What McKinsey says
Not much, and that's the point. McKinsey's Solve page says your performance is considered in conjunction with the rest of your application and any other assessments you're asked to take. It doesn't publish a score, a scale, a pass mark or a cutoff, and it doesn't send candidates a score. You learn the outcome when you're invited to interview or you're not.
So every "you need 70%" or "top 20% pass" you read is somebody's estimate. Some estimates are careful, built from many candidate reports. None of them is a McKinsey number, and a careful site will say so.
What's consistently reported about how it's scored
Two components come up in every serious account:
- Product. Were your answers right. In Red Rock: the calculations, the report entries, the case answers. In Sea Wolf: whether your treatments met each site's targets. In the Sustainable Futures Lab there's no right answer to a single question, so the product is closer to the pattern of your choices.
- Process. How you got there. The assessment is reported to record what you collect, how you move through the screens, whether you use the tools you're given (the calculator, the journal), how long you spend where, and how often you change your mind. One prep site with coaching experience says a clean process with a few wrong answers can outscore a messy process with the same answers. We can't verify the weighting, and neither can they, but the advice that follows from it costs nothing: work in a way you could explain out loud.
What isn't known: how the two combine, how the games are weighted against each other, whether the threshold moves with office or recruiting cycle, and how the Sustainable Futures Lab is scored at all. Prep sites say the assessment compares your pattern to that of successful consultants. That's plausible and unconfirmed.
Pass rates you'll see quoted
The figure that circulates most is that roughly 20 to 30% of candidates who take Solve go on to interviews, which is another way of saying 70 to 80% stop here. It comes from prep sites aggregating candidate reports. It's not a McKinsey statistic, it can't be checked, and it almost certainly varies by office and by how many people applied that cycle. Treat it as "most people don't get through" rather than as a number to aim at. There's no version of the test where you can see the bar, so the only strategy is to do each game as well as it can be done.
What a practice score means
A score on SolveForge, or on any practice site, measures one thing: how well you've learned that site's recreation of the game. It's built from public descriptions of the mechanics, scored by rules the site wrote, on scenarios the site invented. It cannot predict a McKinsey result, and any site that says its score does is guessing.
Used honestly, it's still worth having. Our benchmarks, for our scenarios, are:
| Game | Solid | Strong | Elite |
|---|---|---|---|
| Red Rock Study | 60 | 80 | 90 |
| Sea Wolf | 60 | 80 | 93 |
| Sustainable Futures Lab | 50 | 70 | 85 |
A strong score on a scenario you haven't seen before means the mechanics and the traps aren't costing you points any more, and that's the thing practice can actually deliver. It says nothing about the process side, which no practice tool can see.
What to do with a bad practice score
- Find the one thing. Our results screens show the answer key next to your answers (Red Rock) and the best trio next to yours (Sea Wolf), and the emailed breakdown names the skills that cost you. Usually it's one trap, repeatedly: percentage points, adults versus total, an undesired trait you didn't screen for.
- Fix that one thing. Read the relevant section of the game guide, drill it on paper for ten minutes.
- Run a fresh scenario. A replay of the same scenario will go up whether or not you've improved, because you've seen the answers. A new scenario tells you the truth. That's the honest reason fresh scenarios are the thing we charge for.
- Don't chase the number past the point of learning. If you're scoring in the strong band on new scenarios, more runs mostly buy anxiety. Rest before the day.
FAQ
What score do you need to pass McKinsey Solve?
McKinsey doesn't publish a pass mark, a score range or a cutoff, and it says the result is considered together with the rest of your application. Any number you see on a prep site is an estimate built from candidate reports, not a McKinsey figure.
Do you get your McKinsey Solve score?
Not automatically. You find out the outcome when you're invited to interview or not. One prep site reports that candidates can ask McKinsey for their result and be told a percentile range; we haven't been able to confirm that, so ask your recruiter if it matters to you.
What is the McKinsey Solve process score?
Prep sites and candidates consistently report that the assessment records how you work, not only your answers: what you collect, how you move through the screens, whether you use the tools, how you pace yourself. McKinsey hasn't published how this is weighed. The practical advice is to work in a way you could explain, use the calculator and the journal, and commit to answers.
What does a SolveForge practice score mean?
It measures how well you've learned each game's mechanics on our recreation, not how McKinsey would score you. Our benchmarks are 80 and above on Red Rock and Sea Wolf, and 70 and above on the Sustainable Futures Lab. A practice score is a training signal, not a prediction.
What should I do with a bad practice score?
Find which trap or which step cost you, fix that one thing, and run a fresh scenario. A replay of the same scenario will go up whether or not you've improved, because you've seen the answers. A fresh scenario tells you the truth.
Keep going
How each game is scored in detail: the Red Rock Study guide, the Sea Wolf guide and the Sustainable Futures Lab guide. The whole assessment in one place: the McKinsey Solve guide. Where your practice score sits against other players: the leaderboard.
SolveForge is an independent practice tool and is not affiliated with, endorsed by, or connected to McKinsey & Company. "McKinsey" and "Solve" are trademarks of their respective owners. Practising here does not guarantee any assessment result.