Tutoring Business KPIs for a Solo Tutor or Small Center

Most KPI lists for tutoring are written for centers with a front desk, a roster of tutors and someone whose job is the dashboard. They lead with tutor utilization, customer acquisition cost, conversion rate by lead source and color-coded red, yellow and green thresholds. The thresholds rarely carry a sample, a survey or a year. They are house rules, and comparing your business with an unsourced number is worse than comparing it with nothing, because it feels like information.

A solo tutor's year has a different shape. It runs on the school calendar. Students leave when the exam is done, cancellations bunch up around holidays, and prepaid packages make your bank balance look better than the sessions you have actually taught. The five numbers below are built for that shape.

Which numbers matter when you are the only tutor?

Five, each with a plain definition you can apply the same way every month.

#	Number	How to count it	What it answers
1	Active students	Students with a session in the last four weeks, or one already booked	How big is the book right now?
2	Delivery rate	Sessions delivered ÷ sessions booked, over four weeks	How much of the week you planned actually happened?
3	Sessions owed	Prepaid sessions sold minus prepaid sessions delivered	How much of your cash is still a promise?
4	Term-break retention	Students active before a break who had a session within three weeks after it	Did the calendar take students with it?
5	Hourly yield	Revenue ÷ (session hours + prep + travel + parent messages)	What does an hour of your working week actually pay?

Count students, not percentages, when the book is small. With fourteen students, one family leaving is seven percent of your business, and two in the same month reads like a crisis when it may just be a sibling pair finishing the same exam. A percentage needs a denominator large enough that one person does not swing it.

The research world agrees on the raw inputs even if it disagrees on everything else. The Stanford-based National Student Support Accelerator's recommended student data metrics for high-impact tutoring programs lists, as its minimum, the number of sessions scheduled, the number attended, and for each attended session the date, minutes present, group size and tutor. That is a school-program toolkit, not a business guide, but notice what it starts with: scheduled against attended. So does a tutoring business.

How do you count retention across a term break?

Not as a monthly churn rate. A monthly rate treats September and July as the same kind of month, and for a tutor they are nothing alike. Count retention at the three points in the school year where a tutoring relationship is most likely to end, because each one answers the question the parent was paying to have answered.

The school-year stop points, and what to count in the fortnight before each

When	Why students stop here	Count in the two weeks before
First report card of the year (a quarter or trimester in, set by the district)	A good grade tells the parent the problem is solved. A bad one sometimes tells them tutoring is not working	Which students have a report card coming, and who has not heard from you about progress in the last month
The long holiday (winter break, and above all summer)	The routine breaks. Families travel, schedules reset, and nobody rebooks by default	Who has a session booked for the first week back. Anyone without one is at risk, whatever they said at the last session
The exam is over (finals, state tests, SAT or ACT test dates, AP exams)	The goal was the exam. Once it is sat, the reason to pay is gone	Which students are tied to a single exam date, and whether there is a next goal you could honestly offer

The exam point is the easiest to put a date on. The College Board's 2027 AP exam schedule runs across two weeks in May, May 3 to 7 and May 10 to 14. If a third of your book is AP prep, a third of your book has a natural end date in mid-May, and the fortnight before it is when to have the "what next" conversation, not the week after.

The count itself is simple. Before the break, list every active student. Three weeks after the break, mark each one:

• Returned: had a session within three weeks of term restarting.
• Finished: stopped because the goal was met (the exam was sat, the grade came back where it needed to be). Write the reason next to the name at the time.
• Lapsed: stopped with no stated reason and no reply.

Term-break retention is returned ÷ (returned + lapsed). Leave the finished students out of the denominator. A student who stops because they passed is the product working, and counting them as churn punishes the outcome you were hired for. The fuller version of this idea, a repeat window set to the natural rhythm of the business rather than a calendar month, is in the post on customer retention rate for a small business.

What does a prepaid package do to your real revenue?

It moves cash forward and leaves the work behind. Selling a ten-session package puts ten sessions of money in your account on the day the parent pays, and none of those sessions has happened yet.

Here is a hypothetical September. You sell four ten-session packages at $600 each and teach eight students pay-as-you-go at $65 a session. The account looks like this:

Package cash received:        4 × $600          = $2,400
Pay-as-you-go collected:      31 sessions × $65 = $2,015
Cash in for September:                            $4,415

But the package students only had eleven sessions between them in September, at $60 each once you divide the package price by ten.

Package sessions delivered:   11 × $60 = $660
Earned in September:          $660 + $2,015 = $2,675
Sessions owed at month end:   40 − 11 = 29 sessions ($1,740)

September looked like a $4,415 month. It was a $2,675 month with $1,740 of work still to do, and that work will land in October and November, months that will look weak in the bank because the money came in early. Owners who read cash as revenue tend to feel rich in September and anxious in November, and the anxious month often has the most teaching in it.

So track sessions owed as its own line, per student and in total. It does three jobs:

1. It stops you spending a promise.
2. It tells you who is about to run out, which is the natural moment to ask about the next package.
3. It flags packages that have gone quiet. A package with six sessions left and no booking in three weeks is a lapsed student who has not told you yet.

One boundary, plainly. How prepaid income is treated for tax is a different question from how you read your own month. The IRS explains in Publication 538 on accounting periods and methods that an advance payment for services is generally reported as income in the year you receive it, with a deferral option available only under an accrual method and specific conditions. Whether any of that applies to you is a question for a bookkeeper or accountant. The sessions-owed count is a management number for your own decisions, not a filing position.

How much prep time belongs in your hourly figure?

All of it. Prep, travel, the fifteen-minute handover chat with a parent, writing a progress note and answering messages at nine at night are all hours the business consumed. An hourly figure that leaves them out is a rate card, not a yield.

Hourly yield = revenue earned ÷ (session hours + prep hours + travel hours + admin hours)

A hypothetical week: 14 sessions of one hour at $65 is $910. Prep runs about 20 minutes per session, 4.7 hours. Driving to four in-home students adds 3 hours. Messages, scheduling and progress notes take another 2.5 hours.

Session-hours version:   910 ÷ 14   = $65.00 an hour
Working-hours version:   910 ÷ 24.2 = $37.60 an hour

The rate you quote is $65. The rate the week paid was a little over half of that. That is not a reason to panic, it is a reason to look at which of the ten non-session hours can move. In-home travel and bespoke prep for every single student are usually the two biggest, and they are the two that grow quietly as the book grows.

The same arithmetic, worked at different weekly hours and with the unbillable share measured rather than guessed, is in the guide on how to calculate your hourly rate. Tutoring is one of the cleaner cases for it, because prep scales with the number of distinct students rather than the number of hours.

What does a rising late-cancellation count tell you?

First, when your calendar is. Late cancellations in tutoring bunch around the same places the stop points do: the week before a holiday, sports seasons, exam weeks for other subjects, and the week a report card comes home. A count that rose in December has probably told you about December.

Track late cancellations as a count over four weeks, with one line of reason written at the moment it happens. Late means inside whatever notice window your policy sets. Exclude sessions you moved yourself and sessions a family moved with enough notice for you to fill the slot. The measure is lost teaching hours, not politeness.

Then read the causes in order, because each is cheaper to fix than the next:

1. Reminders. Did your confirmation habit slip? A message the evening before and a message two days out behave differently.
2. The slot. Late-afternoon slots against after-school activities cancel differently from weekend mornings.
3. The student. One family accounting for half the count is a conversation with one family, not a policy change.
4. The approaching stop point. A student who cancels twice in the run-up to a break is often deciding whether to come back after it. Treat it as a retention signal, not a scheduling one.

If the count stays up after the first three, the full sorting method (measure four weeks, tag each miss by cause, match the fix to the cause, and only then think about deposits) is in the post on how to reduce appointment no-shows. Late cancellations on a prepaid package are also where your policy has to be written down before it is needed: does the session come off the balance or not? Decide once, tell every family the same thing, and log it the same way every time.

How does the list change once you add a second tutor?

It gets two new lines, and one old line changes meaning.

The new lines are the ones the center-focused lists open with, and at this point they start being real:

• Tutor pay as a share of revenue earned. Tutor pay ÷ revenue earned in the same period. Use earned revenue, not cash, for the reason above. Paying a tutor out of package cash for sessions they have not taught yet is how a two-tutor business runs short in November.
• Booked hours against offered hours, per tutor. A tutor who offered twelve hours and was booked for six is a cost you carry whether or not the hours sell. This is utilization, and it only exists once somebody else's availability is on your calendar.

The line that changes is active students. It now needs splitting by tutor, because a student who stays with the business but leaves a tutor is telling you something different from one who leaves altogether.

What does not change is the school-year timeline. Two tutors lose students at the same three points one tutor does. The difference is that now somebody has to own the fortnight-before list for each tutor's book, and at two people that somebody is still you.

Where this lives in SMBDashboard

Everything above comes out of two records, which is why SMBDashboard is shaped the way it is: money, customers, tasks and appointments on one screen, no account to create, and your data stays in your browser unless you turn on Pro sync.

For a tutor, that looks like this. Each student is a customer. Each session is an appointment linked to that student in the appointments module, with the note field carrying the package count ("pkg 4 of 10") or the reason for a late cancel. Each payment is a money entry marked paid or open, so a package sold on credit or an unpaid pay-as-you-go session stays in the open list until it clears, and the student's record shows both their sessions and their payments together. The free tier holds 25 customers, 200 money entries and unlimited tasks, which covers a real solo book. Pro removes the caps and adds CSV export, recurring entries and your own branding on the printed report, at $48 every six months (about $8 a month) or $149 once. The general Monday version of this routine is in the numbers to look at every Monday.

Common questions about tutoring business KPIs

How do I track tutoring sessions and payments?

Two lists, linked by the student's name. The appointment list holds every session booked, with a mark when it was delivered or cancelled. The payment log holds every amount received or owed, marked paid or open. Everything in this post, from delivery rate to sessions owed, is a count across those two lists. A spreadsheet does it. A tool that links a session and a payment to the same student saves you the cross-referencing.

How do I account for prepaid tutoring packages?

For your own management numbers, treat the package price as earned one session at a time: package price ÷ sessions in the package, recognized as each session is taught. Keep a running sessions-owed count per student. For tax treatment, which can differ from how you read your own month, check IRS Publication 538 and speak to a bookkeeper or accountant before you decide anything.

How do I work out how many students I need?

Work backward from your hourly yield, not your session rate. Take the income you need in a month, divide by your hourly yield to get working hours, then divide working hours by the hours one student actually costs you per month (sessions plus their share of prep, travel and messages). Then add a margin for the stop points, because the book you have in October is not the book you will have after the summer. The answer is a number of students you can check against your own calendar, not a figure from someone else's business.

Should prep time count when I work out my hourly rate?

Yes. Prep is work the business requires for every session, so it belongs in the hours the revenue is divided by. Leave it out and you have measured your price, not your pay. The honest version usually lands well below the quoted rate, and the gap is the most useful thing it shows you.

Set up next week's sessions so the calendar carries the whole book, then mark each payment paid or open as it lands. Open the appointments module and book the week.