Tutoring and test-prep software in Baltimore: the session is a service, the worksheet is not

The market this is actually about

Start with what the tutoring trade in Maryland actually looks like, because the popular image — a retired teacher and a kitchen table, or a national franchise in a strip mall — misses where the money and the headcount really sit. I pulled the Census Bureau’s County Business Patterns files for 2023, downloaded on 17 September 2026, and counted NAICS 611691, “Exam Preparation and Tutoring,” for the United States, Maryland, and every Maryland county that reports a nonzero figure for it.

Nationally, the industry is not small: 9,820 establishments, 108,755 employees, and $3,014,523,000 in annual payroll — a touch over three billion dollars, averaging $27,718 of payroll per employee. Maryland holds 192 establishments, 2,017 employees and $54,937,000 of payroll, almost exactly proportional to the state’s roughly 1.8% share of the national population. The state’s own per-employee average, $27,237, sits within a rounding error of the national figure. None of that is surprising. What is surprising is how unevenly it is spread once you look at individual counties, and what that unevenness implies about who is actually running these businesses.

Montgomery County is the state’s tutoring capital by volume, and it is not close — 56 establishments and 540 employees, four times Baltimore City’s establishment count and nearly three times its employment, which tracks with a county built around some of the country’s most competitive public magnet programs and a dense private-school market feeding them. But Montgomery’s per-employee payroll, $27,791, sits almost exactly at the state average. The outlier is Baltimore County — the suburban ring, not the city — where 269 employees across 23 establishments carry $52,063 of payroll per head, 91% above the state average and the highest figure on this list by a wide margin. Baltimore City itself, with a comparatively modest 14 establishments, pays $39,140 per employee: 44% above the state average and 41% above Montgomery County’s, despite running on a quarter of Montgomery’s establishment count.

Read together, the shape is this: Montgomery County has the volume — a large number of comparatively ordinary-paying tutoring operations serving a huge, exam-anxious population. Baltimore County and Baltimore City have fewer, better-paid operations, consistent with a market of specialized test-prep and subject-expert practices rather than large storefront chains staffing shifts of part-time college students. That distinction matters for software, because a scheduling tool built for “lots of similar part-time tutors, high turnover, simple hourly billing” is a different piece of software from one built for “a small roster of highly paid specialists, complex packages, and a school contract or two.” Baltimore looks much more like the second. Twelve of the twelve counties reporting a figure for this industry account for 187 of the state’s 192 establishments and 1,974 of its 2,017 employees — the rest sit in counties too small for the Census Bureau to publish even a noise-adjusted figure, which is itself a fact about how concentrated this trade is around the Baltimore–Washington corridor.

Two customers, two rulebooks

The single fact that organizes everything else in this article is that a Baltimore tutoring company can have two entirely different kinds of customer, and Maryland law treats them as though they belong to different industries. One is a parent, paying directly, choosing a tutor the way they would choose a piano teacher. The other is a public school system, paying under a contract, often using state money earmarked for a specific purpose.

That second category exists at real scale because of the Blueprint for Maryland’s Future, the state’s decade-long education funding and policy overhaul — established by Chapter 771 of 2019 and revised by Chapter 36 of 2021 after the General Assembly overrode a gubernatorial veto, originally projected to raise state education funding by roughly $3.8 billion a year at full phase-in. It has not survived contact with the state’s own budget unscathed: facing a multi-billion-dollar structural deficit, the Moore administration has since proposed scaling parts of it back, and nonpartisan legislative analysts have projected the cutbacks compounding to roughly $2.2 billion less through 2029 and $6.2 billion less through 2033 than the law originally promised. Tutoring specifically kept growing as its own budget line even as the broader Blueprint got trimmed: in October 2023, Governor Wes Moore announced the Maryland Tutoring Corps, a $28 million grant program aimed at scaling “high-dosage” math tutoring for middle schoolers in Baltimore City, Baltimore County, and Wicomico County, with a matching structure reported to bring the total investment as high as $30 million and a stated goal of reaching more than 80,000 students and 1,000 tutors. That is real money flowing through real school-system contracts, and every dollar of it turns the vendor on the other end of the contract into an entity subject to a specific stack of Maryland and federal law that a purely private tutoring practice never touches.

The rest of this article follows that fork. Two sections cover what changes the moment you are under a school contract: a doubled background-check requirement, and a data-privacy statute that names you an “operator.” One section covers what governs you instead when there is no school contract: the federal children’s privacy rule. And two sections — tax and the automatic-renewal law — apply to both sides regardless of who is paying, which is exactly the kind of rule that is easy to miss if you build for only one customer type at a time.

The employee you have to interview before you hire them

If your company contracts with a Maryland public school system or a nonpublic school to provide tutoring, you are what Education Article §6–113.2 calls a “contracting agency” — MSDE’s own guidance defines the term as “an entity that contracts with a county board or nonpublic school to provide a service to a school or the students of a school,” and a tutoring vendor is about as clean a fit for that definition as exists. The law has applied to any applicant hired on or after 1 July 2019.

Before you assign anyone to a position involving “direct contact with minors,” the statute requires that person to submit three things: the contact information for their current employer, every former school employer, and every former employer where they held a position involving direct contact with minors, going back ten years; a written consent authorizing those employers to release records relating to child sexual abuse or sexual misconduct; and a written statement disclosing whether they have ever been investigated, disciplined, discharged, or asked to resign over such an allegation, or had a professional license or certificate suspended or revoked over one. §6–113.2(a)(4) defines direct contact with minors as “the care, supervision, guidance, or control of, or routine interaction with, a minor” — language that says nothing about physical presence, and MSDE’s guidance explicitly extends the same reasoning to coaches and substitutes, groups a one-on-one tutor would obviously sit alongside.

Your obligations as the contracting agency do not end with collecting the paperwork. You must request the review from each listed employer, who then has 20 days to respond using an MSDE-approved form; if three documented attempts get no response, you may proceed but must note what is missing. You must maintain a record of each tutor’s review and, before assigning that tutor to a specific school, notify the school of any affirmative response uncovered — the school can then object to that specific placement. None of this is a public record under the Maryland Public Information Act, which is worth knowing before you design where it lives in your own system.

There is a narrow fast-start option, and it has a bite that matters for a solo virtual tutor specifically. A county board or nonpublic school — not a contracting agency — may hire an “emergent employee” for up to 60 days while the review is pending, but only if the applicant has provided all required information, the employer has no disqualifying knowledge, the applicant swears they are not disqualified, and — the condition that matters here — the applicant is not authorized to work alone with minors unless they are in the immediate vicinity of a permanent employee, or, for a driver, under audio and video monitoring reviewed promptly by administrators. A one-on-one video tutoring session is, definitionally, working alone with a minor. Read literally, that closes the emergent-employee shortcut for exactly the staffing model — solo virtual tutors, hired quickly to meet a school contract’s ramp-up — that a fast-growing vendor is most likely to want it for. And in any case the emergent-employee path is only available to the school or county board directly, not to a contracting agency assigning its own staff, so a tutoring vendor cannot use it at all.

And a separate background check, run by a different agency

MSDE’s guidance for §6–113.2 answers this directly, in its own numbered question 6: “Does the Criminal History Background Check required by Family Law Article cover this requirement? No.” The employment-history review above is one obligation. A fingerprint-based state and FBI criminal history check, authorized under Family Law §5–561 and administered under COMAR 12.15.02, is a second, independent one, and a contracting agency’s tutors have to clear both.

The mechanics of the fingerprint check are ordinary for anyone physically present in Maryland: fingerprints taken at a designated law enforcement or approved agency, submitted with a form to the Criminal Justice Information System Central Repository. They stop being ordinary for a remote hire, which is precisely the staffing model an ambitious tutoring vendor wants to use to serve a Baltimore City Schools contract from anywhere in the country. Maryland had to issue a specific clarification on this in July 2025, jointly from the State Superintendent of Schools and the Secretary of the Department of Public Safety and Correctional Services, titled Clarification Regarding Fingerprinting Requirements for Out-Of-State Vendors. It states that COMAR 12.15.02.03 requires the approved DPSCS form but “does not require such individuals to complete this process in-person” — though a local school system retains discretion to insist on it anyway. The out-of-state path: request a Maryland fingerprint card by writing to the Central Repository’s Pikesville P.O. box or calling one of two published numbers, get fingerprinted locally using that card, and mail the card and fee back for processing, with a response typically arriving in 10 to 15 days. Fingerprint cards from other states or the FBI are explicitly not accepted as substitutes.

The fee itself stacks from several published line items rather than one number. COMAR 12.15.01.15 sets a $18 CHRI access fee and a $20 fingerprinting service fee for the Central Repository’s own processing, on top of which the FBI portion of a combined state-and-federal check currently runs about $16.50 — roughly $54.50 in total before any add-on such as a HAZMAT threat-assessment fee, which does not apply here. That is a per-tutor, recurring cost that a hiring pipeline needs to track alongside expiration and renewal, not a one-time line item to forget about after the first hire.

The data the contract hands you

The same school contract that triggers the double background check also makes your company an “operator” under Maryland’s Student Data Privacy Act, Education Article §4–131, originally enacted in 2015. The statute’s own definition ties directly to federal law: an operator is one who “engages with institutions under the school official exception of the federal Family Educational Rights and Privacy Act” — the FERPA provision, 34 CFR §99.31(a)(1), that lets a school share student records with a contracted vendor performing an institutional service — and who operates under a contract with a public school or local school system to provide a site, service, or application that processes “covered information” for a PreK–12 school purpose.

Covered information is defined broadly and specifically, in a 23-item list at §4–131(a)(2), that goes well past what most builders would guess. Alongside the obvious — educational records, name, address, phone, email, grades and test results — it explicitly includes special education information, criminal records, medical and health records, Social Security number, biometric information, socioeconomic information, food purchases, political and religious affiliations, text messages, search activity, photos, voice recordings, disciplinary information, and “persistent unique identifiers” — a defined term that reaches cookie identifiers, device identifiers, hashed emails and phone numbers, and mobile ad identifiers. A tutoring platform tracking which practice problems a student got wrong, recording a video session for the parent to review, or running an ordinary marketing analytics cookie on its own login page is handling covered information under at least three or four of those categories at once.

An operator must protect that information with reasonable security procedures and delete it on the school’s request. It may not knowingly engage in targeted advertising based on information gathered through the service; may not use the data to build a profile on a student outside a PreK–12 purpose; may not, subject to narrow exceptions, sell a student’s information; and may not disclose covered information except through a specific list of permitted channels — legal compliance, safety, a service provider bound by the same restrictions, or with the affirmative, clearly-noticed consent of the student’s parent or guardian. None of this is exotic privacy law by national standards, but it is a real, specific compliance surface that a generic tutoring-scheduling SaaS built for the consumer market is not designed to satisfy, and a school procurement office will ask about it directly before signing.

The good news, if you can call it that, is in subsection (b): the entire statute does not apply to a general audience website, service, or application. A direct-to-parent tutoring platform with no school contract sits entirely outside §4–131, no matter how much student data it collects. That single line is the reason the two sides of this market are, legally, almost unrelated businesses.

The same platform, a different child, no contract

Outside a school contract, Maryland’s own student-privacy law does not reach you. Federal children’s privacy law does, and as of this year it reaches further than it used to. The Children’s Online Privacy Protection Rule, codified at 16 CFR Part 312, applies to any operator with actual knowledge that it is collecting personal information from a child under 13 — a bar that any tutoring platform serving elementary or middle-school students clears without much argument, since age and grade level are exactly the information you need to match a student to a tutor in the first place.

The FTC published amendments to the rule in the Federal Register on 22 April 2025, effective 23 June 2025, with full compliance required by 22 April 2026 — a deadline that had already passed by five months when this was written, which means the amended rule is now the live standard rather than a coming change. Three changes matter most for a tutoring build. The definition of “personal information” now explicitly includes biometric identifiers, which reaches anything from a voice-recorded session to a facial-recognition attendance check. Operators must obtain a separate, specific verifiable parental consent before disclosing a child’s information for a purpose that is not integral to the service itself — a single blanket signup consent no longer covers, say, sharing data with an ad network, even if it once covered the service’s core function. And operators must maintain a written information security program and a written data retention policy, rather than an implicit practice of keeping everything indefinitely. A direct-to-parent platform onboarding families this year needs a consent flow and a retention schedule built to the current rule, not the pre-2025 one still described in a lot of older compliance checklists still circulating online.

The worksheet is not the session

Here is the split that gave this article its headline, and it applies to both sides of the market equally, regardless of who is paying. Since 14 March 2021, following Senate Bill 787, Maryland taxes the sale or use of a “digital product” or “digital code” at the standard 6% sales and use tax rate, codified primarily at Tax-General §11–101(c–4). A digital product is defined as one “obtained electronically by the buyer or delivered by means other than tangible storage media,” and the statute’s own list of examples includes, among other things, a book “transferred electronically” and a newspaper, magazine, or similar product transferred electronically — language a downloadable PDF worksheet or practice test fits without much stretching.

A live tutoring session is not taxed, and it is not taxed for two independent reasons that reinforce each other. First, Maryland only taxes services specifically enumerated under §11–101(m), and tutoring is not on that list — “any service that is not enumerated as a taxable service is not subject to the sales and use tax regardless of how the service is delivered,” per the Comptroller’s own guidance. Second, even if you tried to characterize a video-delivered session as a digital product, §11–101(c–4)(3)(iv) separately excludes “a professional service obtained electronically or delivered through the use of technology” from the very definition of a digital product. A Zoom tutoring session is untaxed on both counts.

A downloadable worksheet or a prerecorded course is where a private tutoring company’s tax exposure actually starts, and this is the part that surprises people who assume “online education” is a single tax category. The statute carves out an exemption for “prerecorded or live instruction by a public, private, or parochial elementary or secondary school… or a public or private institution of higher education” — but that exemption is written around the identity of the seller, not the nature of the content. It names schools. A private tutoring or test-prep company is not a school under this clause, however many students it serves, and does not get the exemption merely because the material looks and functions like classroom instruction. There is a second, narrower carve-out for “instruction in a skill or profession in a buyer’s current or prospective business, occupation, or trade” that is not prerecorded and features a genuinely interactive, contemporaneous element — language clearly aimed at professional continuing-education content, and a poor fit for a tenth-grader’s algebra worksheet, which is neither about the buyer’s occupation nor, being a static PDF, prerecorded in the relevant sense so much as simply not live at all. Practically: a downloadable practice test or a recorded SAT-strategy video course, sold by a Baltimore tutoring company to a Maryland customer, is a taxable digital product at 6%, full stop, unless you are able to construct a genuinely live, interactive delivery that fits the narrow occupation-focused carve-out — which most K-12 academic content simply will not. This is how we read the statute for the businesses we build for, and it is not tax advice; a company selling meaningfully into this line should confirm treatment with the Comptroller’s office or its own accountant before pricing a catalog.

Bundling changes the answer again, and Maryland’s own guidance, Business Tax Tip #29, states the rule in terms that translate directly into a checkout decision: the entire charge for a bundle of a taxable digital product and a non-taxable service is subject to tax if the dominant purpose of the sale is to obtain the digital product, but the tax does not apply where the digital product is “an inconsequential element for which no separate charge is made.” A monthly tutoring package that throws in a free companion workbook, priced as one number, stays untaxed — the sessions are obviously the point. The same package itemized as “$280 for sessions + $20 for materials” is two line items with two different tax treatments, and it is the itemization, not the underlying content, that decides it. This is a genuinely useful piece of information for anyone pricing a package, and it is also a direct argument for why a checkout built for this business needs a per-line-item tax flag rather than a single storewide tax setting.

The subscription itself is regulated

Most tutoring companies sell some version of a recurring package — four sessions a month, an unlimited monthly plan, a semester block that auto-renews. Maryland’s automatic renewal statute, Commercial Law §14–1329, in force since 1 June 2026, reaches every one of them regardless of size. The statute defines an automatic renewal simply as a paid subscription “automatically renewed for a subsequent term” — a draft version of the bill limited this to terms of more than a month, but that qualifier was struck before enactment and survives only in a different subsection governing automatic credit-card charges specifically. There is no minimum term that exempts a short monthly package.

The operative requirement is that cancellation be “at least as easy to use as the mechanism the consumer used to consent,” reachable through the same medium — sign up on the website, cancel on the website, with a prominent link or button, not a phone call required to undo an online signup. Where the initial term runs a year or longer, renewal notice must go out not less than 15 and not more than 45 days ahead; for a free trial longer than 14 days, the window is 3 to 21 days. Enforcement sits exclusively with the Attorney General — the statute rules out a private right of action — with civil penalties running up to $10,000 for a first violation and $25,000 for a subsequent one. For a tutoring platform, the practical build requirement is a cancel path that lives inside the parent’s own account, one click deep, and a notice schedule computed from the customer’s actual current renewal date rather than the date they originally signed up, since a skipped week or a paused month shifts that date without most billing systems noticing.

The trader’s license nobody expects, and the three-visit rule everyone forgets

Two more constraints attach to a Baltimore tutoring business regardless of who its customers are, and both are easy to miss because they are not written with tutoring in mind at all. The first is Business Regulation §17–1807, Maryland’s trader’s license. It attaches to selling goods, not services — pure tutoring never triggers it — but the moment a company keeps and sells physical stock-in-trade alongside its sessions, printed practice books, branded folders, an exam-prep kit, it needs one, and in Baltimore City the fee scales sharply with the value of that inventory: $20 if it is worth $1,000 or less, $40 up to $5,000, $80 up to $10,000, $160 up to $50,000, $375 up to $100,000, $1,000 up to $300,000, $1,500 up to $750,000, and $2,125 above that — an over hundred-fold range across eight tiers, versus a flat $15 or $20 uniform fee that some other Maryland counties offer as an alternative. A company that only ever ships a downloadable PDF never touches this table, because a digital file is not stock-in-trade under the statute’s own inventory-based fee structure; the day it keeps a box of printed workbooks to sell, it does, and the fee is not proportional to how small the company otherwise is.

The second is the same limit every home-based Baltimore business in this series runs into: Baltimore City Code Art. 32, §15–507(d) caps a home occupation at three customer visits a day and ten a week, forbids shipments that draw attention to the residential premises, and bars odors or disturbances detectable off the property. A tutoring practice run from a rowhouse can meet three families a day, in person, full stop — a fourth walk-in is a code violation regardless of how good the tutor is. It is worth being precise about what it does not reach: the cap counts physical visits to the property, not sessions of any kind, so a home-based practice running unlimited video sessions alongside a handful of in-person ones stays inside the rule as long as the in-person count is tracked and enforced, which is exactly the kind of ceiling a scheduling system should refuse to overbook rather than let a family find out by knocking on a door that will not open a fourth time that day.

What a tutoring company’s own storefront can look like

Everything above is about compliance. It is worth showing, concretely, what the commercial upside looks like when a tutoring business builds its own storefront around downloadable materials rather than selling only time.

Edtalk, a Lithuanian tutoring and exam-prep company, runs exactly this pattern, and it is worth looking at as a real, working example rather than a hypothetical. As of this writing its store carries 66 downloadable products — almost entirely mathematics, with a smaller English-language line — organized by grade level, subject, and exam type: mock entrance exams pitched at specific competitive schools, national exam (“VBE” and “PUPP”) practice sets, grade-by-grade test booklets sold as structured trimester sequences, and diagnostic sets aimed at pinpointing a specific knowledge gap before a family commits to paid sessions. Delivery is entirely digital: files reach the buyer by email once payment clears, with no shipping, no inventory, and no physical stock-in-trade at all — the trader’s license question above simply never arises for a catalog built this way.

The pattern generalizes cleanly to a Baltimore market that has none of the equivalent Maryland-specific content yet: a bank of practice items pegged to MCAP grade-level expectations, mock tests calibrated to a specific competitive high school’s admissions exam, or a diagnostic packet a parent can buy for twenty dollars before committing to a $320-a-month package. It is also, precisely because it is a real downloadable-product catalog built by a tutoring company rather than a general-purpose worksheet marketplace, a useful reference point for the tax analysis above: every one of those 66 products would be a taxable Maryland digital product if sold to a Maryland buyer, for exactly the reason a Baltimore company’s own PDF catalog would be — a private tutoring company, however large, is not “a school” in the sense the exemption requires.

What the software stack actually costs, meter by meter

Tutoring-specific scheduling and billing tools price in three distinct ways, and the difference between them compounds enormously at scale. TutorBird prices flat and per-seat: $16.95 a month for a single tutor, plus $4.95 a month for each additional tutor or administrator — predictable, and cheap for a solo operator. Teachworks prices flat plus per-lesson: an entry rate of $15 a month plus $0.25 per student lesson logged — still fairly predictable, since it scales with actual usage rather than revenue. TutorCruncher is the outlier, and it is the one that behaves exactly like the hidden-percentage software this series keeps finding in every vertical: its published tiers are a Pay-as-you-go plan at $30 a month plus 1% of revenue, and a Startup tier priced purely at 0.65% of revenue with no separate base fee at all. A vendor whose price is a percentage of your sales is, by construction, a vendor whose bill grows exactly as fast as your business does — permanently.

On the storefront side, for a company selling downloadable materials, Thinkific’s Basic plan runs $40 a month billed annually as of 2026, focused on the teaching and content-delivery side; Kajabi’s entry plan, restructured in January 2026, runs $143 a month billed annually, bundling in email marketing, funnels, and community tools that a pure worksheet catalog does not need. Payment processing sits underneath either choice at its own standard rate — Stripe’s published US card rate is 2.9% + 30¢ per transaction, a meter that behaves very differently on a $320 monthly subscription charge (about 1.02% of that charge) than on a $12 downloadable worksheet (about 5.4% of it), which is itself a reason a low-priced single-item storefront quietly costs more, proportionally, to run than a subscription business does.

Put a modeled Baltimore company through the whole stack. At 150 subscription students paying $320 a month, with a downloadable-materials storefront doing a modest 8% of that revenue on top, total revenue runs $622,080 a year. TutorCruncher’s Startup tier at 0.65% of the subscription revenue is $3,744. Stripe, charging 1,800 subscription payments across the year, comes to $17,244. A Thinkific plan for the storefront adds $480. The total published stack is $21,468, or 3.45% of revenue.

The percentage barely moves across a nearly sevenfold range in company size, which is exactly the signature of a stack dominated by proportional meters rather than fixed costs — and it is why the question “is this worth building custom” is really a question about how much of that 3.4–3.6% you can convert into a fixed cost instead.

What custom actually costs, and when it does not pay

Our prices are published and fixed: a one-week Prototype Sprint is $3,500, a custom online store starts at $6,000, and a custom app, internal tool, or operations system starts at $12,000 — agreed before we start, and you own every line of code.

The honest question is the same one it always is: which part of that percentage stack does a build actually replace? The payment processing does not go away — Stripe or Square’s rate applies to money moving through your own software exactly as it does through a rented one, and nothing we build makes a card transaction cheaper. What a build replaces is the TutorCruncher-style percentage-of-revenue scheduling fee and the storefront platform fee — at 150 students, $3,744 + $480 = $4,224 a year. Against that, a $6,000 online store pays back in 17.0 months, and a full $12,000 build in 34.1 months.

At 60 students, a payback stretching past six years for a full build is not a reasonable spend, and we would say so on the call. At 150 the online store is a clearly good decision and the full build is a defensible one. At 400, both are close to obvious — not because our price moved, but because the percentage meter on the other side scaled with a business that got bigger while the rent stayed proportional.

What we would actually build

Everything above points to one architectural fact: a tutoring company’s software has to know, per record, which of two very different regulatory regimes applies — and it changes not by company but by customer, sometimes within the same roster. Five structures carry that distinction where a generic CRM or storefront throws it away.

A tutor record carries dual background-check status as first-class fields, not a note in a spreadsheet: the §6–113.2 employment-history review, with its ten-year employer list, consent form, and self-disclosure answers, tracked to completion or to the documented-three-attempts fallback; and, separately, the Family Law §5–561 fingerprint check, with its own submission date, expected 10–15 day turnaround, and result — two independent statuses, because MSDE’s own guidance says plainly that one does not substitute for the other. A tutor cannot be assigned to a school-contracted student until both are clear, and the system should refuse the assignment rather than rely on someone remembering to check.

A student or family record carries a channel flag — school-contracted or direct-to-parent — because that single flag determines which privacy law applies, which consent language is required, and whether the record even counts as “covered information” under §4–131 at all. A school-contracted record needs a data-use agreement reference and a deletion-on-request workflow reachable by the school; a direct-to-parent record needs a COPPA-compliant, separately-scoped parental consent captured at signup and revisited whenever a new data use is introduced.

A session record is a professional service, tagged as such, so it never accidentally acquires a digital-product tax flag no matter how it is delivered or recorded. A product record — a worksheet, a practice test, a prerecorded course — carries the opposite default: taxable digital product at 6%, unless the seller of record is an actual accredited school rather than a tutoring company, in which case the exemption applies and the flag flips. A bundle or package record computes its own tax treatment from a dominant-purpose rule rather than a single storewide setting: sessions with an inconsequential, non-itemized extra stay untaxed; anything itemized with a separate charge for a digital component taxes that component on its own.

And a subscription object is a first-class record with an interval, a current renewal date that recomputes every time a session is skipped or paused, and a cancel path that lives in the parent’s own account, one click deep, satisfying §14–1329’s “as easy to cancel as to sign up” test by construction rather than by policy memo. None of these five structures is exotic engineering. The reason they do not exist in an off-the-shelf tutoring CRM is that a product sold into every state cannot afford to model one state’s dual background-check regime, one state’s digital-product carve-outs, and one city’s trader’s-license table, and a general-purpose builder should not try. If you want to see the shape of what we mean, our demos page has a working e-commerce build you can click through.

Build, or keep paying

The honest test, and the only list in this article:

• Keep renting when you are under roughly a hundred students, when you have no school contract and therefore none of the §6–113.2 or §4–131 obligations to model, when your downloadable-materials line is nonexistent or trivial, and when nobody on your team is retyping the same student record into two systems because your scheduling tool and your storefront do not talk to each other.
• Build when a percentage of every session fee is going to a scheduling platform on top of what card processing already costs, when you are pursuing or have won a school contract and need dual background-check tracking that a consumer-facing tutoring CRM was never built for, when your downloadable catalog is priced and taxed by guesswork rather than by a rule the checkout enforces, or when your renewal notices are scheduled from a signup date rather than the current one and you genuinely do not know whether that is a §14–1329 problem waiting to surface.

Most companies we talk to in this space land in the middle: keep the payment processor, keep a point tool for pure scheduling if it is working, and own the parts where Maryland law actually differentiates — the background-check pipeline, the data-consent flow, and the storefront’s tax logic. That is a $6,000 to $12,000 decision, not a rebuild, and it is the one we recommend most often.

When you should not call us

If you are one tutor with a dozen regular families and a shared calendar link, software is not your bottleneck and we would be a bad spend — a $16.95-a-month point tool and a Venmo request will carry you further than a custom build for a long time. If you are chasing a school contract you have not yet won, do not build the compliance machinery first; talk to the procurement office, understand what they will actually require in a data-use agreement, and come back with something specific. If the thing that is actually hurting is finding enough qualified tutors, or a marketing problem, we cannot fix either of those and will say so on the call rather than after the invoice.

How we work

Fixed price, agreed before we start. You talk to the people writing the code — there are two of us and no account layer. You own every line, in your own repository, on your own infrastructure, from the first commit. We start with the one screen costing you the most, which in this trade is usually either the background-check pipeline for a school contract or the storefront checkout for a downloadable-materials line. If a week of work would tell us both whether the rest is worth doing, that is what the $3,500 Prototype Sprint is for, credited against the full project if you go ahead.

Questions we get from tutoring and test-prep owners

Does a private tutor need a license in Maryland?

No. Maryland has no state license for tutoring the way it licenses a cosmetologist or a home improvement contractor. You do not need a teaching certificate, a degree, or a permit from any state agency to charge a family for algebra help. What Maryland regulates instead is not the act of tutoring but three things around it: who you are allowed to be alone with and under what screening, whose data you are allowed to touch, and how your invoice is taxed depending on whether it describes a session or a shipped or downloaded item. A single-person tutoring practice working directly with families triggers almost none of this. The moment you sign a contract with a public school system, or start selling a downloadable worksheet, that changes.

What background checks does Maryland require for a tutor working under a school contract?

Two, run under two different laws, and MSDE’s own guidance is explicit that neither substitutes for the other. First, under Education Article §6–113.2, your company becomes a “contracting agency” the moment it contracts with a county board or nonpublic school to provide a service to a school or its students, and before assigning anyone to a position involving “direct contact with minors” — defined as “the care, supervision, guidance, or control of, or routine interaction with, a minor” — you must review that person’s employment history for the past ten years, including a written consent to release child sexual abuse and misconduct records and a written self-disclosure. Second, under Family Law §5–561 and COMAR 12.15.02, that same tutor separately needs a fingerprint-based Maryland and FBI criminal history check through the Criminal Justice Information System Central Repository. MSDE guidance for §6–113.2, question 6, states plainly: this “is a separate requirement from the Criminal History Background Check (fingerprinting) requirements… You must continue to meet the criminal history record check requirements, as well as the requirements of this law.” Neither check is optional and neither one covers the other.

Does the fingerprint background check apply to virtual tutors who never physically enter Maryland?

Yes, and Maryland had to write a clarification to say so in plain terms. A July 2025 joint guidance from the State Superintendent of Schools and the Secretary of the Department of Public Safety and Correctional Services, titled “Clarification Regarding Fingerprinting Requirements for Out-Of-State Vendors,” addresses exactly this: prospective out-of-state virtual tutors. COMAR 12.15.02.03 requires the approved DPSCS form but “does not require such individuals to complete this process in-person,” and an out-of-state applicant instead requests a Maryland fingerprint card by mail from the Criminal Justice Information System Central Repository in Pikesville, has it taken locally, and mails the card and fee back for processing, with a response in 10 to 15 days. A tutor who has never set foot in Maryland and never will can still be a “covered individual” who must be fingerprinted before a Baltimore-area school system will let them near a student.

Is my tutoring company an “operator” under Maryland’s Student Data Privacy Act?

Only if you are under contract with a public school or local school system. Education Article §4–131 defines an “operator” as an individual or entity that “engages with institutions under the school official exception of the federal Family Educational Rights and Privacy Act” and operates “in accordance with a contract or an agreement with a public school or local school system” to provide a site, service, or application used for a PreK–12 school purpose. Subsection (b) carves out any “general audience” website, service, or application entirely — so a direct-to-parent tutoring platform with no school contract is not covered by this statute at all, whatever data it collects. A vendor delivering Blueprint-funded high-dosage tutoring under a Baltimore City Schools contract is squarely inside it; a family paying you directly through your own website is outside it.

What counts as “covered information” under Maryland’s Student Data Privacy Act?

Twenty-three enumerated categories, and the list is broader than most builders assume. Education Article §4–131(a)(2) defines it as information that could identify a student “with reasonable certainty” and specifically lists educational records, first and last name, home address and geolocation, phone number, email, test results and grades, special education information, criminal records, medical and health records, Social Security number, biometric information, socioeconomic information, food purchases, political and religious affiliations, text messages, student identifiers, search activity, photos, voice recordings, disciplinary information, online behavior linked to a student, persistent unique identifiers such as cookie IDs and device IDs, and “confidential information” as defined by the Department of Information Technology. An operator may not use it for targeted advertising, may not build a profile from it outside a PreK–12 school purpose, and may not sell it, subject to narrow exceptions — and must delete it within a reasonable time if the school asks.

Does COPPA apply to an online tutoring platform?

If you have actual knowledge you are collecting personal information from a child under 13 — which any elementary or middle-school-facing tutoring platform does — yes, and the rule got materially stricter in 2026. The FTC published amendments to the Children’s Online Privacy Protection Rule (16 CFR Part 312) in the Federal Register on 22 April 2025, effective 23 June 2025, with full compliance required by 22 April 2026 — a deadline that had already passed by the time this was written. The amended rule adds biometric identifiers to the definition of personal information, requires a separate, specific verifiable parental consent before disclosing a child’s data for any purpose that is not integral to the service itself (targeted advertising being the clearest example), and mandates that operators maintain a written information security program and a written data retention policy rather than keeping data indefinitely by default.

Is tutoring taxable in Maryland?

The session is not. A live tutoring session is a personal service, and Maryland only taxes services that are specifically enumerated in Tax-General §11–101(m) — tutoring is not one of them. It is also independently excluded from the definition of a taxable “digital product” regardless of delivery method, because §11–101(c–4)(3)(iv) states a digital product “does not include… a professional service obtained electronically or delivered through the use of technology.” A downloadable worksheet, practice test, or prerecorded video course is a different animal. Since 14 March 2021, Maryland taxes digital products at the standard 6% rate, and a downloaded PDF or a prerecorded course sold by a private tutoring company — as opposed to an actual public, private, or parochial school — does not qualify for the school exemption in §11–101(c–4)(3)(i), because that exemption names schools, not tutoring companies.

Do I have to charge sales tax on a bundle of live sessions plus a workbook?

It depends on how you price it, and Maryland’s own bundling rule turns on which part is the point of the purchase. Business Tax Tip #29 states that the entire charge for a bundled transaction involving a digital product and a non-taxable service is subject to tax if the dominant purpose of the sale is to obtain the digital product — but the tax does not apply to a personal or professional service that involves a digital product only “as an inconsequential element for which no separate charge is made.” Practically: twelve live sessions with a free companion workbook thrown in, priced as one line, stays untaxed, because the sessions are obviously the point. A package billed as “$40 for the session, $15 for the workbook PDF” on the same invoice is two line items with two tax treatments, and the $15 owes 6%.

Does Maryland’s automatic renewal law apply to a monthly tutoring package?

Yes, and there is no minimum term that exempts a short package. Commercial Law §14–1329, in force since 1 June 2026, defines an automatic renewal as any paid subscription that is “automatically renewed for a subsequent term” — a qualifier limiting this to terms of “more than 1 month” was struck from the bill before enactment and survives only in a different subsection about auto-charging a credit card. A four-session-a-month package renewing monthly is inside the law. The cancellation mechanism must be “at least as easy to use as the mechanism the consumer used to consent” and reachable through the same medium — sign up on the website, cancel on the website. Enforcement is Attorney General only, with civil penalties up to $10,000 per violation and $25,000 for a repeat one.

Do I need a Baltimore City trader’s license to sell workbooks or test-prep kits?

Not for tutoring itself, which is a service — a trader’s license under Business Regulation §17–1807 attaches to selling goods, not sessions. It attaches the moment you keep and sell physical stock-in-trade: printed practice books, exam-prep kits, branded supplies. In Baltimore City the fee is tiered by the value of that stock-in-trade on hand: $20 up to $1,000, $40 up to $5,000, $80 up to $10,000, $160 up to $50,000, $375 up to $100,000, $1,000 up to $300,000, $1,500 up to $750,000, and $2,125 above that. A tutoring company that only ships downloadable PDFs never touches this table at all, because a digital file is not stock-in-trade; the day you keep a box of printed workbooks in a closet to sell, you do.

Can I run a tutoring business from my home in Baltimore City?

Within the same limit every other home-based business in this city runs into. Baltimore City Code Art. 32, §15–507(d) caps a home occupation at three customer visits a day and ten a week, bars deliveries shipped from the premises in a way that draws attention, and prohibits odors or nuisances detectable off the property. For a tutoring practice that means: in-home sessions are capped at a small, countable number, which is exactly the kind of ceiling a scheduling system should enforce automatically rather than let a parent discover by being turned away. It does not limit virtual sessions at all — the cap is about physical visits to the property, not about how many students you tutor online.

How much does tutoring business software cost?

For a modeled Baltimore tutoring company running 150 subscription students at $320 a month plus a modest downloadable-materials storefront, the published stack runs about $21,468 a year, or 3.45% of $622,080 in revenue. TutorCruncher’s Startup tier alone is 0.65% of revenue with no flat fee — $3,744 a year at that scale. Stripe’s standard 2.9% + 30¢ on 1,800 subscription charges a year adds $17,244. A Thinkific Basic plan for the storefront is $480 a year at the 2026 annual rate. Simpler point tools price lower on paper: TutorBird starts at $16.95 a month for one tutor plus $4.95 per additional tutor or admin; Teachworks starts at $15 a month plus 25 cents per student lesson. The percentage meters, not the list prices, are where the money actually goes at scale.

Is it worth building custom software for a tutoring or test-prep company?

It depends heavily on your size and on how much of your revenue is subscription-based, because that is where the percentage meters compound. In our modeled example, the replaceable layer — the scheduling percentage fee plus the storefront platform — is about $4,224 a year at 150 students, against which a $6,000 online store pays back in roughly 17.0 months and a $12,000 full build in about 34.1 months. At 60 students those stretch to 36.4 and 72.8 months, and we would tell you to wait. At 400 students they compress to 6.9 and 13.8. Our packages are published: a one-week Prototype Sprint is $3,500, a custom online store starts at $6,000, and a custom app, internal tool, or operations system starts at $12,000.