Most Capgemini Invent candidates walk into the room with a clean profitability tree and no plan for what happens after the answer. That is the wrong bet. Invent's own careers page says the case is guided, lasts about 45 minutes, and scores you on structured problem solving, quantitative and qualitative analysis, communication under pressure, and thinking holistically, including identifying risks and mitigation tactics. The sheet is grading a miniature transformation engagement, not a pure strategy puzzle.
The thesis is simple: treat every Invent case as a strategy-plus-delivery problem. Land a recommendation, name the two risks that would kill it, and say what you would do in week one. After this piece you should be able to run that loop on a digital or operating-model prompt, do the math without freezing, and behave in a group exercise like someone who has sat on a real project team.
What Invent actually is (and why the brand matters)
Capgemini Invent is Capgemini's management consultancy brand for innovation, design, and transformation. On its brand page, Invent describes itself as combining strategy, technology, data science, creative design, and engineering. The parent group is large: Capgemini reported €22.1 billion in 2024 revenue and about 341,100 employees at year-end. Invent sits inside that engine as the strategy-and-change layer that often hands work to technology delivery.
That split changes what interviewers listen for. A clean MECE tree that never mentions data, customer experience, operating model, or change risk sounds like an MBB candidate who wandered into the wrong building. An Invent interviewer wants to hear how the answer would actually get built.
How the process usually unfolds
Invent is explicit that process depends on country and role. Capgemini Invent's Australia careers FAQ lists application, phone or video interviews, and possible assessment centres with group exercises. The US Invent students page focuses on a guided case of roughly 45 minutes. The UK Accelerate graduate path publishes a tighter sequence: online application, digital interview, then a half-day assessment centre with a group exercise and a strengths-based interview (UK Accelerate recruitment process).
| Stage (typical) | What you face | What they are filtering for |
|---|---|---|
| Screen / digital interview | Motivation, resume, "Why Invent" | Clarity, commercial curiosity, brand fit |
| One-on-one case | Guided business problem, ~45 minutes (US Invent description) | Structure, math, judgment, risks |
| Assessment centre / group exercise | Shared problem with peers (UK and some other markets) | Collaboration without dominance |
| Fit / strengths interview | Behavioral stories, motivation | Delivery mindset, teamwork |
Candidate folklore online often freezes group-case timings at "4–6 people, 10 minutes reading, 30 minutes discussion." Capgemini's public pages confirm group exercises exist in some markets; they do not publish a universal timing sheet. Prep for collaboration. Do not memorize a countdown clock you cannot verify.
What the score sheet rewards in a one-on-one case
Invent publishes the skills it assesses. Map your practice to that list, not to a generic case checklist.
| Published skill (Invent US careers) | What I score in the room | Common leak |
|---|---|---|
| Structured problem solving | A simple issue tree in under two minutes | Jumping to a favorite framework |
| Quantitative and qualitative analysis | One clean calculation plus a judgment call | Math without a so-what |
| Communication under pressure | Short synthesis every five to seven minutes | Narrating your notes |
| Holistic thinking / risks and mitigation | Two real risks with owners and mitigations | Ending on "I'd recommend X" with no downside |
Their sample prompt on the students and graduates page is an automotive client weighing electric and self-driving vehicles: market size, positioning, customer value, and who to target. That is strategy with technology and market design baked in. If your practice deck is only "profit down 10%, find the driver," you are under-preparing.
Ned's rule. On an Invent case, I stop listening when the recommendation has no risk line. Name the two ways the plan fails, who owns each risk, and what you would check in the first two weeks. That is what "holistic" means on their sheet.
A worked Invent-style case: Meridian Appliances
Original company, original numbers. Reproduce the math with a pen.
Prompt. Meridian Appliances sells mid-market kitchen appliances in North America. Revenue last year was $1.2 billion. Connected "smart" products are 8% of units sold but 18% of revenue because of higher ASP and a subscription add-on. The CEO wants a three-year plan: should Meridian push a full digital service stack (app, predictive maintenance, parts marketplace), and what is the revenue upside if they do?
Structure I want to hear in sixty seconds.
- Market and offer: who buys connected appliances, what jobs the service stack must do.
- Economics: unit lift, subscription attach, incremental cost.
- Operating model: data platform, service network, partner vs build.
- Risks and sequencing: cybersecurity, dealer channel conflict, capability gaps.
Math (do this out loud).
Assume 4 million units sold last year. Connected share is 8% of units → 320,000 connected units. Those units drive 18% of $1.2B → $216 million revenue from connected products, or about $675 ASP equivalent vs a blended ASP of $300 on all units ($1.2B / 4M).
Suppose Meridian raises connected penetration to 25% of units in year three (1.0 million units) and attaches a $60/year service subscription on 40% of those units.
- Incremental connected units vs today: 1,000,000 − 320,000 = 680,000.
- If each incremental connected unit carries $200 of hardware ASP premium vs a dumb unit: 680,000 × $200 = $136 million hardware upside.
- Subscription: 1,000,000 × 40% × $60 = $24 million recurring.
- Rough year-three upside before cost: ~$160 million, or about 13% of current revenue.
Then force the qualitative judgment Invent wants. The $160 million only shows up if Meridian can (a) instrument the field service network, (b) keep dealer margins from collapsing when parts go digital, and (c) fund a data platform that is not a science project. My recommendation would be a staged push: instrument the top 20 SKUs, launch a thin subscription for filter and service reminders, prove attach above 25% in two regions, then decide on marketplace build vs partner. Risks: dealer revolt (mitigate with shared margin on parts), data privacy incidents (mitigate with a security gate before any app launch), and CapEx without attach (kill criteria at month nine if attach is under 15%).
That answer is what the sheet is built for: structure, numbers, recommendation, risks, next steps.
How to behave in the group exercise
Where Invent uses an assessment centre, the group exercise is not a debate tournament. Interviewers are watching for the same habits that keep a transformation program from falling apart: framing, timeboxing, building on others' points, and a clean close.
In the first two minutes, offer a shared map ("demand, cost, delivery risk") rather than an opinion. Midway, pull a quiet voice into the data. In the last five minutes, volunteer to synthesize so the room lands a recommendation. Dominating airtime is an easy way to fail while feeling productive.
I score group rounds on a simple three-point grid: Did you make the group smarter? Did you protect the clock? Did you leave with a decision someone could brief a client on? Talking volume is not on the grid.
Fit questions Invent interviewers actually use
Expect "Why Capgemini Invent rather than a pure strategy firm or a pure tech firm?" Your answer needs both halves: you want strategy work that ships through technology and design, and you understand Invent's place inside Capgemini's dual digital-and-sustainable positioning. Capgemini's own 2024 results release frames the group around that dual transition; Invent is where much of the strategy-and-change work starts.
Also prepare one story about leading without authority, one about recovering a messy analysis under time pressure, and one about explaining a technical trade-off to a non-technical stakeholder. Keep each story under two minutes with a clear outcome.
Practice this week: one Invent loop
Do this once, timed, with a pen.
- Open a transformation-flavored case on CoachNed cases or take the five-minute first rep at /start.
- Force a four-bucket structure in under two minutes: market/offer, economics, operating model, risks.
- Complete one calculation and state the so-what in one sentence.
- Close with recommendation, two risks with mitigations, and a week-one next step.
- If you want voice pressure, run the live case at /interview (~18 minutes, seven-score debrief).
Everything on CoachNed is open for seven days, no card; then $120 for a recruiting season or $49 a month. CoachNed is not affiliated with Capgemini or Capgemini Invent.
Frequently asked questions
How long is the Capgemini Invent case interview?
Invent's US students and graduates page describes a guided case of approximately 45 minutes. Other markets and experienced-hire tracks may run shorter or longer screens, video interviews, or assessment centres, so confirm with your recruiter for your country and role.
Does Capgemini Invent always use a group case?
No. Capgemini Invent states that recruitment can include group assessments depending on country and role. The UK Accelerate graduate path publishes a half-day assessment centre with a group exercise; other offices emphasize one-on-one cases. Treat the group round as likely in graduate assessment-centre tracks, not as a universal final for every Invent hire.
Are Capgemini Invent cases more digital than MBB cases?
Often in flavor, yes. Invent's public sample involves EV and self-driving strategy, and the brand sits at the intersection of strategy, design, data, and technology. You still need classic profitability and market-sizing muscle; the difference is that an answer without delivery, data, or change risk usually under-scores.
What should I practice the week before an Invent interview?
One candidate-led case per day with a forced risk-and-mitigation close, twenty minutes of mental math, and one timed synthesis drill. If your office uses an assessment centre, add one mock group case with friends where your only job is to frame, timebox, and close.
Is Capgemini Invent the same as Capgemini Consulting?
Capgemini Invent is the group's current innovation, design, and transformation consultancy brand. Older "Capgemini Consulting" branding was folded into Invent; interview for the Invent role you applied to, not a discontinued label.
Sources
- Capgemini Invent brand page — positioning as strategy, technology, design, and transformation consultancy. Checked 2026-09-24.
- Capgemini Invent: careers for students and graduates (US) — case interview length (~45 minutes), assessed skills including risks and mitigation, sample automotive EV prompt. Checked 2026-09-24.
- Capgemini Invent Australia careers — process varies by country and role; group assessments possible. Checked 2026-09-24.
- UK Accelerate recruitment process — application, digital interview, assessment centre with group exercise and strengths-based interview. Checked 2026-09-24.
- Capgemini full-year 2024 results — Group revenue €22,096 million and headcount context for 2024. Checked 2026-09-24.