TEACHERS
WHO
STAY.
Former CNN journalist. Stanford LBAN graduate. Co-author of the Innovation Advantage™ curriculum with the University of Central Florida and Harvard's Project Zero. Founder of Young Innovators Academy, three Florida campuses — 117 teachers across 11 years.
"I have hired, lost, and re-hired the same kinds of teachers enough times to be honest about the pattern."
I have written six books for the small-school operator. This is the one I have been the most reluctant to write, because to write it honestly I have to be honest about myself — about the years I micromanaged my lead teachers and could not understand why they left, about the times I hired the credential and lost the human, about the Sunday-night Brightwheel messages I sent because I could not let go.
I am writing this book anyway, because the operators I work with now are doing the same things I did, and they deserve a clean version of the playbook I had to assemble in pieces over a decade. Teacher turnover in a small school is almost never a labor market problem. It is an owner-behavior problem disguised as a labor market problem. That is the uncomfortable sentence. It is also the most actionable one in this book.
What follows is the way I now hire — The Classroom Audition — and the way I now keep teachers once they're in. The compensation bands. The first-30-days framework. The exit-interview question that changes everything. And, candidly, the ten things owners do that drive turnover without realizing they're doing them. I am the owner in most of those examples first.
Read it twice. Forgive yourself for the first read. Then change something for the second read.
The book is sequenced the way a working operator should rebuild their hiring and retention practice. If you have an immediate hiring need, jump to Chapter 4 · The Classroom Audition this week. If you have a new lead teacher about to start, the most important page in the whole book for you is Chapter 5. If your team is fully staffed and stable, read in order — Chapters 7 and 9 will do more work on you than you expect.
The 10 owner behaviors in the Interlude (p. 22). Read first, even though it's late in the book. You are one or two of them.
Run The Classroom Audition (Ch 4) on your next candidate. Score with the rubric. Compare to your gut. Notice the gap.
The 30-day model in Ch 5. Shadow → co-lead → reverse-shadow → solo. Counterintuitive at first; data-backed every time.
Comp bands, Freedom-With-Structure, the exit-interview question, and the referral flywheel. Compound over years.
You run a small school — somewhere between 30 and 250 children, one to three campuses, fewer than 25 lead teachers in total. You are losing 1–2 lead teachers a year more than you'd like. You have tried raising wages and it has not solved the problem.
If you run a 600-child chain, some of this applies but the comp math is different. If you are a national operator with HR business partners, you have systems this book does not address. If you are a solo founder with three teachers, you are exactly who this is written for.
When a teacher gives notice at a small school, the conversation that follows in the owner's head almost always starts in the wrong place. "The labor market is brutal right now." "She had a better offer." "We just can't compete with the chains." All of these can be partially true. None of them is the place to begin if you actually want to fix the problem.
The place to begin is a quieter question, and a more difficult one: What was the experience of working at this school that made a good teacher conclude that another job was better? The answer to that question is almost never primarily about money, and is almost always partially about owner behavior. I know this because I have lost teachers I should have kept — and when I have done the honest post-mortem, the deciding factor was almost always something I did, or did not do, in the previous 90 days.
This reframe matters because it changes what you do with the budget you have. If turnover is a labor market problem, the move is to raise wages, run a recruiter, and hope. If turnover is partially an owner-behavior problem, the move is to change three or four operating habits — none of which costs money — and then the wage-and-recruiter strategy starts to actually work.
That is what this book is for. The wages still matter, and we will get to them in Chapter 6. But the wages alone will not retain a great teacher whose owner overrides her in front of children every other Tuesday.
Four numbers, memorized, change how you make every hiring and retention decision for the rest of the year. Most operators have never written any of them down for their own school.
The dollar spent on recruiting a new teacher produces one teacher. The dollar spent on retaining a current one keeps producing — through that teacher's classroom culture, through the parents she'll refer, through the candidates she'll vouch for, through the new hire she'll onboard for you next year. A retention dollar is a recruiting dollar two years from now, and a curriculum dollar four years from now, and an enrollment dollar continuously.
The numbers that change when you retain instead of replace are not subtle. Parent retention climbs because parents are not adapting to a new teacher every year. Word-of-mouth referral volume climbs because long-tenure teachers refer their best friends — and only their best friends. Time-to-fill drops because half your hires are now referrals. Onboarding time drops because the senior teachers who do the onboarding are the same senior teachers year after year.
There is a benefit nobody puts on the slide. The owner of a school with stable lead teachers gets her Sunday nights back. She does not write the seven Brightwheel messages with "small fixes for Monday." She does not redo the lesson plan. She does not interview on Saturday morning. The single largest underestimated benefit of retention is its effect on the operator's own quality of life, which then loops back into making her a less anxious, less controlling boss — which retains the next round of teachers.
The cost model most operators use to think about teacher turnover counts the wrong things. It counts the recruiter fee, the Indeed-sponsored post, the printing of new training binders. What it leaves out — and what is usually the largest line item by an order of magnitude — is what happens to the families on the other side of the classroom door.
When a lead teacher leaves a small pre-K classroom, somewhere between one in four and one in three families with a child in her room will be gone from the school within 60 days. They will not tell you that the teacher leaving is the reason. They will say "we're moving," or "his grandparents have offered to take him three days a week," or "we wanted to try something different." Those are the polite reasons. The actual reason, in the conversations they will have with each other but never with you, is that the relationship they trusted is gone, and they are not waiting to see whether the replacement is as good.
The implication is uncomfortable. The implication is that a $2,000 retention bonus to keep a great teacher who is on the fence costs less than what you will lose, in tuition, from one disenrolled family in the first six months following her departure. That math is not subtle. Most operators run it at the wrong level of resolution and conclude they cannot afford the bonus. They can. They cannot afford the alternative.
What one departure actually costs, line by line. Write your school's version of this on a single page. Tape it to the office wall.
Run this for your own school. Adjust the tuition figure to your market. The point of the exercise is not the precise number — it's the rounding. A teacher departure costs you between $25,000 and $45,000 in a small school. Once that number is on the wall, every retention decision you make for the rest of the year is calibrated correctly.
Theresa, owner-director, 84-seat pre-K in Sarasota, FL. Three lead-teacher departures over an 18-month period.
Wrote the replacement math on the wall in the office. $31K per departure × 3 = $93,000 in 18 months. Reallocated $14,000 of that anticipated cost into a $2,000 annual retention bonus for the remaining four lead teachers, plus $1,500 each for new comp-band rollouts (Ch 6) and $500 each for PD (Ch 8).
Zero lead-teacher departures in the following 12 months. Parent attrition rate cut roughly in half. Theresa's anxiety about staffing, by her own report, "gone." The investment was 15% of what the alternative would have cost.
The credential-first hiring model is comfortable because it is measurable. CDA, AA in early childhood, years of classroom experience — these check boxes. They also, in my eleven years of hiring, predict almost nothing about whether a teacher will be great with four-year-olds, will stay three years, or will be the one parents request by name on the second tour.
Every great teacher I have ever hired — the ones with seven, eight, eleven years in our rooms — has shared three traits. None of these traits is a credential. None of them appears on a resume. All three can be screened for in 25 minutes if you know what to look for.
1. Asks "why" 5× more than she says "no." A teacher whose default move when a child does something unexpected is curiosity will run a different classroom than a teacher whose default is correction. You can hear this in their answer to a single interview question: "Tell me about a time a child did something you didn't expect. What did you do?" Listen for whether the teacher's first action was a question or a correction.
2. Keeps a physical or digital notebook of observations. Not for the curriculum, not for the report. For herself. The teachers who keep stayed in our rooms longest all had some version of this — a Moleskine, a Notes app, a folder of voice memos. Ask: "What's something a child has done in the last two weeks that you wrote down?" If the answer comes fast and specific, you're talking to a teacher.
3. Can name three things she changed about her practice in the last six months. Specific changes. Why she made them. What happened. The teachers who answer this well are growing on their own — they will continue to grow under any director, including a mediocre one. The teachers who can't are reliant on you to make them grow. That is exhausting on both sides.
I have hired teachers with CDA credentials and 10 years of experience who could not answer any of these three questions well. I have hired teachers six months out of college, with only their student teaching, who answered all three beautifully. The first group, on average, lasted 14 months. The second group, on average, lasted four years. The trait predicts. The credential does not.
Five stages. A 45-minute observed lesson with real children at the heart of it. The single highest-leverage change you can make to your hiring practice this year.
The 18% pass-through at Stage 3 is the right number. If your audition is passing 50% of candidates, you are running a tour, not an audition. If it is passing 5%, you are scoring perfection instead of trajectory. Adjust.
The 45 minutes of the audition are divided into four observable segments. Score each on a 0–1–2 scale (not present · partial · clearly present). Total possible: 8 points. Hire above 5. Below 5, do not hire even if you like the candidate — your gut is being overruled here by the children.
What to score: Does the candidate greet children before launching the activity? Does she crouch to their height? Does she learn one or two names before starting?
Why it matters: A teacher who launches without connecting is a teacher who will run the rest of her year that way.
What to score: When a child asks an off-plan question, does the candidate follow the question or push back to the plan?
Why it matters: This is the single most predictive 60 seconds of the entire audition. The teachers who follow the question stay longer and produce more memorable classrooms.
What to score: Does she extend the lesson based on what the children just did? Or does she revert to a generic "free play"?
Why it matters: Extension is the move that signals she sees children as collaborators, not audience.
What to score: Does she involve children in cleanup, or do it for them? Does she close the loop on what happened, or just exit?
Why it matters: Cleanup is where the strongest teachers do their best work. It's also where ordinary candidates reveal themselves.
Forty-seven percent of small-school teacher departures happen in the first 90 days. The first 30 of those 90 are where the outcome gets decided. Most schools throw the new hire into the room on day one and assume she will adapt. Most of the ones who leave at day 65 had concluded by day 12 that they were not going to stay.
The model below runs counter to the instinct of every small-school owner I have ever met, including me ten years ago. The instinct is "we are short-staffed; we need her teaching by Wednesday." The data is that the schools that let the new hire shadow for the first week — fully, without leading — see 90-day retention move from roughly 70% to roughly 92%. That is not a marginal improvement. That is, on the replacement math, $14,000 of saved cost per hire.
Week 1 · Trust week. Shadow only. No leading. The new hire learns names, routines, where the supplies are, and — most importantly — sees a senior teacher run a classroom well. She is not under pressure. She is learning the system.
Week 2 · Apprenticeship week. Co-lead one activity per day with a senior teacher. The senior leads; the new hire supports. Reverse during one activity per day by the end of the week.
Week 3 · Reverse-shadow week. New hire leads activities. Senior teacher observes silently, in the corner. End-of-day debrief: 10 minutes. What worked, what to adjust tomorrow.
Week 4 · Solo week, with daily check-in. Full lead. 5-minute end-of-day check-in with the director, every day. By Friday, the check-in becomes a weekly cadence.
The objection: "We can't afford to have a new hire shadowing for a full week. We need her in the rotation." The answer: You can afford it. The math is on the previous page. A failed hire at day 65 costs you $31,000. A shadow week costs you, at most, two days of overlapping wages — about $400 — and reduces your failure probability by roughly half. That is a 70× return.
Print this. Hand the new hire a copy on her first day. The transparency itself is one of the retention moves — she knows exactly what is expected of her in each phase, which removes the anxiety that drives most early departures.
1. Does she initiate questions? The new hires who stay are the ones who, by day 8, are asking the senior teachers specific questions about why a routine is the way it is. The ones who leave by day 60 tend to nod through week 1 and never ask anything.
2. Does she remember the children's names by Friday of week 1? Not a high bar. The ones who don't are usually not going to invest in the relationships, period.
3. Does she modify her plan based on what week 1 showed her? Her first solo activity in week 3 should be visibly different from what she would have planned in week 0. If it's the same, she's not absorbing.
4. Does she initiate the end-of-day check-in or wait for it? By the end of week 4, the keepers are coming to you. The leavers are waiting to be called.
Owner-operators of small schools cannot match the starting wages of regional daycare chains. The chains compete on signing bonuses and starting rate. That is the wrong field to fight on. Small schools win on the other four things — clarity, growth, smaller ratios, real say — and they need a wage structure that is honest, published, and predictable. Discretion as a comp philosophy ("we give raises at our discretion") drives more turnover at small schools than low wages do.
The band example below is calibrated for a Central Florida small school of about 80 children in 2026. Adapt to your market. The structure is what matters more than the precise numbers.
1. Published, not discretionary. Every teacher knows what the next step looks like and what it pays. Discretion looks like favoritism even when it isn't, and favoritism is one of the top three reasons small-school teachers leave.
2. Tied to observable milestones, not to anniversary dates. Year-2 raise contingent on a positive performance review. Y3 contingent on the CDA. Y5 contingent on a defined promotion path. Time alone is not the trigger.
Almost every small-school owner I have ever met — including me — used to be a lead teacher. We know what a great room looks like because we have run great rooms. This is the thing that makes us good at this work. It is also the thing that, if we are not careful, makes us bad at retaining our teachers.
The temptation is constant. The dramatic play area was set up wrong this morning, and you cannot help yourself. You rearrange it. The lesson plan that came in on Sunday night is going to lose the kids by 10 a.m., and you cannot help yourself. You send three Brightwheel messages with "small tweaks." The daily report a parent received last night used the word "okay" twice, and you cannot help yourself. You redo it before pickup. None of these things, by themselves, is a fireable offense. The compounding effect over six months is what drives your best teachers out the door.
The director's job is to set the structure and audit weekly. It is not to redesign the activity on Monday morning. The minute I stopped redesigning my teachers' Monday mornings, my best teachers stopped leaving. I lost two years of retention before I figured this out."
The hardest version of this for me was the Brightwheel messages on Sunday night. I would see a lesson plan come in and notice three small things I would have done differently. I would type the suggestions. I would tell myself it was helpful, that good teachers want feedback. I would hit send.
What I did not notice for years was that the Sunday-night feedback landed on my teachers as a quiet weekly reminder that their Monday morning was being evaluated before it started. The strongest teachers — the ones with options elsewhere — were the ones who picked up on this fastest. Those were the teachers I lost.
The rule I now follow: weekly walkthrough on Friday afternoon, in person, kindly, specific. No Sunday-night Brightwheel. No Monday-morning rearranging. If something needs to change, it gets raised once, on Friday, and the teacher decides how to implement it.
Almost every small school has a tiny PD budget — somewhere between $300 and $600 per teacher per year. Most operators deploy this on one regional conference and call it a year. That is a wasted budget. The five-line allocation below stretches the same dollars into a development practice that is one of the strongest retention forces a small school has access to.
The two free ones do the most work.
Innovation Friday is a 30-minute slot every Friday afternoon where one teacher shares one new thing she tried that week. Rotates through your whole team in 6–8 weeks. Three benefits compound: teachers learn from each other; every teacher gets recognized publicly four times a year; the act of preparing the share itself is the highest-quality PD a teacher does all month.
Cross-classroom observation weeks happen twice a year. Each teacher spends two mornings observing in a peer's classroom. Comes back with notes. The notes are the basis of the next month's Innovation Fridays. Free, and one of the highest-retention practices on this list.
A $500/year PD budget per teacher, deployed this way, runs an annual cost of about $4,000 across a small school of eight teachers. The replacement math says one prevented departure pays for the entire program for seven years. The Innovation Friday slot alone has been cited by three of my long-tenure teachers as the reason they stayed when they had higher-paying offers elsewhere.
Every school does exit interviews. Most ask "Why are you leaving?" and receive a polite, unhelpful answer. "Looking for new opportunities." "Family reasons." "I needed a change." These answers are not lies. They are also not actionable, which is why the exit interview becomes a ritual instead of a learning instrument.
There is a single question that consistently produces specific, useful, surprising answers. I have been asking it for six years. I have never had a teacher dodge it.
The shift is from past blame to future possibility. "Why are you leaving" asks the teacher to indict the school, which she will refuse to do out of professional courtesy. "What would have been different to keep you" asks her to imagine a version of the school she could have stayed at, which she will tell you about because she still cares — and because the question is now about her imagined future, not about your failures.
The answers I have collected over six years cluster around four themes, almost regardless of school or market. Compensation appears, but it is rarely the top theme. Recognition appears more often than money. Autonomy appears most often. And the fourth, which surprised me the first time and has never stopped surprising me, is predictability of schedule. Teachers will accept lower wages for a more predictable schedule. They will not stay at a school where the schedule is reactive even at a higher wage.
The retention flywheel. Once you have two or three lifers, your recruiting problem inverts.
Referrals from senior teachers are the single highest-quality recruiting channel a small school has access to. In my data across three campuses, referral hires have 2.3× the average tenure of Indeed hires, screen through the audition at ~55% versus the platform's ~18%, and cost the school somewhere between $0 and $500 each compared to roughly $1,200 for a sponsored Indeed post that produces a hire.
The mechanism is straightforward: long-tenure teachers will only refer people whose work they trust, because those new hires will become their classroom neighbors for the next several years. They have skin in the referral. They will not refer someone who will embarrass them.
1. Ask explicitly. "We are hiring an assistant teacher for the threes room. You know teachers I don't. Who comes to mind?" Asked at the team meeting, every time you have an open role. Not implied. Asked.
2. Pay $500, structured. $250 at hire, $250 at the new hire's six-month mark. The delayed half matters — it signals you take retention seriously, and it gives the referring teacher a reason to mentor through the first six months.
3. Let the senior teacher do the first phone screen. She knows what to ask better than you do. She will ask questions you would not. The candidates she advances are the ones she would be proud to sit next to.
4. Make the referring teacher visible at the hire announcement. "Brooke joined us this week, referred by Ms. Carolina." Naming the referrer publicly compounds the social capital and produces more referrals.
The referral flywheel does not exist until you have two or three teachers with 3+ year tenure. Below that threshold, your senior teachers do not have enough confidence in the school to put their reputations on the line. Above it, the flywheel becomes the dominant channel within 12–18 months. The investment in retention (Chapters 6–9) is what unlocks the referral channel. They are not separate problems.
I have done at least seven of these. Probably eight. The point is not blame. The point is that these are the things to stop, in order of damage.
The single most corrosive owner behavior in a small school. The damage compounds invisibly. Wait. Talk privately. Always.
Every senior teacher on your team registers this as a signal about how seriously you take the work. They are right.
A delayed raise without communication is a broken promise. Communicate. Even "I cannot give it to you until October" is dramatically better than silence.
"We do this whole STEM thing here" — said while standing in Ms. Brooke's room, where Ms. Brooke is the one who built the program. She knows you did this. She tells two other teachers at lunch.
Your teachers have kids they need to pick up too. The meeting that ends at 5:15 on Friday is the meeting that loses you the teacher with a second-grader.
You are signaling that Monday is being evaluated before it starts. The strongest teachers — the ones with options — pick up on this fastest.
The teacher reads this as: she matters less than the parent emergencies, the leaky sink, the inspector visit. She is partly right.
Even a small correction in front of a parent ("we usually do it this way") is a public undercut. Save it for the Friday walkthrough.
The new hire shows up Monday morning to a senior teacher who has not been consulted about her. That dynamic is corrosive on day one.
If your teachers cannot eat lunch in 18 quiet minutes, you have a retention problem disguised as a space problem. The room matters more than you think.
Six numbers. One sheet. Reviewed once a month at the senior leadership meeting — not weekly. Retention is a slow-moving signal; reviewing it weekly creates anxiety without information. Monthly is the right cadence.
The single most predictive metric in the dashboard is #4 — referral hire rate. When it climbs above 40%, retention is working downstream. When it falls below 20%, your senior teachers no longer trust the school enough to refer their friends. That is a signal to read Chapters 7 and 9 again, honestly.
If reading this book surfaced more about your own pattern than you were expecting, you are not alone — that is the most common feedback I get on Chapters 7 and the Interlude. Three tiers of engagement, all designed around the same hiring and retention sequence in this book. Tier 2 is the recommended starting point for most operators.
The Classroom Audition built and trained for your school. Comp band review. One on-site working session. Designed as a clean off-ramp if Tier 2 isn't right yet.
Tier 1, plus the onboarding model rolled out with your senior team. Comp bands published. Innovation Friday installed. Three on-site visits + weekly working calls. Includes one full hiring cycle run together.
Tiers 1 + 2, plus the full referral flywheel, dashboard implemented, monthly board-level reporting, exit interviews redesigned, and a senior-leader coaching arc on the owner-behavior patterns in Chapter 7 & the Interlude.
Engagements start with a 25-minute free audit conversation. We talk through your last three departures and the practices in this book. If it's a fit, we go. If it's not, you walk away with a one-page action plan.
Six books with Gryphon House, written across eleven years of running pre-K classrooms. The companion to Teachers Who Stay is The Basics of Leading a Child-Care Business — the operational philosophy in this playbook is its direct extension.
Leading is where the Freedom-With-Structure principle was first written out in long form. Starting covers the financial side. Together they are the closest things to a small-school MBA you can put on a single shelf.
A weekly podcast for school founders, directors, and the curious operators trying to do this work better. Episodes drop every Tuesday.
Listen on: Apple · Spotify · YouTube
Recommended episodes for this book: the Freedom-With-Structure conversation (Ep. 38), exit interviews redesigned (Ep. 51), and the Classroom Audition rollout with three working directors (Ep. 56).
The three Young Innovators Academy campuses host quarterly Operator Days. Half a morning in classrooms, a working session with our leadership team on hiring and retention practices, lunch, and a Q&A.
Campuses: Winter Garden, Winter Park, Oviedo (Central Florida).
The third one is the most important. Stay interviews are more valuable than exit interviews, and almost nobody does them.
Curated for the hiring and retention side specifically. The HR-and-PD tools section is unique to this playbook.
Free. No pitch unless you ask for one. We'll walk through your last three departures honestly, identify the two or three owner-behavior patterns most likely at play, and you'll leave with a one-page change plan whether you hire me or not.
BOOK THE CALL