How To Study for the GMAT - A Data-Backed Guide to Test Prep Resources
For many years, the best data available on GMAT preparation stemmed from anecdotal reports posted online in forums like Reddit and GMAT Club. That’s because nobody had ever conducted a systematic analysis evaluating GMAT preparation options among a sample large enough to be statistically robust.
That situation changed dramatically in January 2019, when Poets and Quants released an exclusive study of GMAT preparation techniques. The survey’s sample consisted of 859 readers from the publication’s worldwide audience. These readers tend to be ambitious MBA and specialized business master’s degree candidates who want to attend elite business schools in the United States. Such schools encompass those typically ranked among the U.S. News and World Report top 25, including the super-elite M7 schools like the Harvard Business School, the Stanford Graduate School of Business, and the University of Pennsylvania’s Wharton School.
Traditionally, candidates like these aim to score above a 700 on the GMAT’s 200 to 800 scale, which in 2019 ranked at or above the 88th percentile. But, according to the test’s administrators, the Graduate Management Admission Council (GMAC), only about 32,500 of the roughly 260,000 GMAT scores around the world each year exceed 700.
In the study, all the respondents had taken the GMAT twice. Before their first attempts, the examinees evidently had studied independently without hiring a preparation service. But before the second time they sat for the test, those candidates received training conducted by 11 of the most popular test preparation vendors.
By comparing the scores before and after training, the researchers ranked the effectiveness of the vendors’ online platforms, classes, tutoring, or some combination of those services. The authors also ranked the vendors’ effectiveness on a variety of dimensions besides score increases, like customer satisfaction.
Several aspects within the data struck us as contradictory and confusing. Accordingly, we’ve simplified and streamlined the results. And what’s more, we distilled the study’s most essential conclusions into actionable insights that MBA applicants planning to sit for the GMAT should find useful.
What’s Best: A GMAT Class, Tutor, or Online Platform?
Before the Poets study, many experts believed that the best odds of receiving a strong score on the GMAT required hiring a private tutor. But the surprising results of the study suggest that premise might amount to an oversimplification.
That’s because one unequivocal conclusion emerged: GMAT registrants should enroll in classes, tutoring, and online platforms. In other words, they should enroll in all three preparation modalities.
Why? Buried deep within the report, the data shows that the maximum score increase—a staggering average improvement of 118.8 points—was achieved by Manhattan Prep clients who bought all three of that vendor’s offerings. Those services include classes and tutoring, as well as Manhattan Prep’s online study platform. Overall, the clients who bought the full range of Manhattan’s services blew away those who bought from competitors because on average the non-Manhattan clients only improved their scores by 83.3 points. That 35.5-point difference works out to a whopping 43 percent score boost.
Now, keep in mind some important points about Manhattan’s 119-point surge. First, that’s merely the average gain logged by these clients. Although Poets doesn’t provide us with the variance around that mean value, it’s likely that several examinees trained by Manhattan Prep gained even more than 119 points. Indeed, some may have gained a lot more.
Second, that gain seems even more remarkable because, according to GMAC, Manhattan’s 119-point average equals an increase of roughly a full standard deviation. In other words, that mean value roughly equates to 68 percent of the variance in all GMAT examinees’ scores.
Third, not many respondents engaged in Manhattan Prep’s comprehensive “kitchen sink” approach. Only 16 examinees did, which works out to merely 14 percent of the 113 MP clients within the sample, and less than two percent of all the respondents.
But fourth, of all the sample’s respondents, those who did utilize all three of Manhattan’s modalities didn’t spend the most time prepping. The narrative reports those 16 MP clients spent 36 hours on average, divided between 26 classroom hours and ten hours with a tutor (it’s not clear how much time they spent with MP’s online platform).
Interestingly, about 5 percent of the sample, or 41 examinees, spent 40 hours or more prepping, but they only realized a 93.7-point average score boost. That increase only accounts for about 79 percent of the increase the 16 Manhattan preppers attained—even though the latter group invested less instructional time.
Fifth, this 119-point increase doesn’t tell us much about how difficult that gain would have been to accomplish. The reason is that we don’t know its starting level. For example, a gain of 119 points from a first score of 500 would be a 619, and that score would be well within the capabilities of many examinees. But if an examinee started with an initial score of 681, a 119-point gain would result in a “perfect” 800 score.
And except for a genius GMAT superstar like Dan Edmonds—an expert tutor whom GMAC banned for life from taking the test after three consecutive 800 results—a score at or above 760 on the GMAT is so difficult to attain that it’s virtually impossible.
Top GMAT Training Modalities
But what about the examinees who aren’t enthusiastic about utilizing (and paying for) all three of these training modalities from Manhattan Prep?
If financially feasible, it still makes good sense to combine two of the three modalities: classes and tutorials. Across all vendors, the examinees who bought tutors along with classes boosted their scores by 23.2 points over those who exclusively prepped with an online platform. And in percentage terms, that’s more than a 30 percent score improvement.
|Average GMAT Gain||Sample Size|
|Class & Tutor||100||46|
Top GMAT Prep Vendors
About 22 percent of the survey respondents bought GMAT preparation services from only two vendors that were responsible for the greatest average score increases.
The firm with the largest increase, New York-based Manhattan Prep, sold services to 113 examinees, or 13 percent of the sample. Incidentally, Kaplan Inc. acquired Manhattan in 2009 while 2020 Presidential candidate Andrew Yang served as MP’s CEO.
Bought by 8 percent of the sample, Berkeley, California-based Magoosh ranked as the vendor with the second-greatest score gain. Magoosh was founded in 2009 by four students at the Haas School of Business at the University of California at Berkeley. During the early days of online video, the firm’s first product was a collection of video solutions to quantitative GMAT questions.
Here’s how the survey’s top three vendors stack up:
|Average GMAT Gain||Sample Size|
Maximizing GMAT Performance Requires Instructional Time
In general, according to the Poets data, a close correlation between the instructional time invested and higher scores appears to exist.
The data shows that only 13 examinees couldn’t spare more than 20 instructional hours in classes or tutoring. Their gains were the lowest of the sample, at only 76.2 points, which ranks significantly below the 83-point average.
Contrast that lackluster performance with the 41 examinees we mentioned above who invested at least 40 instructional hours with classes or tutors. Their average score boost was 93.7 points, which is a 23 percent increase over the group that invested the least instructional time.
The only exception appears to lie with the timesavers who bought all the services available from Manhattan Prep. But in every other case, those who invested more hours with classes and tutors earned higher scores.
|Hours Size||Average GMAT Gain||Sample|
|11 to 20||76.2||13|
|21 to 30||88||40|
|31 to 40||90.7||31|
|40 and above||93.7||41|
Client Satisfaction Doesn’t Always Match High GMAT Scores
Satisfaction with GMAT prep vendors was the subject of a separate follow-up Poets report that based its conclusions on a smaller sample of 596 respondents. In that case, it turned out that large score gains didn’t always match customer satisfaction ratings.
One might surmise that the two vendors responsible for the largest overall gains—Manhattan Prep and Magoosh—would also win the top client satisfaction rankings. But that didn’t happen. Oddly enough, the 20 clients who bought services from Target Test Prep awarded this upstart the highest satisfaction rankings of any firm in the survey—a commanding 8.8 on a 10-point scale.
Instead, in the client satisfaction rankings, Manhattan Prep finished in third place with an 8.64 rating. And Magoosh finished in seventh place. Magoosh’s surprisingly low showing might have something to do with the curious outcome that examinees who bought the firm’s classes, as well as online platforms, actually did worse than those who studied using Magoosh’s online platform alone.
Do Simple Explanations Correlate with Strong GMAT Score Gains?
In the test prep business, much of the variance in customer satisfaction rankings can result from the vendor’s skill at efficiently and effectively explaining challenging concepts. But what’s surprising is that the firms ranked best at explaining difficult GMAT concepts weren’t always the ones that delivered the largest score gains.
That’s because the data showed that three firms appear to have chalked up the highest client ratings when it came to explaining difficult concepts. However, small sample sizes raise doubts about the credibility of these rankings.
For example, besides winning the customer satisfaction rankings, Target Test Prep also came out on top when it came to explaining difficult concepts. But only 20 examinees within the sample hired Target. That’s only 18 percent of Manhattan Prep’s sample size, and merely 2.3 percent of the overall sample. Moreover, Empower and the Economist, the next highest-ranked vendors in this category, account for only 13 and 21 respondents respectively. Those small samples may very well be accurate, but larger sample sizes would certainly inspire more confidence.
Interestingly enough, Magoosh placed sixth, with 127 responses. And those examinees who consistently need difficult concepts explained clearly might want to think twice about hiring Manhattan Prep. Despite stunning showings elsewhere in this survey, 103 respondents awarded that vendor a dismal ninth place in this category, with a score merely equivalent to the average among all 11 firms.
The Impact of Covid-19 on Testing – Should I Take the GMAT or GRE?
It might seem like preparing for the GMAT during Covid-19 wouldn’t be any different than before the pandemic. But this is actually a complicated topic that probably deserves its own guide here on BSchools. For the time being, here’s a brief summary of tips to consider.
At the time of this writing in October 2020, many of the test centers around the world that used to offer the GMAT remain closed. Meanwhile, GMAC has introduced a home version of the test designed for desktop and laptop computers with broadband Internet connections. This initiative partly developed as a response to archrival Educational Testing Service, which had started to roll out a home version of its Graduate Record Exam (GRE). And these days, almost all business schools requiring the GMAT allow MBA applicants to substitute the GRE—a smart option for many candidates.
BSchools believes that the home version offers the only safe way to sit for the GMAT for the foreseeable future. But should one actually do so? Now, that’s a separate question entirely.
Should MBA Applicants Take the Home-Based GMAT?
Notable differences exist between the online and test center versions. The main content difference involves deleting the Analytical Writing Assessment (AWA) from the home version, which is a big plus. But there are also big differences in the way GMAC administers this strictly closed-book test that don’t involve content but do require advance preparation. Several such distinctions have to do with security—and technology.
For example, despite what GMAC says, a desktop computer with a camera built into the top of a large screen—like Apple’s iMac models—won’t work for the online GMAT. Because GMAC’s proctors use the webcam to carefully look for evidence of cheating, examinees with widescreen desktop computers will need to buy a peripheral webcam that they can pick up and move around.
In fact, before an examinee can begin the home test, a proctor will conduct a thorough visual inspection of the room, which requires the examinee to pick the camera up and sweep around the visual field, both vertically as well as horizontally, while listening to instructions from the proctor through the computer’s sound system. The proctor is looking for evidence of “cheat sheets” taped to walls or hidden in obscure areas.
GMAC claims that this “check-in” inspection should take about 15 minutes. But it can take longer. Furthermore, at any time during the test, a proctor can interrupt the examinee and ask them to refocus the camera’s field of vision from themselves to objects within the room. And GMAC records the entire three hours worth of the camera input, the microphone audio, and the activity on the computer’s desktop, including any software “whiteboard” calculations.
Through admissions to top business schools, high scores on the GMAT have often translated to millions of dollars in additional earnings over the course of an MBA graduate’s career, and have frequently earned scholarships that defrayed significant proportions of business school tuition. For the overwhelming majority of examinees, sitting for the GMAT is already a high-stakes, high-stress proposition.
That said, we have to wonder what sort of examinee would be comfortable taking the test under even more stress, like this kind of intense scrutiny that GMAC demands from home examinees, or the restriction that home test scores cannot be canceled and only one retake at home is permitted.
It’s also important to recognize that some home examinees have reported extreme dissatisfaction with GMAC following requests for technical support and customer service that weren’t timely or adequately addressed.
Furthermore, incendiary allegations of widespread cheating have been levied in online forums against GMAC. Some individuals with inside knowledge of the home GMAT anonymously claimed that unlike the test center version, the home test used the same question bank for as long as 60 days without changes. What could go wrong? After receiving reports from clients who had previously sat for the test, that interval would have given tutors and test prep firms an ample opportunity to solve test items and share the solutions with new clients.
However, for the time being, what admissions testing alternatives do potential MBA program applicants have if they don’t want to risk exposure to SARS-CoV-2? Do candidates have any viable alternatives at all?
Skip the GMAT For Now?
Yes, many MBA applicants do indeed have a viable alternative: Don’t take the GMAT, the GRE, or any other admissions test.
Because of Covid-19, a rapidly-growing list of business schools—including several with top-25 rankings—has decided to waive their requirements for current admissions test scores, or to make their applications “test-optional.” The first to waive testing was the Darden School of Business at the University of Virginia, a disruptive strategy we first covered here on BSchools in April 2020.
Applications to Darden quickly shot up by 364 percent. That application explosion at Darden sent shockwaves through the admissions offices at competing business schools afraid of losing top candidates.
The latest business school to waive testing is the Sloan School of Management at MIT. And before Sloan, the Kellogg School of Management at Northwestern University, the McCombs School of Business at the University of Texas at Austin, and several other business schools had announced that they also wouldn’t require the GMAT or the GRE.
Keep in mind that a compelling precedent exists for such a paradigm shift. The Harvard Business School did not require any admissions testing at all for eleven years starting in 1986. An internal admissions audit HBS performed found no loss in the quality of their applicant pool—much like the results of recent audits performed by the Kenan-Flagler Business School at the University of North Carolina at Chapel Hill and the Rotman School of Management at the University of Toronto.
But overall these days, in 2020, how many MBA applicants actually have to take the GMAT in the first place? Here on BSchools, we’ve already published several guides explaining why many MBA applicants—and especially those applying to online and part-time programs—probably don’t need to take the GMAT or GRE at all. Our guides explain how most professionals with several years of experience should be able to obtain GMAT waivers in exchange for their work experience; at Kenan-Flagler, that waiver is automatic for any applicant who’s worked at least seven years.
And our guides even present examples of MBA programs that in recent years have not mandated any admissions testing—just like the Harvard Business School.