Is Math Tutoring Worth It? An Honest Cost-Benefit Guide for Parents

Is math tutoring worth it? It depends entirely on which of five common problems your child actually has. For two of them, a specific skill gap and a home where nobody can do the teaching, tutoring is the right money.

For two others, a possible learning difficulty that has never been assessed and a curriculum that does not fit your child, tutoring is the wrong purchase at any price. And for the fifth, lost confidence, it depends on how the tutoring is delivered.

This site sells no tutoring, runs no tutor marketplace, and earns nothing from any provider named in this guide, which is why the guide can give that answer instead of the one every tutoring company gives.

Below: the real 2026 costs from the providers’ own pages, what the research actually found (including the part the brochures skip), the five-problem diagnostic, and an audit plan for anyone already paying.

Quick answer: Worth it when the problem is a specific skill gap, or nobody at home can teach and supervise, AND the tutoring is delivered the way the research describes: the same tutor, two to three shorter sessions a week, a named target, and a review in about ten weeks.

Not worth it when the real problem is a possible learning difficulty that has never been assessed, or a curriculum that does not fit your child, because both have cheaper and more effective first moves, and tutoring either one is paying hourly rates to stand still.

The price reality: a private tutor averages about $25 an hour on Care.com and runs $35 to $60 an hour for most tutors on Wyzant, so two hours a week works out to roughly $215 to $380 a month, and the big learning centers publish no prices at all.

The research in one line: one-to-one tuition adds about five months of progress on average across subjects in the Education Endowment Foundation’s synthesis of 123 studies, but about two months in mathematics specifically, and the gains belong to frequent, targeted, time-boxed tutoring, not to an open-ended weekly hour.

What Math Tutoring Costs in 2026

Three stacks of coins of different heights beside a blank notebook, for what math tutoring costs in 2026 across formats
About $25 an hour on one marketplace, $35 to $60 on another, and no published price at all from the big centers.
FormatTypical costSource (October 2026)
Private tutor via Care.comAbout $25/hour average, posted rates to about $36care.com
Private tutor via WyzantMost tutors charge $35-$60/hour, tutor-setwyzant.com
Online 1:1 platform (Brighterly)$17.70-$21.40 per 45-minute lesson, about $142-$171/month at 2 lessons a weekbrighterly.com/pricing
Online marketplace 1:1 (Outschool)Generally $15-$60 per session, teacher-set, pay per classoutschool.com
Learning center (Mathnasium)No price published; the local center quotes. One independent tutoring company’s October 2026 survey sampled two centers at $449-$579/month plus a $149-$199 enrollment feemathnasium.com (no figure published)
Worksheet center (Kumon)No price published; tuition, registration, and materials fees quoted by the local centerkumon.com (no figure published)
Practice app (IXL, math only)$9.95/month or $79/year per childixl.com
Self-teaching curriculum (Math Mammoth / Teaching Textbooks)$42.50 per grade, once / $48.95 per student per yearmathmammoth.com / teachingtextbooks.com
Nonprofit platforms (Khan Academy / Zearn)$0khanacademy.org / about.zearn.org

Private tutors. The national average rate for math tutors on Care.com is $24.94 an hour, as listed on care.com at the time of checking, November 2026, across roughly 36,000 listed math tutors.

On Wyzant, tutors set their own rates and the platform states that most charge between $35 and $60 an hour, as listed on wyzant.com at the time of checking, November 2026.

The arithmetic on those published rates: two one-hour sessions a week work out to about $215 a month at the Care.com average and about $390 a month at the middle of Wyzant’s band, and one session a week halves both figures.

Online platforms. Brighterly publishes a per-lesson ladder for live 45-minute 1:1 lessons on brighterly.com: $21.40 per lesson on a 3-month plan, $20.20 on a 6-month plan and $17.70 on a 12-month plan, on packages built around two lessons a week, which works out to about $141.60 to $171.20 a month.

Note the shape honestly: the cheapest per-lesson price requires the longest commitment, which is a reason to prove the fit on a short plan first.

Outschool is a marketplace rather than a tutoring company: teachers set their own prices, families pay per class, and Outschool’s own site puts 1-on-1 sessions generally in the $15 to $60 range, with no contract and fit depending entirely on the individual teacher.

Some platforms, Varsity Tutors among them, publish no rates at all and quote only after a consultation call.

The centers publish the least.

Mathnasium’s own homepage FAQ asks what Mathnasium costs and answers with no figure at all, describing an enrollment fee and regular attendance much like a gym membership, then directing families to a local center; the national site does state that most families attend two to three times a week and that the model starts with a diagnostic assessment.

Kumon’s national site publishes no tuition, registration, or materials fee anywhere, and routes families to their local center for everything financial.

None of that makes either a bad provider, since franchise pricing genuinely varies, but it does make the first visit a sales conversation, so get the full fee stack (tuition, enrollment, materials, assessment, and cancellation terms) in writing before money moves.

A research-pattern block is worth pricing against these routes: three 30-minute sessions a week for ten weeks with a private tutor at the Care.com average works out to about 15 hours, roughly $375, and at Wyzant’s mid-band roughly $525 to $900.

What the Research Says Tutoring Can Do

The strongest evidence comes from the Education Endowment Foundation’s Teaching and Learning Toolkit. Its one-to-one tuition strand, built on 123 studies, finds about five additional months of progress on average, with primary-school children gaining more (about six months) than secondary students (about four).

The same page carries the caveat no brochure prints: in mathematics specifically, the one-to-one effect appears substantially lower, at about two months, against about six months in literacy.

The Foundation also warns that a large share of the underlying studies were not independently evaluated and that provider-connected evaluations report larger impacts, which is the institutional way of saying to discount vendor-run studies, including any a tutoring company shows you.

The small-group numbers change the buying decision. The Toolkit’s small group tuition strand (62 studies) finds about four additional months of progress at low cost, for groups of two to five pupils, with the mathematics effect at about three months, and the Foundation itself suggests trialing small-group tuition before one-to-one on cost-effectiveness, because one month less of measured gain for a fraction of the price is usually the better trade. Effectiveness falls once a group passes six or seven pupils. A large meta-analysis of tutoring programs in randomized trials by researchers Nickow, Oreopoulos and Quan (published through the Annenberg Institute) found a pooled effect of about 0.37 standard deviations, roughly three to four months of extra learning for early elementary students in the high-dosage studies and among the largest effects recorded for any education intervention, shrinking to about 0.20 to 0.25 standard deviations in larger real-world programs. A family buying commercial tutoring is buying the real-world version, not the trial version, and should hold the smaller expectation.

The conditions matter as much as the headline. The Education Endowment Foundation’s optimum pattern for one-to-one work is short, regular sessions of about 30 minutes, three to five times a week, over a set period of up to ten weeks.

The National Student Support Accelerator at Stanford distills the same research into non-negotiables: at least three sessions a week, small groups (one-to-one is ideal and up to three-to-one still works), the same consistent tutor, and trained tutors with support, with the largest meta-analysis effects coming from teachers and trained paraprofessionals rather than untrained helpers.

Targeting matters too: tuition aimed at a diagnosed, specific need outperforms tutoring aimed at math in general. Now hold the commercial default pattern next to that: one 60-minute session a week, open-ended, with whoever is on shift, aimed at this week’s homework.

Those are different products wearing the same word, and a parent paying for the second while expecting the first pattern’s results is the central disappointment this guide exists to prevent.

Nothing in these sources supports promises of one to two grade levels of gain in a few months for a typical child, and dosage actually delivered, not dosage purchased, is the variable that decides outcomes.

The 5 Problems: Which One Does Your Child Have?

Five blank index cards in a row with one pencil, for the five problems that make parents consider math tutoring
Behind in one topic, confidence gone, a possible learning difficulty, a curriculum that does not fit, or no time to help: five problems, five different right answers.

Problem 1: Behind in one specific topic. It presents as a child who was fine last year and lost this term: a missed unit, a school move, a shaky fractions term, homework that collapses on one topic while the rest holds up.

This is tutoring’s home turf, because a defined gap is exactly what the research rewards. The right purchase is a time-boxed block aimed at the named gap, with a before-and-after check, and the early-grade version of this problem most often hides in place value or regrouping.

The cheaper first move for a narrow gap is a topic worktext (Math Mammoth’s Blue Series topic books run $2 to $9 each as downloads, as listed on mathmammoth.com at the time of checking, October 2026) or free targeted lessons, with the child’s written work checked step by step against a fraction calculator when fractions are the gap.

If the gap resists two to three weeks of that, or nobody at home can teach the topic, buy the tutor, for the gap, with an end date.

Problem 2: Confidence has collapsed. It presents as avoidance, tears, guessing, and a child saying “I’m bad at math”, while the skills on paper are closer to grade level than the child’s self-story suggests.

Tutoring can help here, and parents consistently report confidence as its most visible effect, but the mechanism decides: confidence rebuilds from experienced success at the right level, and a tutor who re-runs the same too-hard classwork at a slower speed re-teaches the failure.

A good tutor for this problem starts below the failure point and engineers wins, using models a discouraged child can succeed with, like values on a number line, and the parent should say exactly that in the first conversation.

Cheaper first moves often solve it: practice dropped to the level the child can succeed at (a free placement test finds that level), short daily wins, and low-stakes math separated from schoolwork. If anxiety is severe, the first step is a conversation with the school, not a tutoring contract.

Problem 3: A possible learning difficulty. It presents as persistent, broad-based number difficulty that does not respond to ordinary good teaching, across topics and across years.

A generic tutor is the wrong first purchase here, because tutoring delivers more teaching and a learning difficulty needs identification first: the right intervention is specialist and number-sense-focused, and the wrong one spends the exact money and months the family can least spare.

The right first moves are to learn the signs (this site’s guide to the signs of dyscalculia was written for exactly this reader), to ask the school in writing about screening and evaluation, and only then to choose support, which may well be a specialist tutor chosen for the diagnosis.

Nothing in this guide diagnoses any child; it routes families to the people who can.

Problem 4: The teaching itself is the mismatch. It presents as a child fighting one program’s format daily: the mastery child trapped in a spiral book, the hands-on child drowning in a symbols-first text, the school child whose class sequence is not the one they can follow. Tutoring papers over a mismatch at hourly rates, forever.

The fix is changing the material, which costs $42.50 to $48.95 a year or $0 for the verified free spines, not $200 to $400 a month.

The tell-tale test: if the child does fine on the same math presented a different way, with a video, with manipulatives, or with a pictured model like the area model, the problem is fit, not ability, and a tutor is the most expensive way to rent a different presentation.

This site’s curriculum decision guide walks through that change step by step.

Problem 5: Nobody at home can help. It presents as ordinary skills and ordinary gaps with no available teacher: the parent works evenings, or never learned this math, or teaching their own child reliably ends in tears for both.

This is the second legitimate tutoring purchase, and it should be bought honestly as what it is, outsourced teaching and supervision, judged on reliability and fit rather than miracle claims.

The research conditions still apply (same tutor, a plan, a target), and the cheaper structures deserve a first look: a small group or center with fixed scheduling, an online platform with the same tutor weekly, or a self-teaching curriculum or interactive free program carrying the daily teaching with a human tutor bought monthly for check-ins.

What not to buy for this problem is homework-doing, because if the sessions’ output is completed homework rather than a child who can do the next set alone, the family is renting the problem’s disguise.

When Tutoring Is Worth It, and When It Is Not

Worth it: the gap is specific and named, and the block is time-boxed with a review date; nobody at home can teach or supervise, and the purchase is understood as outsourced teaching; the delivery matches the research pattern, same tutor, two to three shorter sessions a week, a written target, between-session practice assigned and checked; small-group tuition is available, because the measured cost-effectiveness beats one-to-one for most families. Not worth it: the difficulty is broad, persistent, and unassessed, where the money belongs in evaluation first; the real problem is a curriculum or classroom mismatch, fixable for the price of one tutoring session; the offer is an open-ended weekly hour with a rotating cast and no target, which is the pattern the evidence does not describe; the contract demands months of prepayment before fit is proven. Every line of both lists traces back to the costs and the research above, and a provider who argues with the shape of the first list is telling you which list they sell.

How to Tell If It Is Working

An hourglass beside an open planner and two pencils, for measuring whether math tutoring is working over about ten weeks
Measure before you start, agree the target in writing, and re-measure at about ten weeks with a test the tutor did not write.

For the parent already paying, the audit runs in six steps. One, fix the baseline with one measurement the family controls and the tutor does not: the free Math Mammoth placement tests on mathmammoth.com are end-of-year tests grouped by topic with printed score bands, free, tutor-independent, and repeatable, and a school benchmark or recent class test is the alternative. A tutor who welcomes an outside baseline is a good sign. Two, agree the target in words at the start: the named gap or goal, not “improve at math”, and if the provider cannot state it, the family states it and asks for written confirmation. Three, set the review point in advance at eight to twelve weeks of the agreed schedule, because confidence and homework mood may improve within weeks while the attainment check belongs at the block’s end. Four, re-measure with the same instrument and read the school’s own signals alongside it, comparing the two scores the way you would compare any before and after: whether the new one is less than, equal to, or greater than the baseline, strand by strand. Improvement on the tutor’s own worksheets alone is not evidence, because transfer to independent work is the definition of working. Five, audit the dosage actually delivered: count the sessions that happened, with the same tutor, at the agreed frequency, because most “tutoring did not work” is really “the tutoring in the research never took place”. Six, decide: continue if the target moved and a next target exists; change the tutor or format if delivery was faithful and the target did not move, because fit and teaching are the variables; stop and re-diagnose if two faithful blocks produce nothing, because at that point the problem is probably Problem 3 and the money belongs in assessment and specialist support.

Red Flags: Tutoring That Wastes Money

No assessment before the sales pitch. Every credible model in the research starts from diagnosis, and a provider who quotes a program before knowing the child’s gap is selling hours, not targeting. No written plan or named target.

“We will work on math” is not a plan; a plan names the gap, the materials, the session pattern, and the review date. Sessions that are homework-doing. Suddenly perfect homework with unmoved class tests is the signature. A rotating cast of tutors.

The consistency condition in the research is not a nicety, and a provider who cannot keep the same tutor with your child is not selling the product the evidence describes. Once a week, open-ended, no end date. The default commercial pattern sold as if it were the research pattern.

Long contracts and prepaid packages before fit is proven. The cheapest per-lesson prices belong to the 12-month commitments, and paying a premium for a short plan first is buying information, which is usually the right trade. Prices you cannot see until a consultation or an assessment visit.

The full fee stack belongs in writing before money moves. Guaranteed grade-level jumps. No institutional source supports them. No practice between sessions. The research and every honest tutor agree that an hour a week with nothing between sessions is close to money for nothing.

Before paying anyone, ask these eight questions: What will you assess before we start, and will I see the results? What exactly is the target for the first block, and how will we both know when it is met?

Who will teach my child each week, and what happens when that person is away? How many sessions a week, how long, and for how many weeks before we review? What should my child do between sessions, and who checks it?

What is the total cost of the first block, including enrollment, materials, and fees, and what are the cancellation terms? How will you report progress to me, and how often? What is your experience with children this age and this kind of difficulty?

Cheaper Alternatives, Priced Honestly

The $0 routes come first, because for two of the five problems they are the correct route and not a consolation prize.

Khan Academy’s complete grade courses with mastery tracking are free from a nonprofit, and Zearn’s interactive lessons are free on an individual account, both verified on the providers’ own pages in October 2026.

For diagnosis and practice, IXL’s family membership for math is $9.95 a month or $79 a year per child, as listed on ixl.com at the time of checking, October 2026, and its diagnostic shows exactly which skills are missing.

For a curriculum that does not fit, the change costs $42.50 per grade once for Math Mammoth or $48.95 per student per year for Teaching Textbooks, which teaches and grades inside its app with no parent teaching required.

Fact fluency at home rides on fact families and short daily practice, and this site’s free tools belong in that rotation for any route: a worksheet generator for between-session practice sets, a multiplication chart for the facts, and a long division calculator that shows every step when a child checks written work.

The labeled arithmetic that does this section’s persuading: one month of two-hours-a-week private tutoring at the Care.com average, about $215, equals about five Math Mammoth grades, about four and a half years of Teaching Textbooks for one child, or about two and a half years of IXL math for one child.

Alternatives replace tutoring for the confidence and curriculum-fit problems and support it for the gap and no-time problems; they do not replace a human for a child who genuinely needs one.

Frequently Asked Questions

How much does an elementary math tutor cost?

It depends on the route, and every figure here was checked on the provider’s own pages in November 2026. A private tutor found on Care.com averages about $25 an hour nationally (care.com), and most tutors on Wyzant charge $35 to $60 an hour (wyzant.com).

An online 1:1 platform, Brighterly, publishes $17.70 to $21.40 per 45-minute lesson, which works out to about $142 to $171 a month at two lessons a week.

The big learning centers, Mathnasium and Kumon, publish no prices at all; the local center quotes tuition, enrollment, and materials fees, so get the full stack in writing.

At two hours a week, a private tutor runs roughly $215 to $380 a month by simple arithmetic on the published rates.

How many tutoring sessions a week does a child need?

More often than most families buy. The research pattern that produces the measured gains is short and frequent: the Education Endowment Foundation’s synthesis found the optimum at about 30-minute sessions, three to five times a week, over a set period of up to ten weeks, and the National Student Support Accelerator’s research summary sets at least three sessions a week with the same tutor as a baseline condition. One open-ended hour a week is the commercial default, not the researched pattern, and it should be bought with that difference understood.

How long does it take for tutoring to work?

Plan on a block, not a session. The studies behind the effect sizes typically ran for up to about ten weeks of frequent sessions.

Confidence and homework mood often improve within a few weeks, but the honest check is at eight to twelve weeks: re-measure with the same test you used at the start (the free Math Mammoth placement tests work well for this) and compare.

If two faithful blocks move nothing, the problem is probably not effort; it may need assessment for a learning difficulty instead of more of the same tutoring.

Are online math tutors as effective as in-person tutors?

The evidence does not punish the screen.

The Education Endowment Foundation notes that studies involving digital technology show broadly similar effects for one-to-one tuition, and the large tutoring meta-analysis by Nickow, Oreopoulos and Quan includes effective online programs; the National Student Support Accelerator’s work finds delivery mode matters less than dosage: enough minutes, frequently, with a consistent trained tutor.

For young elementary children, the practical questions are attention span on a screen and whether the format lets them show their working, which is a fit question to settle with one trial session, not a research question.

Is tutoring worth it if my child is behind in math?

Usually yes, if “behind” means behind in identifiable topics. Targeted help for diagnosed gaps is the best-evidenced use of tutoring there is: one-to-one tuition averages about five extra months of progress across subjects in the EEF’s synthesis, and about two months in mathematics specifically.

Buy it as a time-boxed block aimed at the named gaps, with a before-and-after check. If the difficulty is broad, persistent across topics and years, and unresponsive to ordinary teaching, spend the first money on assessment for a possible learning difficulty instead, because generic tutoring is the wrong tool for that problem.

What can I do instead of hiring a tutor?

Match the alternative to the problem. For a narrow gap: free lessons and targeted practice (Khan Academy’s grade courses are free; a Math Mammoth topic worktext costs $2 to $9 as a download).

For practice and diagnosis: IXL’s family plan is $9.95 a month for math, with a diagnostic that shows exactly which skills are missing.

For a curriculum that does not fit: change the curriculum, at $42.50 a grade (Math Mammoth) or $48.95 a year (Teaching Textbooks), or free with Zearn’s interactive lessons. For confidence: supervised success at the right level, which a free placement test can locate.

A tutor is the right answer mainly when the gap is specific and stubborn, or when no adult at home can do the teaching.

The question was never really “is tutoring worth it”. It is “which problem does my child have, and is tutoring the fix for that one”. A parent who can name the problem will spend less and get more, whichever route they choose.

Our free math tools page holds the practice and checking helpers that sit inside every route: a worksheet generator for targeted sets, step-by-step calculators for checking work, and a multiplication chart for the facts underneath it all.

Try It With a Free Lesson

Every topic clicks faster with practice. Our Resource Library has free lesson packs with slides, printable worksheets and full answer keys, ready to download and teach.

Members unlock every lesson in the library, including the full Fractions, Multiplication, Place Value, Addition & Subtraction, Time & Money and Geometry & Shapes units. See the full library

Leave a Comment