PrepMatter Review 2026: What the Solve Simulator Claims

What PrepMatter Sells, and Which Price Is Published

The flagship is software. The Solve simulator page offers to let a candidate "Practice all 3 McKinsey Solve games with an interactive simulation", names them as "Redrock Study, Sea Wolf, and the newest addition, Sustainable Futures Lab", and states the term as "12-month access, one-time charge". Its own FAQ confirms the term ends: "you'll have access to the course for 12 months. After this period, your access will not renew automatically, and no further charges will apply."

What that page does not contain is a price. We grepped the raw markup for a dollar sign, for USD, for EUR and for the word price. Zero hits. A buyer reaching that page learns what the product does and has to start a checkout to learn what it costs. An Internet Archive capture of the same page from 14 March 2026 did carry a figure, $239 for the twelve months, and we report that only as what the page said in March.

The coaching is priced in the open. The coaching page lists $309 for one session, $1,495 for a block of five and $2,890 for a block of ten, and states that "Each session lasts 75 minutes". Both numbers in one place is what makes a real comparison possible further down.

The catalog is wider than the one product. The terms name six assessments the platform sells preparation for, covering McKinsey Solve, two BCG assessments, two Bain assessments and a games based test, alongside case and fit interview material. The same document puts the operating entity in a Dubai free zone under the law of the United Arab Emirates.

Repetition, Then a Report on Each Attempt

The mechanism is stated without decoration, which is refreshing. "Our simulation is dynamic, so you can practice as many times as you want and get a different scenario each time." And then: "After each attempt, you'll receive a detailed report showing how your answers map across the key traits McKinsey evaluates, along with question-by-question feedback to help you improve your score next time." Read 7 October 2026.

So the method is volume plus a scored report. Not a theory of problem solving, and the page does not pretend to one. For a timed assessment that is a defensible choice, because what degrades performance on a timed task is unfamiliarity with the interface and the clock, and both of those yield to repetition.

The coaching has a separate and more conventional method, described by one of the coaches in the first person: "I typically support them over 6 to 8 weeks. We meet every 3 days or so, and in each meeting, we cover one topic. Since there are 10 topics in total, we typically finish all topics in a month."

One sentence in the simulator FAQ carries more weight than the rest, and it needs quoting whole: "The Prepmatter simulation is based on real candidate experiences and reflects the latest 2026 format." Hold that next to McKinsey's assessment integrity expectations, which ask candidates to refrain from "sharing information about the content of their assessment/interview with others (such as specific questions, problem solving scenarios)", read 7 October 2026. We are not asserting how any content was obtained, and we have no way to know. We are putting three published sentences in one place, the third being the vendor's own statement that its scenarios are unique to it, and leaving the reader to hold all three.

Why No PrepMatter Review Can Rate the Accuracy

One structural limit on this product is permanent, and the firm sets it. McKinsey does not publish the assessment. Its content, scoring and weighting are held by the firm, and the firm asks candidates who sit it to keep it that way. So no simulator can be verified against the original by anyone, including the people who build it and including us. Accuracy is unknowable rather than doubtful, and that is a property of the market rather than a flaw in a product.

The second limit is also the firm's doing rather than the software's. McKinsey's own interviewing page describes Solve as "a gamified assessment that tests your natural problem-solving abilities" and then says "There's no need to prepare", read 7 October 2026 in the archive capture of 27 September 2026. Bain says the same about one of its two digital assessments in three words, "No preparation is needed", and for the other it points candidates to the test vendor's own free sample questions. Whether a candidate believes the firms is their business. Reading both sentences before paying is the point.

That last row applies here more sharply than on any other page in this series, so read it twice. We publish six assessment simulator pages of our own and our own product page states that none of the six is live. We score typed and spoken case answers. We do not currently score a gamified assessment, and we will not imply that we do.

What PrepMatter Does Well, Starting With Its Refund Terms

Most of this market promises a refund in a sentence and defines it nowhere. This vendor defines it in numbers, and that deserves saying first. The headline reads "Pass McKinsey Solve or get 50% back", the mechanism is "Just forward the email from McKinsey confirming you didn't place in the top 25%", and the thresholds are printed: eligibility at "the 7th decile or below", with an alternative of a numeric score of "510 or below". Read 7 October 2026.

Agree or disagree with the level, that is a falsifiable promise. A candidate can read it before paying and know exactly what would qualify them. Compare that with a guarantee whose conditions appear only in a terms document after purchase.

The second thing it does well is disclaim. The simulator page carries its own notice: "It is not an official McKinsey product, nor is it endorsed or sponsored by McKinsey & Company. While our simulation offers a similar gameplay experience and covers comparable skills, the scenarios, content, and interface are unique to Prepmatter." That is the correct thing to write, it is written plainly, and it is on the product page rather than buried.

The third is the form itself. A gamified timed assessment is the one part of a consulting process where software genuinely is the right medium. You cannot rehearse a timed interactive task on paper or with a friend. If a candidate wants repetitions of a clicking, dragging, timed exercise, something that runs in a browser is the only thing that can supply them, and this vendor builds exactly that.

What Our Grader Reads in a Sizing Chain

Our method is practice plus a verdict on what you produced, with the weights printed because they are ours. No firm prints a marking scheme, which is exactly why nobody can grade a simulator against one.

A market sizing answer is scored Scope 25, Assumption chain 30, Arithmetic 25, Sanity check 20. Those numbers come from src/lib/grading/rubrics.ts and the full set sits on our method page.

Assumption chain at 30 is the heaviest line, and it is the one a timed clicking exercise never reaches. It asks whether each number you invented follows from the one before it and whether you said so out loud. A report that tells you your score moved from the sixth to the seventh decile does not tell you that your third assumption was unsupported, because a decile is a rank rather than a diagnosis.

Sanity check at 20 is the other line worth naming. It asks whether you looked at your own answer and said whether it was plausible. Candidates skip it under time pressure, and a timed task trains them to skip it faster. Scope at 25 asks whether you defined what you were counting before you counted. Arithmetic at 25 is the only one of the four that a timed drill improves directly, and it is a quarter of the total.

Seventy Five Minutes, and What an Hour Really Costs

This vendor publishes a session length, which most do not, so the hourly conversion is available instead of guessed. Our own published prices, read in src/lib/pricing.ts: $49 a month, $120 for 90 days, $249 for a year. Divide 120 by three for $40 a month and 249 by twelve for $20.75, which we round to about $21 and say that we rounded.

Now the coaching, converted honestly.

1. One session costs $309 for 75 minutes. Divide 309 by 75 and a minute costs $4.12, so an hour costs $247.20.
2. Five sessions at $299 each is $1,495, which is $239.20 an hour.
3. Ten sessions at $289 each is $2,890, which is $231.20 an hour.
4. Ten sessions is 750 minutes, so 12.5 hours of expert attention for $2,890.

Read what those figures show. Buying ten sessions instead of one saves $16 an hour, which is about 6 percent, so there is no strong volume argument for committing early. And the gap between the two product lines is the real content of the table: $2,890 buys 12.5 hours of a person, and $249 buys twelve months of an automated verdict. Those are different goods, and a candidate who needs the first should buy the first.

The simulator row is empty for a reason we want stated plainly. The page prints no price, and a number from a March archive capture is not today's price. Anyone comparing on cost has to begin a checkout to find out, which is a fact about the product rather than a criticism of it.

The PrepMatter Review Verdict, as Four Decisions

The verdict is a decision rule. Four readers, four answers.

Read the firm's sentence first, then decide about the simulator. McKinsey says there is no need to prepare for Solve, and Bain says no preparation is needed for one of its two tests. If you accept that, this product is not for you. If you want the repetitions anyway because an unfamiliar timed interface rattles you, that reason is real and the software is the only thing that supplies them.

Try the free demo before any purchase, and ask for the price. The vendor offers a free Sea Wolf demo on the simulator page. Use it. Since the page carries no figure, the price is something you have to extract before you can compare, and you are entitled to it without starting a payment.

Buy one coaching session rather than a block. At $247.20 an hour against $231.20 in a block of ten, the discount for committing to twelve and a half hours is about 6 percent. Buy one, find out whether that coach suits you, then buy more if they do.

It stops being enough the moment your problem is the spoken case. A simulator trains a clicking task and a report ranks you on it. Neither touches the ninety seconds where you state a recommendation to a partner. If your rejections come after a live interview rather than after an online test, the money belongs somewhere else.

Where to Put the Practice a Report Cannot Mark

The gap this page measured is specific. A simulator can hand you an unlimited number of timed attempts and a rank. It cannot tell you which link in your reasoning gave way, because a rank is not a diagnosis.

Our market sizing drill scores an estimate on Scope 25, Assumption chain 30, Arithmetic 25 and Sanity check 20, and names the assumption that failed rather than giving you a percentile. Our math drill scores Answer 40, Working 25, Units 15 and Interpretation 20. Our pricing is $49 a month, $120 for 90 days, or $249 for a year, which is about $21 a month on the annual plan.

What we do not do, stated plainly: we do not simulate a gamified assessment, our own six assessment pages say that none of those simulators is live, and we do not sell a coach. For the Solve exercise specifically, the most useful free thing in existence is the firm's own material, and BCG likewise offers "a free job simulation that provides insights into the challenges we solve".

For context first, our guide to McKinsey's online assessments covers what the firm itself publishes about them, and our page on case interview tools covers what each category of software actually does.

Which Claims Here Are the Vendor's and Which Are Not

Every figure attributed to this vendor was checked 7 October 2026 on its own pages, with a browser user agent over a compressed request. The simulator description, the access term, the refund thresholds and the absence of a price came from the Solve product page. The session prices and the 75 minute length came from the coaching page. The operating entity and the list of six assessments came from the terms.

The firm statements are separate and come from the firms. McKinsey's description of Solve and its four free sample cases were read through an archive capture of 27 September 2026, and its integrity expectations through a capture of 20 May 2026, because that domain answers neither a direct request nor a fetch. Bain's digital assessment page and BCG's case interview page were read live the same day.

Four things we refused to assert, and we would rather name them. We do not state a current price for the simulator, because the page prints none and the March figure is historical. We do not state how the simulation's content was built beyond quoting the vendor. We take no position on whether the simulation resembles the real assessment, because the firm forbids publishing the content that would settle it, so nobody on earth can show that comparison honestly. And we found no independent first hand user account we were willing to cite: the review platforms refuse our requests, the forum that would carry them is closed to our user agent, and the remaining results are published by this vendor's competitors.