What to take away
- The rubric weights analytical thinking at 50%, presence and communication at 25%, and track record and drive at 25%
- Fourteen named skills sit inside those three dimensions: five in analytical thinking, four in presence and communication, five in track record and drive
- Every skill is scored 1 to 5 against a fixed written standard, not against the other candidates in the room that day
- 3 means you met the first-round standard on that one skill; 3.5 on the weighted average is where we set the line for a credible hire
- A short, case-only session cannot score track record and drive, so analytical thinking and presence and communication reweight to fill the gap
- The top of the scale is reachable by anyone who performs at that standard, not reserved for a rare flash of brilliance
What is the case interview scoring rubric, exactly
Ask most case-prep sites how a case interview gets scored and you'll get some version of "you're judged on structure, math, and communication." That's true, and it tells you almost nothing. It doesn't say what a 4 on structure looks like next to a 2, or how much your math is worth relative to your closing recommendation, or whether being nervous costs you anything at all. A rubric you can't check your own performance against isn't a rubric. It's a description with a label on it.
What follows is the actual scorecard: the one our AI coaches mark against on every case, unedited. Three weighted dimensions, split into fourteen named skills, each with a written description of what a 1, a 3, and a 5 look like. It's modeled on the structured scorecards the two of us used as interviewers before we built this, translated into our own vocabulary and our own weights rather than reproducing any one firm's internal document. If a prep product tells you it grades against "the real rubric" without showing you the rubric, that's worth noticing.
| Dimension | Weight | Sub-dimensions |
|---|---|---|
| Analytical thinking | 50% | Framing, commercial judgement, rigor, creativity, synthesis |
| Presence & communication | 25% | Presence, precision, listening, relationship-building |
| Track record & drive | 25% | Self-direction, collaboration, depth, growth mindset, results delivered |
What are the five analytical thinking skills, and what does 1, 3 and 5 look like on each
Analytical thinking is half the score, so it's the highest-return place to look first. It covers whether you can build a structure that fits the specific problem in front of you, reason through it with real numbers, and land on a recommendation a partner could act on. Five skills sit inside it.
| Skill | 1: not yet shown | 3: on the bar | 5: excellent |
|---|---|---|---|
| Framing | Reaches for a memorised framework with no fit to the problem, or buckets with no logic | A sound, broadly MECE structure that fits the problem and is usable, even if a layer is thin | A structure tailored to this problem, genuinely MECE, a stated or clearly-implied hypothesis, and adapts when new information lands |
| Commercial judgement | Assumptions are arbitrary or business-illogical; explores a low-value branch first | Plausible assumptions, a defensible first branch, reasonable trade-offs | Assumptions reflect real business logic and are named explicitly; picks the highest-value branch first; makes a sharp trade-off unprompted |
| Rigor | Avoids the math, or is wrong by an order of magnitude with no recovery | Gets the math right, at least in order of magnitude, but slowly, or only sense-checks when prompted | Correct on the first pass, states the calculation aloud, and sense-checks the result against a known anchor without being asked |
| Creativity | Only the obvious; no original angle | One non-obvious idea, or a useful angle the interviewer didn't fully seed | A genuinely non-obvious insight, a cross-domain connection, or a reframe that sharpens the answer |
| Synthesis | No recommendation, "I'd need more analysis," or recommends before diagnosing | A clear recommendation in the right direction with reasons, even if missing a risk or a quantified ask | Recommendation first, two or three reasons with numbers, a material risk plus mitigation, and a concrete next step |
Two of these skills get misjudged constantly by candidates marking themselves, so they're worth a beat each.
Framing is scored on the structure you finished, not the one you were still building
We score the structure a candidate had committed to by the time they finished laying it out. A first layer that's still thin because you asked to verify direction before going deeper doesn't cost you here. What costs you is a set of buckets with no underlying logic, or a memorised framework bolted onto a problem it doesn't fit.
Rigor doesn't penalise asking for a number to be repeated
This is worth stating plainly because it's the single most common thing candidates get wrong about their own performance: asking the interviewer to repeat or confirm a figure you didn't catch is never a deduction on rigor. We score the reasoning, the arithmetic, and whether you sense-checked the result, not whether you needed a number said twice.
What are the four presence and communication skills, and what does 1, 3 and 5 look like on each
This dimension asks a narrower question than it sounds like: not "were you likeable," but whether your thinking actually reached the interviewer clearly, under pressure, without them having to do extra work to follow you. Four skills, a quarter of the overall.
| Skill | 1: not yet shown | 3: on the bar | 5: excellent |
|---|---|---|---|
| Presence | Visibly rattled; speeds up under pressure; hedges almost every claim | Generally composed; voice may tighten on the hard parts | Calm and paced throughout, comfortable with silence, states a view without hedging it away |
| Precision | Rambling; several ideas tangled per sentence; vague placeholders | Gets the point across, signposts sometimes, mostly clear | One idea per sentence, signposts before drilling in, uses real numbers and named entities |
| Listening | Talks over the interviewer's pushback; misses corrections; answers a different question | Follows the interviewer's nudges; occasionally needs something repeated | Plays the question back, incorporates small corrections, asks to clarify a number rather than guessing |
| Relationship-building | Transactional or stiff; no curiosity beyond passing the test | Warm enough and professional; would be fine to work with | Feels like a real teammate: warm, genuinely curious about the problem, asks for help without losing momentum |
- Presence is about pace and composure, not confidence in the abstract
- Precision is about density: one idea per sentence beats three ideas tangled into one
- Listening is about catching what the interviewer actually said, including deliberate small corrections
- Relationship-building is about whether a partner would want you in the room with a client, not whether you're charming
What are the five track record and drive skills, and what does 1, 3 and 5 look like on each
This dimension is scored entirely from your behavioural stories, not from the case itself. It's asking a different question from the other two: not whether you can do the analytical work in this room, but whether you'll still be doing it well, unsupervised, in year three. A quarter of the overall, five skills.
| Skill | 1: not yet shown | 3: on the bar | 5: excellent |
|---|---|---|---|
| Self-direction | No example of acting on their own initiative; waited to be told what to do | One clear example of starting or driving something themselves, with usable ownership | Repeatedly moves under their own steam, pursues outcomes past the comfortable point, holds a considered view under pushback |
| Collaboration | No sense of how they work with others; credit and roles are vague | Names their role on a team and how they worked with others | Separates their own contribution from the team's, credits specific people for what they did, handled disagreement constructively |
| Depth | Claims with no substance; folds at the first probe | Backs claims with some specifics and holds up to one layer of probing | Numbers, names and sizes of impact; holds up to two layers of probing without retreating |
| Growth mindset | No genuine failure named, or defensive when challenged | Names a real setback and something they changed afterwards | A real failure, not a humble-brag, a concrete change made because of it, and openness to disagreement in the conversation itself |
| Results delivered | Ambitions or activity, but no shipped outcome | At least one real outcome traceable to their action | Quantified results across multiple examples, clearly attributable to them rather than to circumstance |
Depth is the skill candidates most reliably underestimate in themselves. A story that sounds complete on first telling often has nothing behind it once you push a second and third layer down: who specifically disagreed, what you said back, what happened after the part where the story usually stops. Build stories that survive that, not stories built to be told once.
What does the 1-to-5 scale mean, independent of any one skill
Every cell in the three tables above is a specific application of one general scale. It's worth seeing that scale on its own, because the mistake we see most often is candidates treating 3 as "average" and 5 as "perfect." Neither is right. This is a criterion-referenced scale: each number is a fixed description of a standard, not a rank against the other candidates interviewing that day.
| Score | What it means |
|---|---|
| 5 | Did this as well as a strong hire does it: fully realised, or a genuinely insightful non-standard move that achieves the same end. Reachable by anyone who performs at this level |
| 4 | Clearly above the first-round bar, with at most a minor rough edge |
| 3 | On the bar: met the first-round standard for this skill. Solid and usable; would not hold the candidate back |
| 2 | Developing: a real attempt with partial signal, but a gap a coach would flag and the candidate needs to close |
| 1 | Not yet shown: no usable evidence of this skill in what was attempted. For genuine absence, not a weak attempt |
Two things follow from a scale built this way. A 5 doesn't require brilliance, only doing the thing to the standard a strong hire does it, which means candidates leave 5s on the table constantly by assuming the top is reserved for something extraordinary. And a 1 specifically means the skill never showed up at all, so a candidate who attempts the math and gets the wrong answer while reasoning aloud usually scores a 2 or a 3, ahead of one who says "I'm not great at numbers, could you walk me through it" and never attempts rigor in the first place.
Why is 3.5 the first-round bar, not 3.0
The fourteen skill scores average into three dimension scores, and those three combine on the 50/25/25 weighting into one overall number from 1.0 to 5.0. That number maps to a hire read with published cut-points.
| Overall score | Reads as |
|---|---|
| 4.5 and above | Outstanding |
| 3.75 to 4.49 | Strong, clearly above the bar |
| 3.0 to 3.74 | Solid, keep building |
| 2.0 to 2.99 | Developing |
| Below 2.0 | Early days |
Notice that 3.0 to 3.74 is a wide band, and it contains both a candidate who's genuinely on the edge and one who's clearly ready. That's why we don't treat 3.0 as the pass line even though it's technically "on the bar" for every skill. A real first round wants you comfortably above the standard across all fourteen skills at once, because the interviewer is extrapolating a thirty-minute conversation out to three years of client work. An overall that lands exactly at 3.0 usually means a couple of 4s propping up a 2, and that 2 is precisely what gets raised in the debrief.
So we set 3.5 as the readiness line, roughly the midpoint of the "solid" band, rather than 3.0. It's a deliberately conservative reading: below it, keep drilling before you treat yourself as ready; at or above it, you're in credible territory for a real first round.
How do the fourteen skill scores turn into one overall number
The arithmetic is simple once you see it laid out, and seeing it laid out changes how you read your own results. Two steps.
Step one: average the skills inside each dimension
Each dimension's score is the plain average of its own skills. Five numbers in analytical thinking average into one; four in presence and communication average into one; five in track record and drive average into one.
Step two: weight the three dimension scores
Those three numbers combine at 50% analytical thinking, 25% presence and communication, 25% track record and drive, which is a weighted average, not a simple one. This is the step that has the biggest practical consequence: a weak dimension score in analytical thinking costs you twice what the same weak score costs you in either of the other two.
That example is a genuinely strong-looking case on the surface, several 4s, nothing below a 2, and it still lands under the readiness line. That's the arithmetic worth internalising: two skills stuck at 2 pull an otherwise solid case down by roughly a fifth of a point each on their dimension, and analytical thinking's weight doubles that effect. Closing your weakest skill moves the overall further than polishing a skill that's already a 4.
Does every practice session score all fourteen skills
No, and this matters for reading your own results correctly. Our default session is 30 minutes, case-only, with no behavioural conversation in it. Because track record and drive is scored entirely from behavioural stories, a case-only session structurally cannot assess it. There's nothing in that transcript for a coach or an interviewer to score on those five skills.
When track record and drive is absent, the other two dimensions don't just get ignored, they reweight to fill the full 100%. Analytical thinking moves from 50% to roughly 67% of the overall, and presence and communication moves from 25% to roughly 33%. The 60-minute session adds a behavioural part and scores the full rubric at the original 50/25/25 split.
| Dimension | 60-minute session | 30-minute, case-only session |
|---|---|---|
| Analytical thinking | 50% | ~67% |
| Presence & communication | 25% | ~33% |
| Track record & drive | 25% | not scored |
How should I actually use this rubric to prepare
Having the full rubric changes what "getting better at case interviews" means. It stops being one vague goal and becomes fourteen specific ones, most of which you can score yourself against without a partner.
Score your last case skill by skill, not as one impression
Write all fourteen skills down the page, or the five in analytical thinking if that's what you have transcripts for, and give each a number with one line of evidence. "That went okay" is not a score; a number with a quote behind it is.
Find your lowest number in analytical thinking first
It's worth twice as much as either of the other dimensions, so a 2 there costs you more than a 2 anywhere else. If framing or rigor is your floor, that's where the next several practice sessions should point.
Check presence and communication for a skill that's costing more than it feels like
Listening and precision are the two candidates most often underrate in themselves, because they don't feel like "case skills." Record a session and count how often you answer more than what was asked.
Build three track record stories before you assume you're ready
You can't self-mark this dimension from a case transcript, because there isn't one. Write three stories, one strong on self-direction, one on collaboration, one on results delivered, and rehearse each against the follow-up questions a real interviewer would ask.
Re-score after your next case and check whether the same skill is still the floor
Improvement on this rubric looks like your lowest number moving up, not your highest number moving further up. If the same skill is still your floor three sessions later, the drilling isn't targeting the right thing.
“My recommendation is to enter the market through the regional distributor, not a direct launch.”
“Two reasons: distributor entry gets us to 40% regional coverage in year one versus roughly 15% direct, and it caps our upfront investment at around 2 million euros against the 8 million a direct launch would need.”
“The main risk is margin: the distributor takes an estimated 18 to 22 percent cut, which we'd want to negotiate down as volume scales, and we should protect that in the contract from day one.”
“Next step: get real distributor economics from the client's regional team before we commit, since our 18 to 22 percent is an assumption, not a confirmed number.”
Common questions
What are all fourteen skills on the case interview scoring rubric?
Analytical thinking (50%) covers framing, commercial judgement, rigor, creativity and synthesis. Presence and communication (25%) covers presence, precision, listening and relationship-building. Track record and drive (25%) covers self-direction, collaboration, depth, growth mindset and results delivered. Every skill is scored 1 to 5 against a written standard for that specific skill.
Is a 3 a good score, or just an average score?
A 3 means you met the first-round standard for that skill, which is a real, usable outcome, not a mediocre one. This rubric is criterion-referenced rather than curved, so there's no forced distribution and no guarantee that some candidates land below 3. A candidate scoring all 3s across fourteen skills is, by definition, meeting the bar everywhere.
What happens to the rubric in a 30-minute case-only practice session?
Track record and drive isn't scored at all, because it depends on a behavioural conversation that a case-only session doesn't include. Analytical thinking and presence and communication reweight to fill the full 100%, moving from 50/25 to roughly 67/33. A full 60-minute session includes the behavioural part and scores all three dimensions at the original weights.
Can a candidate realistically get a 5 on every skill?
In theory yes, since the scale is criterion-referenced and every level is reachable by anyone performing at that standard. In practice almost no real transcript scores all 5s, because fourteen different skills rarely peak in the same thirty minutes. A strong, hireable case typically shows a mix of 3s and 4s with one or two genuine 5s, not a flat row of top marks.
Where does this rubric come from?
It's modeled on the structured scorecards the two of us used while interviewing candidates at McKinsey and Bain, translated into our own vocabulary, our own weights, and our own written anchors rather than reproducing any firm's internal document. MBB Ready is not affiliated with McKinsey, BCG or Bain.
Does asking the interviewer to repeat a number cost points on this rubric?
No, on either of the two skills it could plausibly touch. On rigor, we score the reasoning, the arithmetic and the sense-check, not whether a number needed repeating. On listening, asking to clarify a figure rather than guessing is explicitly scored as a strength, not a weakness.
Which skill should I fix first?
Whichever skill is your lowest score inside analytical thinking, because that dimension carries 50% of the overall, twice the weight of either of the other two. A 2 on framing or rigor costs roughly twice what a 2 on presence or listening costs, so closing that gap moves your weighted overall furthest for the same amount of practice.