Victor Cheng Review 2026: LOMS, Frameworks, and Dates

Victor Cheng's Catalog, and the Order Form Price

The product is recordings. The program page states it plainly: "The Look Over My Shoulder program consists of 20 hours of audio recordings and transcripts of these recordings. The program includes both candidate-led case formats and interviewer-led/command-and-control formats (McKinsey)."

The page itself carries no sticker. Every button on it opens an order form, and that form prints "$297.00", read 7 October 2026. The page does print the alternative: "With the payment plan, you can make 4 monthly payments of $87." Four payments of $87 each is $348, which is $51 more than paying once, and the page does not say so.

The refund promise is on the page and it is a strong one: "We also offer a 100% satisfaction guarantee, so if you are not happy with the program within 30 days, we will issue you a full refund."

Two other things belong in any account of what is on sale. The products page lists a resume toolkit and a cover letter toolkit alongside the flagship, and offers free access to "6 hours of training videos" plus "PDFs of my Case Interview Secrets video program" in exchange for an email address. And the coaching page is a two line notice: "Case Interview Coaching is no longer available. We have discontinued our Case Interview Coaching Services."

The Four Frameworks, and the Claim Attached to Them

The method is a deliberately short list, and the frameworks page argues for shortness: "Other case interview prep resources often will give you dozens of frameworks to use in your preparation. My approach is different in that I have found it easier to master just 3-4 key frameworks that apply to the vast majority of cases. If you get really good at the frameworks and are naturally analytical, you'll be able to do extremely well in 85% of case situations." Read 7 October 2026.

The four are named on that page: profitability, business situation, mergers and acquisitions, and capacity change. There is a second, separately named method for experience questions, the PARADE Method, which the site describes as an acronym a candidate can use "to prepare concise and powerful answers to commonly asked consulting resume questions". That one covers the resume conversation rather than the case.

Here is the detail that answers the question people actually type. The site names four frameworks. The catalogued contents of the author's 2012 book name three, omitting capacity change. So a reader who buys the book and a reader who reads the site get different lists, and no page on either side dates the difference.

The 85 percent figure carries no denominator, no sample and no method, and it is the author's estimate of his own material. We report the sentence and treat the number as a claim.

What Victor Cheng Does Well in Twenty Hours

Audio is an unusual choice and the page defends it well: "The audio delivery of the program forces you to focus more on the content, allowing you to improve your case skills instead of focusing on the body language of candidates, which is not as relevant to your case skills." Read 7 October 2026.

Take that argument seriously, because it is a real one. A case interview is mostly spoken. What an interviewer hears is a sequence of sentences, and most preparation material is read rather than heard. Twenty hours of listening to real attempts, including weak ones, trains an ear for pace and for the moment a candidate loses the thread. Very little else on this shelf does that, and nothing in print can.

The second strength is the format coverage. The program covers both the candidate led case and the interviewer led case, which are different exercises with different failure modes, and a candidate who has only rehearsed one is exposed. Naming both and recording both is the right call.

The third is a short, unfashionable list. Four frameworks is a position, and it is defensible against the alternative of memorizing twenty. A candidate who knows four shapes deeply and can build a tree will go further than one carrying a folder of grids.

The fourth is the refund window. Thirty days and a full refund, stated on the product page rather than in a terms document, with no clause about how much of the material you may consume first. That is a shorter route to a refund than most of this market offers.

What a Recording Cannot Hear, and What a Date Cannot Change

A recording cannot hear you. It can model a good answer and it can model a weak one, and it cannot tell you which of the two you just gave. That is a property of recorded audio and it will not change.

The second limit is the calendar, and the site supplies the evidence itself. The program page's own update notice reads: "We have updated LOMS for the 2021 Recruiting Season." That is the most recent update the page announces for itself. The coaching page says the coaching is gone. Both read 7 October 2026.

What sits in that gap is a whole screening stage. McKinsey's current interviewing page describes a gamified assessment called Solve running ahead of the interviews for most consulting roles, read 7 October 2026 in the archive capture of 27 September 2026. Twenty hours of case audio has nothing to say about it, and should not be expected to.

The last row is what makes this page readable as a comparison. We put numbers on structure, arithmetic, sizing, brainstorming, exhibits and synthesis. We do not put one on whether you held the room, and the room is part of what this audio trains an ear for. Several of our simulators are still being built.

One claim on the site deserves naming rather than passing over. The author page states "9 out of 10 new hires at McKinsey, Bain and Boston Consulting Group are caseinterview.com students." No source, period or definition of a student appears with it, and none of the three firms publishes such a figure. The same page gives "over 105,000 active members" while its own signup box invites readers to "Join over 144,119 aspiring consultants". We report both numbers and adopt neither. For context, Bain's case interview page does publish a shortlist of external resources it recommends to its own candidates, and this one is not on it.

What Our Grader Hears in a Spoken Close

Our method pairs the same listening with a verdict on what you said. The weights are ours, which is why we publish them. No firm publishes a marking scheme.

A spoken close is scored Answer first 35, Evidence 25, Risk 20, Next step 20. Those numbers come out of src/lib/grading/rubrics.ts and the whole set sits on our method page.

Answer first carries 35, the heaviest of the four, and it is the exact thing twenty hours of audio is good at teaching and cannot possibly measure. Listening to a strong candidate lead with their conclusion teaches you what it sounds like. Whether you did it this morning is a fact about your recording, and the program has no way to hear your recording.

Risk at 20 is the line this material raises more than most, since a recorded weak attempt usually fails by asserting a recommendation without naming what would break it. Next step at 20 is the same shape. Evidence at 25 is the one a listener can judge for themselves, because the case data was stated out loud.

Twenty Hours Against a Month of Marked Attempts

A one time program and a subscription compare on use rather than on sticker, so here is the arithmetic. Our prices, read in src/lib/pricing.ts, are $49 a month, $120 for 90 days and $249 for a year. Ninety days divided across three months is $40 a month. A year at $249 divided by twelve is $20.75 a month, which this page rounds to about $21 and says that it rounded.

1. Look Over My Shoulder costs $297.00 once, for 20 hours of audio, which is $14.85 an hour of listening.
2. The payment plan is four payments of $87, so $348 in total, which is $17.40 an hour and $51 more than the single payment.
3. Our annual plan at $249 is less than the program's single payment by $48, and it marks every attempt you submit.
4. On the monthly plan at $49, thirty marked attempts in a month cost $1.63 each.

Read what that table shows and no more. It prices listening against marking, which are different goods. Twenty hours of recorded attempts at $14.85 an hour is reasonable for what it is, and the free six hour offer is the cheapest honest way to find out whether the voice works for you. The last column is the dividing line, and the top four rows all read the same way.

Who Victor Cheng Still Fits, and Who Should Pass

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

Take the free six hours first, always. It costs an email address, it includes video and PDFs, and it tells you within an hour whether this teacher's pace suits you. Nobody should pay $297 before doing that.

Buy it if you learn by ear and your commute is long. This is the only serious audio program on the shelf, it covers both case formats, and twenty hours of hearing real attempts is a genuinely different input from reading. The thirty day full refund makes the trial cheap.

Pay once rather than over four months. The plan costs $348 against $297, which is $51 for the convenience. If you can pay once, do.

Skip it if your process starts with an online test, or if you need a mark. The program's newest announced update is for the 2021 season and the coaching is discontinued, so nothing here speaks to the gamified screen that now sits in front of the interviews at McKinsey. And no recording can tell you whether your own answer was any good, which is the question most candidates reach after twenty hours of listening.

Where to Send the Answer the Audio Cannot Hear

The gap this page measured is one of direction. Twenty hours of audio sends words at you, and the thing that decides your interview is the words you send back.

Our synthesis drill scores a close on Answer first 35, Evidence 25, Risk 20 and Next step 20, and names which of the four you dropped. Our structure drill scores a typed tree on Coverage 45, Separation 25, Depth 20 and Concision 10. 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.

The honest pairing is both, in order. Listen for the pace, then be marked on your own attempts. What we do not do, stated plainly: we publish no audio library, we do not judge how you come across to a partner, and several of our scoring simulators are still being built.

For context first, our guide to case interview frameworks sets out where the firms' own published instructions disagree, and our page on case interview videos covers what recorded material can and cannot teach.

What the Site Publishes, and What It Does Not

Every figure attributed to this brand was checked 7 October 2026 on its own pages, with a browser user agent over a compressed request. The program description, the payment plan, the refund promise and the 2021 update notice came from the program page. The $297.00 came from the live order form that page links to. The frameworks and the 85 percent claim came from the frameworks page. The discontinuation came from the coaching page. The membership and new hire claims came from the author page.

Two sources are not the site. McKinsey's interviewing page supplies the current interview sequence and the screening assessment, read through an archive capture because that domain answers neither a direct request nor a fetch. Bain's case interview page supplies the fact that a firm publishes its own shortlist of outside resources. The Penn Libraries record supplies the catalogued contents of the 2012 book, which is how we can say the book names three frameworks where the site names four.

Four things we declined to assert. We do not repeat the claim about nine in ten new hires at three firms as a fact, because no source, period or definition accompanies it and the firms publish nothing like it. We do not state a membership figure, because the same page carries two. We do not confirm or dispute the author's stated background, his offers or his ranking, all of which appear only on his own pages. And we found no independent first hand user account later than a decade ago that we were willing to cite: the forums that carried them are closed to our user agent, the review platforms return nothing, and the remaining results are published by this brand's competitors.