commercial growth prompts.
You do not need every case in the library. You need the two or three that match the practice named on your invitation. A Technology & Transformation interview that starts with a profitability tree for a consumer brand wastes the first minute translating the wrong object. A Human Capital prompt that never mentions workforce utilisation or change adoption will sound like a strategy MBA who never sat through an implementation review.
| Practice flavour (from Deloitte's tool categories) | What to train | Wrong default |
|---|---|---|
| Strategy / commercial growth | Candidate-led structure, quantified recommendation, next steps | Stopping at "enter the segment" |
| Technology / architecture | Process, systems, and risk trade-offs with a phased plan | Forcing a pure revenue tree |
| Cyber / forensic / valuation | Issue framing under incomplete data; clear assumptions | Pretending you know the technical answer |
| Human capital / talent | Adoption, capacity, and stakeholder sequencing | Treating people as a residual cost line |
The tool itself says cases differ by interview length and that Deloitte is grading how you think through a problem, not subject-matter trivia. That matches what I score: structure, judgment under ambiguity, and a professional close.
Worked example: Midland Claims Portal backlog
Prompt. Midland Mutual, a regional P&C insurer, launched a self-service claims portal six months ago. Digital intake is 38% of new claims; the target was 60%. Average adjuster handle time on portal claims is 47 minutes versus 31 minutes on phone intake. Leadership wants a 90-day plan. Headcount is frozen. What do you recommend, and what should they expect by day 90?
Understand. Success is portal share toward 60% and handle time closer to phone, without hiring. Ask whether contractors, overtime, and UI changes are in scope. Ask how many claims arrive per week.
Assumptions (say them). Midland Mutual writes about 4,000 new claims a week. Portal UX can be changed in six weeks. Adjusters cannot be hired; overtime is capped at 5%. First-pass completeness on portal forms is the main unknown.
Structure. Three peer branches: (1) intake quality and rework on the portal form, (2) adjuster workflow after intake, (3) customer routing (who is steered to portal versus phone). Test intake quality first because rework inflates handle time and kills adoption.
Math you can reproduce. 4,000 claims a week. Portal share today: 0.38 × 4,000 = 1,520 portal claims. Phone: 2,480. Extra handle time on portal: 47 − 31 = 16 minutes. Extra minutes a week: 1,520 × 16 = 24,320 minutes ≈ 405 adjuster-hours. At 35 productive hours per adjuster-week, that is roughly 11.6 FTE of friction. If first-pass incomplete rate on portal forms is 35% and each incomplete claim adds one extra touch of 12 minutes, incomplete touches alone are 0.35 × 1,520 × 12 ≈ 6,384 minutes ≈ 106 hours a week, or about 3 FTE. Cleaning the form will not close the full 16-minute gap, but it is the first test with the freeze.
Suppose a redesigned form lifts first-pass completeness from 65% to 85% (a 20-point gain). Incomplete rate falls from 35% to 15%. Hours saved: 0.20 × 1,520 × 12 / 60 ≈ 61 hours a week ≈ 1.7 FTE equivalent. Portal share might rise if customers stop abandoning after failed submissions; a cautious estimate is +8 points of share in 90 days if the form ships in week six and marketing nudges phone users for four weeks.
Recommendation. Do not hire. Ship a mandatory evidence checklist on the portal form in six weeks, pull two senior adjusters onto a forms war-room for four weeks, and run a soft redirect for low-severity auto claims to the portal.
Next steps and expected impact (the fifth step). Sample 300 incomplete portal claims this week; report the top five missing fields to the claims ops lead by Friday. Expected result by day 90: first-pass completeness near 85%, handle time from 47 minutes toward the high 30s if workflow tweaks land, portal share from 38% toward the mid-40s. Risk: agents push back if the form feels longer; mitigate with a two-field pilot before full rollout.
That close is Deloitte-shaped: a decision, a constraint respected, a number, a date, and a risk.
How this differs from McKinsey and BCG without the mythology
McKinsey live cases are typically interviewer-led: exhibits and questions arrive in sequence. BCG and Bain live cases are usually candidate-led: you drive the tree and ask for data. Deloitte's consulting case is closer to the candidate-led shape. You still follow the interviewer if they take control of a question.
The useful difference is not "harder" or "easier." It is the published fifth step and the client mix. An MBB-trained close that ends on a crisp recommendation without an implementation test can still sound unfinished in a Deloitte room. Group exercises show up in some Deloitte finals; treat them as confirmed only when your invitation names them. When they do, assessors watch whether you integrate someone else's point, disagree without dominating, and land one shared recommendation.
Behavioral is separate content on the same loop. Deloitte's behavioral interview tips ask for STAR stories and evaluate achievement, communication, leadership and teaming, and aptitude. Prepare stories that end in adoption or a measurable client result, not analysis alone.
Drill the close this week
- Read Deloitte's five steps once, then sit two interactive practice cases from the US case prep hub in the practice area on your invite.
- On every close, force one sentence with a date, a number, and a risk. If you cannot say it aloud in under twenty seconds, rewrite it.
- Run one timed candidate-led case on CoachNed at /start (three typed turns, instant scores, no account) or the live voice case at /interview. Everything is open for seven days, no card; then $120 for a recruiting season or $49 a month.
- If your invite names an online assessment or job simulation, prepare that format on its own terms. It is not a substitute for case reps.
Frequently asked questions
Is a Deloitte case interviewer-led like McKinsey?
No. Live consulting cases are generally candidate-led: you structure, ask for data, and synthesise. Follow the interviewer if they redirect a question. Confirm the exact format in your invitation, because offices and service lines vary.
Does every Deloitte candidate sit a group case?
No. Group exercises appear in some final rounds and offices, not as a universal global rule. Practise them when your recruiter or schedule names a group exercise; otherwise prioritise the live case and behavioral.
Where should I practice if I have only a few days?
Use Deloitte's own 15-to-20-minute practice cases via the US case prep hub first, then one full-length candidate-led case with a scored next-steps close. The firm's tool is free and unmatched for flavour; CoachNed is for reps and feedback under time pressure.
Is Monitor Deloitte the same interview?
No. Monitor is Deloitte's strategy practice with its own filter and, in many routes, a written case. Confirm the practice name on the invitation before you choose prep material.
Sources
- Preparing for the Case and Scenario Interview | Deloitte US — five-step approach, logic over "right" answer, links to case and scenario practice tools. Checked 2026-09-24.
- US recruiting case prep hub — Deloitte US entry point to interactive consulting case prep. Checked 2026-09-24.
- Deloitte reports FY2025 revenue — US$70.5B aggregate revenue, 470,000+ people, SR&T and T&T growth. Checked 2026-09-24.
- Behavioral Interview Tips | Deloitte US — STAR method and what recruiters evaluate. Checked 2026-09-24.
- Tips to Prepare for an Interview | Deloitte US — general interview prep and mock-interview guidance. Checked 2026-09-24.
CoachNed is independent and not affiliated with Deloitte.