When a score rises, how much credit should the tutoring company receive?
A student enrolls in an expensive SAT program, attends dozens of hours of class, completes hundreds of practice questions, takes multiple practice tests, and raises their score by 150 points.
The company places that increase in a testimonial: Our student improved 150 points.
But who actually produced those points?
Was it the instructor? The student's 60 hours of homework? Another semester of Algebra II? Repeated exposure to the test? A parent enforcing the study schedule? Free Bluebook tests and Khan Academy practice? Or some combination of all of them?
This is not an argument that tutoring does nothing. Good tutoring can save students time, identify gaps they cannot see, and turn random effort into focused work. The problem begins when a company treats every point earned during enrollment as proof that its program produced the entire increase.
Before paying thousands of dollars, parents should know what they are actually buying. Sometimes they are buying specialized guidance. Sometimes they are buying a large amount of practice packaged inside a schedule. Those are not the same service, and they should not automatically command the same price.
Volume is useful—but it is not always special
Many SAT programs are built around the same basic promise: more instruction, more questions, more homework, and more practice tests.
That can help, especially when a student is just beginning. Students need enough exposure to understand the test, notice recurring question types, rebuild forgotten content, and develop basic stamina. Research supports a measured version of this: test preparation can improve performance, but its effect is generally small and varies according to what the preparation includes. A 2025 review of 28 studies found that workbooks, emotional and motivational support, and direct instruction in test-taking skills could matter, while practice tests and sample questions alone did not show consistent benefits (Hao et al., 2025).
So volume is not worthless. It is simply not proprietary.
A family can create a basic volume plan at home with official Bluebook tests, College Board's free resources, Khan Academy, one carefully chosen workbook, and a weekly schedule (College Board). A student who needs help getting started may only need a lower-cost class, a few targeted sessions, or periodic check-ins—not the largest package available.
The important question is not How many hours does this program include? It is What is happening during those hours that my student could not reasonably do alone?
If the answer is mostly watching lectures, completing standard worksheets, and taking tests that are available elsewhere, the family may be paying a premium price for structure and volume. Structure has value—but parents should recognize it for what it is.
The higher the score, the less useful generic volume becomes
A student beginning at 900 and a student beginning at 1400 do not need the same program.
The student at 900 may have broad content gaps. They might need to relearn linear equations, punctuation, transitions, percentages, and basic reading relationships. At this stage, substantial practice can be productive because there are many repeatable skills to strengthen and many points still available.
The student at 1400 usually has a narrower problem. They may be losing points to a few advanced math concepts, subtle reading distinctions, rushed decisions, or one recurring pacing mistake. Giving that student another enormous set of general lessons can waste time and create fatigue without addressing the real limitation.
As the score rises, preparation should become less about total hours and more about precision. Which question types still produce errors? Are the misses caused by content, misreading, pacing, or execution? Does the pattern repeat across official tests? What is the smallest useful intervention?
This is where quality matters more than quantity. The student may need ten carefully chosen problems and a strong analysis of their thinking—not another 200-question assignment.
The Princeton Review's own guarantees make this point more clearly than most advertisements do. Its 1400+ course does not promise every student the same increase. The guarantee is satisfied if the student reaches 1400 or improves by 150 points, depending on the starting score. Its 1500+ tutoring works similarly: the advertised outcome is a 1500 or a 180-point increase, with the result based on a superscore after two official tests and refunds varying according to the final score (1400+ course; 1500+ tutoring). Under the company's broader Better Scores Guarantee, a student beginning at 1500 or higher is not eligible for the SAT score-improvement guarantee at all (guarantee terms).
That structure is revealing. A student beginning just below 1400 may satisfy the guarantee with a relatively small increase by crossing 1400, while a student starting much lower may satisfy it through a larger point gain without ever reaching 1400. Both can be counted as successful outcomes, but they did not make the same journey or have the same room to grow. And once a student is already at 1500, the company will not make even the broader promise that the score will rise. The fine print therefore acknowledges the exact limitation parents should keep in mind: starting score changes both the amount of realistic growth and what the company is willing to call success. Any advertised result that blends those students together without showing their baselines leaves out the context needed to judge the program's contribution.
Who did most of the work?
Suppose a student attends 20 hours of tutoring and completes another 70 hours of homework, practice tests, videos, corrections, and independent review.
The tutor may have made those 70 hours more productive. They may have selected the right materials, corrected misconceptions, and held the student accountable. That is a legitimate contribution.
But the student still performed most of the work.
This is why families should be cautious when a company says, "Our students improve by an average of 150 points." It leaves out outside study, other resources, months of school, and prior test exposure.
Older SAT research estimated about 20 math points and 10 verbal points in one analysis—and warned that even those effects might be too generous because coached students can differ in motivation, academic background, and family resources (Briggs, 2009). The exam has since changed, so those numbers cannot predict today's results. The warning still holds: a score earned after tutoring was not necessarily caused entirely by tutoring.
Good tutors should be able to explain their contribution without taking possession of the student's entire result.
The missing students in an “average increase”
An average is only as honest as the students included in it.
When a company advertises an average increase, ask whether that calculation includes:
- students whose scores stayed the same;
- students whose scores decreased;
- students who completed the course but never submitted a final score;
- students who withdrew because the program was not meeting their needs;
- students who did not finish every assignment; and
- every enrolled student, or only the students who qualified for the guarantee.
If scores are submitted voluntarily, large improvers have more reason to report them while disappointed students may quietly leave. The resulting average may describe the visible success stories rather than everyone who paid.
Guarantees narrow the group further. Princeton Review's 1400+ terms require full attendance, homework, scheduled practice tests, and an official SAT within the required period (Princeton Review 1400+). Those rules are understandable, but results among perfectly compliant finishers do not represent every purchaser.
A useful report would show how many students enrolled, completed the program, and submitted comparable final scores; how many improved, stagnated, or declined; and how the outcomes differed by starting score.
Without those numbers, an “average increase” may be accurate and still tell parents very little.
A large price tag does not guarantee a better match
High prices can buy experienced instruction and careful tracking. They can also pay for branding, advertising, long classes, and an overwhelming amount of content.
In a recent Reddit account, a parent reported spending about $2,000 on Prep Expert without seeing measurable improvement. The parent believed the generalized course failed to address the student's math, reading, confidence, and processing needs. Another student described a roughly $3,000 package, seven-hour online classes, an unrealistic workload, and insufficient help with math foundations (Reddit discussion).
These individual accounts do not prove a typical result. And this is only one of many accounts that exist around the topic. They illustrate the question every family should ask: Does this program match the reason my student is struggling? A cheaper specialist may outperform a large course for one student; another may genuinely benefit from the course's structure. Price and size do not determine fit.
What should families do next?
The right next step depends on the student's starting point and ability to work independently.
If the student is just beginning
Take a realistic Bluebook test. Sort errors into content, interpretation, pacing, and careless execution. Use free official resources plus one strong workbook on a consistent schedule. If you cannot interpret the results, a few targeted tutoring sessions from even a mildly to moderatly experienced SAT tutor can set priorities without committing you to months of instruction.
If the student has broad foundational gaps
Do not mistake SAT tricks for missing education.
A student struggling with algebra, grammar, or reading comprehension needs teaching that goes deeper than test tricks. The support should teach the underlying skill, check independent understanding, and then connect it to SAT wording. This stage may require volume—but volume should follow instruction, not replace it.
If the student is already scoring well
Avoid restarting the entire curriculum every time the score stalls.
Track errors across official tests and target the smallest set of skills or decisions producing the remaining losses. The best support may be shorter, more specialized, and more analytical than a large course. The closer the student gets to the goal, the more each paid hour should have a specific reason to exist.
What good tutoring should provide
You are paying a good tutor to notice what the student cannot yet notice alone: the difference between a content gap and a one-time mistake, what the student should stop studying, which practice matches the current level, and whether a new strategy improves accuracy without damaging pacing. That is more valuable than simply assigning hours.
The fairest claim is not, "We gave this student 180 points." It is:
"We identified the student's recurring gaps, built a focused plan, monitored the results, and helped make their independent work more effective."
The score still belongs to the student.
Seven questions to ask before paying for a program
- What will my student receive that we could not reproduce with official materials and a schedule at home?
- How does the plan change for someone at my student's starting score?
- How much work happens outside paid instruction?
- Who is included—and excluded—from your advertised average increase?
- How do you diagnose content, comprehension, pacing, and execution separately?
- What happens if the student completes the work but the score stagnates?
- As the student improves, will the program become more targeted, or will it keep assigning the same volume?
- What are the stipulations of your guarantee? What would disqualify me from the guarantee? (if applicable — some of the best companies don’t promise score a increase, just quality support)
If a company cannot answer those questions clearly, do not let an enormous package or a dramatic guarantee answer them for you.
Volume can get a student moving. Guidance can make that volume more useful. But as the score rises, quality, precision, and fit should matter more than the number of hours being sold.
Pay for what your student actually needs—not simply for more SAT.
Sources
- Briggs, D. C. (2001). The Effect of Admissions Test Preparation: Evidence from NELS:88. Chance, 14(1), 10–18. https://doi.org/10.1080/09332480.2001.10542245
- Briggs, D. C. (2009). Preparation for College Admission Exams. National Association for College Admission Counseling. https://files.eric.ed.gov/fulltext/ED505529.pdf
- College Board. The SAT: Practice and Preparation. https://satsuite.collegeboard.org/sat
- Hao, Z., Baird, J.-A., El Masri, Y., & Double, K. (2025). The Impact of Test Preparation on Performance of Large-Scale Educational Tests: A Meta-analysis of Experimental Studies. Review of Educational Research. https://doi.org/10.3102/00346543251360775
- Powers, D. E., & Rock, D. A. (1999). Effects of coaching on SAT I: Reasoning test scores. Journal of Educational Measurement, 36(2), 93–118. https://doi.org/10.1111/j.1745-3984.1999.tb00549.x
- The Princeton Review. Better Scores Money Back Guarantee Terms and Conditions. https://www.princetonreview.com/legal/guarantee-better-scores
- The Princeton Review. SAT 1400+ Course. https://www.princetonreview.com/college/sat-honors-course
- The Princeton Review. SAT 1500+ Tutoring. https://www.princetonreview.com/college/sat-1500-tutoring
- Witty-Strain-1111. (2026). Buyer Beware: Our Experience with Prep Expert SAT Prep [Online forum post]. Reddit. https://www.reddit.com/r/Sat/comments/1tef4hb/buyer_beware_our_experience_with_prep_expert_sat/
Evidence note
The 2025 review combined different educational tests and preparation methods, so its effect size cannot be converted into a fixed number of digital SAT points. The older SAT studies examined earlier versions of the exam; they help explain the score-attribution problem but cannot predict the exact result of a current digital SAT program. Reddit accounts represent individual experiences and are included as illustrations, not proof of a typical result.