McKinsey PST cutoff score, and how Solve is scored today

The PST cutoff: about 19-20 out of 26

McKinsey never published a pass mark for the Problem Solving Test. My estimate was about 19-20 correct answers out of 26, which is roughly 75%. I based that on my own result, on what candidates I coached reported after their tests, and on what recruiters hinted at.

Candidates also reported that the bar looked similar across offices and across entry levels (undergraduate, MBA, experienced hire). I never saw evidence of a lower bar for any group. Treat 75% as a working target, with the caveat that it is an estimate.

Why 75% was harder than it sounds

Scoring 75% on a test of 26 questions means you can afford about six or seven wrong answers. With roughly 2 minutes 18 seconds per question, most candidates did not finish comfortably. A typical failing pattern looked like this:

• Two questions wrong because of reading errors (answering a different question than the one asked).
• Two wrong because of a math slip (a misplaced zero or the wrong base for a percentage).
• Three or four unanswered or rushed guesses on the last case.

That adds up to the margin. Preparation was mostly about removing the first two categories and buying time to avoid the third.

The guessing math

The PST had no penalty for wrong answers. That one rule shaped the right strategy.

On a test where 19-20 was the target, three blind guesses were worth roughly 0.75 of a point. Two well-eliminated guesses were worth a full point. That difference often decided the outcome.

My rule for candidates was simple:

1. With five minutes left, stop working on any question that needs a long calculation.
2. For each remaining question, eliminate any option that is clearly wrong in direction or order of magnitude.
3. Guess among what is left and fill in every answer before time runs out.

Stephan's note: BCG's legacy Potential Test worked the other way, with a deduction for wrong answers. Candidates who sat both tests had to switch strategy between them. The details are in the BCG Potential Test article.

The PST is retired: how Solve is scored

McKinsey replaced the PST with Solve, a gamified online assessment. McKinsey's official page says your Solve performance is considered alongside the rest of your application. It does not publish a pass mark, a score scale or the weight of each game.

Prep providers describe two dimensions:

• Product score: the quality of what you produce, such as correct calculations and correct final selections.
• Process score: how you work, inferred from the sequence and timing of your actions, how you gather information and how often you change answers.

According to these providers, scores are compared with a benchmark group, and the effective threshold shifts with how competitive the office is and how many people it is hiring that cycle. That makes a fixed "Solve cutoff" less meaningful than the PST's rough 75%.

Percentile feedback on Solve

Some prep sites report that candidates can request their Solve result and receive a percentile range. Others note that a rejection email alone does not reveal a score. Both can be true: feedback is usually something you ask for.

If you take Solve, ask your recruiter whether feedback is available, whatever the outcome. A percentile range tells you whether you missed narrowly or by a wide margin, which matters if you plan to reapply. See what happens when you fail for waiting periods and next steps.

What the scoring change means for strategy

The PST rewarded a simple rule: answer everything. Solve, as prep providers describe it, rewards a slightly different discipline.

1. Finish every task. An unfinished Redrock report or an incomplete site selection in Sea Wolf scores poorly on outcome, whatever the process.
2. Work in a deliberate order. Read the objective, gather the relevant data, calculate, then decide. Random clicking and frequent answer changes are reported to weigh on the process score.
3. Watch each game's timer separately. Candidate reports say each game has its own clock, so time saved in one game cannot be spent in the next.

The underlying lesson from the PST still applies: points are lost to haste and misreading far more often than to difficulty.

What to aim for in practice

Since Solve has no public target, set your own practice targets on the underlying skills:

• Written case sets: 80% or better on timed 26-question practice tests, finishing with at least two minutes spare.
• Mental math: percentage and growth calculations accurate within 2% in under 30 seconds.
• Exhibit reading: state the key insight of a chart in one sentence within 20 seconds.

Those targets sit comfortably above the old PST bar and build the habits Solve rewards. The Key to the PST is a full 26-question timed practice test with a complete answer guide, built for the PST format. The PST is retired, and the book remains a direct way to measure yourself against the 75% benchmark. Find it on the Key to the PST page, and follow the 4-week practice plan to structure your preparation. For the full Solve format, read the McKinsey test in 2026.