The short answer
The 1,000-Day Framework is Bill Canady’s operating method for a private-equity-backed CEO. A hold runs five to seven years, but the window in which a CEO can still change the outcome is roughly the first 1,000 days — about two years and nine months. After that you are no longer building value. You are presenting it, because the numbers you will be sold on have already been made.
Those 1,000 days are not one job. They are four different jobs done in sequence, and the skills that win the first will lose you the third:
- Days 1–100 — Take control. Deliverable: command. By Day 100 anyone in the building can state the goal, the strategy, and who owns what. Verbatim, not approximately.
- Days 101–365 — Earn the right to grow. Deliverable: margin, EBITDA and cash visibly turned by Day 365. Not a plan to turn them. Turned.
- Days 366–730 — Scale what matters. Deliverable: a flywheel — growth that continues while the CEO is on a plane.
- Days 731–1,000 — Always be exiting. Deliverable: the premium — the multiple, worked on purpose.
The stakes are arithmetic. Roughly 70% of PE-backed CEOs are replaced during the hold, and the replacements cluster hard between months 18 and 24 — the moment the second phase comes due and the bill arrives for a year of expansion that got mistaken for growth. Meanwhile a deal underwritten today needs 10–12% annual EBITDA growth to return what the sponsor promised, against roughly 5% a decade ago. Twelve is the new five, and multiple expansion is no longer showing up to bail anyone out.
One rule governs the whole structure: you cannot skip a phase, you can only defer it, and deferred phases get more expensive every quarter.
Why does a 1,000-day clock exist at all?
When a sponsor buys a company, they buy it with an exit already modeled and dated. The model has a date on it — call it five years, sometimes six. Holds have stretched; they are the longest they have been in about twenty-five years, and more than half of PE-backed companies are now held past four years. But the stretch does not extend your working window. It extends the waiting.
The part of the hold where a CEO can materially change the outcome is the front of it. Roughly a thousand days. Past that point the diligence is running, the story is being packaged, and the operating decisions that will show up in the sale price have already been made and have already had time to compound. Nobody has ever fixed a margin structure in a data room.
So the clock is not a motivational device and it is not marketing. It is arithmetic about when your decisions still have time to work. A decision made on Day 200 has eight hundred days to compound into the number a buyer underwrites. The same decision made on Day 800 has two hundred, and buyers price a two-hundred-day track record exactly as thinly as it deserves.
This is also why the failure clusters where it does. The replacements are not spread evenly across five or six years. They bunch between month eighteen and month twenty-four, because that is the moment the second phase comes due — when a year of top-line work has to show up as EBITDA and doesn’t. The board stops asking how it is going and starts asking what the plan is. Two calls later there is a co-pilot nobody asked for.
Why is a hold four different jobs instead of one?
Here is what almost nobody says out loud when a CEO takes a PE-backed seat: the thousand days are four completely different jobs done in sequence, and the skills that win you the first one will actively lose you the third.
That is the trap, and it explains most replacements. The CEO hired into this seat is usually hired for one quality: they are a closer. Decisive, commercial, able to go get revenue. So on Day 1 they do the thing they are good at, and it works. Quarters three and four look strong. The board is pleased.
Underneath it, the business is getting more complicated every week. More customers, more SKUs, more one-off pricing, more exceptions and configurations, more small orders from small accounts that somebody in operations is expediting on a Tuesday afternoon. None of it shows up as a line item, because there is no expense category called “too much stuff.” So it hides — in capacity, in cycle time, in the 14% discount somebody granted in 2019 that nobody ever revisited.
Then around month fifteen the growth curve flattens, because the business ran out of capacity rather than demand. The margin curve, drifting down the whole time, finally crosses in front of the revenue curve where everyone can see it. Revenue up eleven percent, two record months, EBITDA flat.
That CEO did not fail through laziness. He worked harder than anyone in the building. He spent nineteen months making the business bigger without making it better, because he ran the Phase One playbook for four phases straight. Each phase has exactly one job, exactly one deliverable, and one deadline — and if you do not finish the deliverable, you do not get to start the next phase. You just carry the debt forward and pay for it later at a much worse rate.
What has to be true by Day 100?
Days 1–100. Take control. The job is command — not strategy, command. The deliverable is that by Day 100 every person in the building can state the goal, the strategy, and who owns what. Not approximately. Verbatim.
The common mistake is defensible and expensive. A thoughtful new CEO says, correctly, “I don’t know enough yet — I’ll spend ninety days listening.” So they run a listening tour. Every plant. A hundred and twenty one-on-ones. It is good leadership behavior and it is the wrong use of the only clean pass you will ever get. Because whatever you intend, a listening tour communicates one thing: nothing has changed yet. And the organization is watching closely for whether something has changed, because until they see it, every person runs the old playbook while telling you in the one-on-one how excited they are about the new direction.
On Day 101 you are no longer the new CEO. You are just the CEO, and everything you did not change in the first hundred days is now something you are endorsing.
Phase One is four meetings, and the sequence matters as much as the content. Meeting One — Get a Goal (weeks one and two): one measurable goal for the enterprise, one sentence. Meeting Two — Set the Strategy (weeks three through six): where the company will and will not compete to hit that number, because strategy without a number attached to it is a preference. Meeting Three — Reorganize the Company (weeks six through nine): structure follows strategy. Meeting Four — Take Action (weeks nine through twelve, then forever): the cadence starts.
The artifact of Meeting One is one sentence with four load-bearing parts — a number, a unit, a date, and a denominator. Improve EBITDA margin by three hundred basis points in twelve months. The denominator is what stops the goal being gamed: “grow revenue fifteen percent” can always be won by spending, and you will hit it and be poorer.
Failure mode: the laundry list. Twenty-five initiatives. A beautiful, comprehensive deck in which every function is represented so nobody’s feelings were hurt — and which is functionally identical to having no plan. A leadership team can carry about three genuine changes at once on top of running the company. Hand eight people twenty-five items and within six weeks the organization sorts the list for you, down to the six that are comfortable rather than the six worth the most.
Learn it in: The Clock and Get a Goal.
What does “earn the right to grow” actually mean?
Days 101–365. The job changes completely, and this is where the closer starts to struggle. The job is not growth. It is becoming the kind of business that can grow profitably — and only then growing. The deliverable is margin, EBITDA and cash visibly turned by Day 365.
Most CEOs try to grow before they have earned it. Grow a business that has not earned it and you do not get growth — you get expansion. Bigger, more complicated and less profitable at the same time. You have scaled the dysfunction. The distinction is mechanical, not moral: growth is adding a dollar of revenue and keeping a larger share of it than last year; expansion is adding a dollar and keeping a smaller share. They look identical on the top line and diverge on the bottom line about four quarters later — exactly long enough to have hired against the wrong one.
There is a number that tells you which one you are about to do. The Right-to-Grow Ratio: material margin divided by total fully-loaded employee cost. Two lines off your own P&L, about thirty minutes, one verdict. Below 2.5 you have not earned it. 2.5 to 3.5 is the healthy middle-market band. 4.0 is exceptional. Calibrate the level for your industry’s capital intensity, then watch direction of travel harder than level — a company that moved from 1.4 to 1.7 in three quarters is doing better work than one sitting flat at 2.8.
Phase Two is the longest phase and it carries five of the ten episodes, because it contains the actual engine work:
- Segment the business. The top 20–25% of customers produce 105–150% of the profit — which is only possible if the rest produces negative profit. The bottom 20% account for roughly 5% of revenue, 30% of capacity and 60% of complaints. None of it is visible on a P&L, because a P&L is an average and profit lives at the intersection of a specific customer buying a specific product.
- Retire the complexity. Quad 4 — B-customers buying tail products — is about 4% of revenue and 20–40% of capacity. Audit the Dirty Dozen, size each in dollars, take the top three by dollars rather than by drama.
- Fix price first. On a $150 million business, one point of price is $1.5 million, and it flows through at roughly 100%. Buying that same $1.5 million out of volume takes $5 million of new revenue at a 30% incremental margin; out of cost it takes 5% of overhead and a year before it stops bleeding back. Order the bridge price, mix, share gain, M&A, cost — cost last, because you cannot know which costs are bad until you know which revenue is good.
- Rebuild the budget from zero. Last year plus five percent is how you lock in every bad decision you have ever made.
Failure mode: flinching. You build the analysis, you see the tail clearly, and somebody says the sentence — “but that’s still four million dollars of revenue.” Flinch there and the whole phase is decoration. The four percent is not the cost. The four percent is the bait.
Learn it in: The 80/20 Engine, Earn the Right to Grow, The Complexity Tax, Price First, Cost Last and Zero-Up.
Why does Phase Three end more tenures than any other?
Days 366–730. Scale what matters. Now you have earned it. Now you grow. And the job changes again — this time in a way personal effort cannot solve.
Phases One and Two can be pushed through on personal force. You can establish command by sheer presence. You can drive a simplification agenda by being the most stubborn person in the room. Phase Three cannot be done that way. Phase Three requires the business to run without you in the room — and if the only functioning system in the company is your personal attention, you have hit your ceiling. The ceiling is not the market. It is you.
The deliverable is a flywheel by Day 730: growth that continues while you are on a plane, a leadership team where each person owns a number and presents it themselves, and an operating cadence — weekly, monthly, quarterly — that runs whether or not you feel energetic.
Two things build it. The first is architecture: the Rule of Three. Direction (the Visionary — where are we going, what are we not doing), Execution (the Operator — by when, and then who), and Truth (the Prophet — the one person empowered to use data to kill sacred cows and complexity). The Prophet seat is empty in roughly nine out of ten companies, and you cannot delegate to an empty chair. Around 87% of large companies hit a major growth stall and fewer than one in ten fully recover — and the causes are overwhelmingly controllable, made inside the building.
The second is rhythm: the Command Center. Meetings fill a calendar; a cadence runs a company. The weekly (45–60 minutes, one page, five to ten numbers). The monthly business review (owners present their own numbers; variances get causes and countermeasures, not adjectives). The quarterly reset (kill it, keep it, or double down — focus is subtraction). The whole system costs about half a day of a CEO’s week, roughly eight percent of the calendar, and you are already spending that half day in fragments on interruptions. The cadence does not add time; it converts time you are already losing from reaction into governance.
Failure mode: the hero CEO. If you feel like the Chief Everything Officer, that is usually not micromanagement. Most CEOs who feel that way are the opposite — they hand out authority freely and are irritated that everything comes back anyway. They are carrying the load because the architecture was never designed to carry it without them. That is a blueprint problem, not a character flaw, and you cannot fix a blueprint by trying harder. It is fixable in Phase Three. It is not fixable in Phase Four, when the buyer’s advisor asks who else can run this.
Learn it in: The Team Lock and The Cadence.
How do you build the premium instead of hoping for it?
Days 731–1,000. Always be exiting.
Enterprise value is EBITDA times a multiple. Every operator works the first term. Almost nobody works the second one on purpose — and the second one is where the money is. A turn of multiple on twenty-eight million dollars of EBITDA is twenty-eight million dollars, and you know exactly how hard you had to work for the last two points of margin.
Two businesses, same industry, both sold inside the same eighteen months, both putting up about twenty-eight million of EBITDA in the final year. One cleared seven times. The other cleared ten and a half. Ninety-eight million dollars of difference. The seven-times business had one customer at thirty-one percent of revenue, a forecast that had missed by double digits four quarters running, a third of its EBITDA in add-backs only the CFO could explain — and four seconds of silence when the buyer’s advisor asked who runs this if the CEO leaves.
The multiple is not handed down by the market. It is substantially a function of how risky the business looks to somebody who has to write a check and then live with what they bought. Seven things move it, and every one is inside your control: customer concentration; owner and CEO dependency; forecast accuracy; management team durability post-sale; quality of earnings and how much of EBITDA is add-backs; revenue predictability and contract structure; and data-room readiness. Score yourself one to five on each — that is the rubric, and it is the Phase Four work order.
Enterprise value has four levers, and pulled in order they multiply rather than add: revenue quality, margin, capital efficiency, and the multiple. Lever four multiplies the value of the other three, which is why owner dependency is the single most expensive problem in a middle-market company.
Failure mode: treating the business like your baby. Inside this structure the company is a product — designed, improved, packaged and sold at the right moment in the best possible condition. Built to last is not the required condition here; such companies are intended to go from good to gone, and gone is the design, not the failure. Operators who make peace with that early make materially more money than those who make peace with it in the data room, four days after the first offer comes in low.
Learn it in: Always Be Exiting.
Why can’t the four phases be rearranged?
Because each phase is the precondition for the next one, and the dependencies are mechanical rather than stylistic.
You cannot do Phase Two without Phase One, because the Phase Two work is unpopular. Freezing discounts, exiting a tail, re-pricing a distributor, killing a product line — every one of those decisions gets tested within days, usually by your largest account, usually before your own team has finished reading the memo. An organization that cannot state the goal will not hold a line it does not understand. Command is what makes simplification survive contact with Monday morning.
You cannot do Phase Three without Phase Two, because scaling an unearned business is expansion by another name. Growth initiatives land on top of a leadership team that is already fully occupied, and occupied is not available. Capacity has to be released by a decision before it can be reallocated — somebody has to raise a price, set a minimum order value, move an account to a distributor, or say no. Free first, then reallocate, in that order, every time. Skip it and six months later the team is doing everything it was doing before plus a growth agenda, and the CEO cannot understand why a team this strong is moving this slowly.
And you cannot do Phase Four without Phase Three, because the multiple only re-rates once the business has genuinely de-risked. Margin and cash move first because they are the fastest levers you own. Everything that lifts the multiple is a form of track record — eight quarters of forecast accuracy, a bench with named successors in development, add-backs converted into realized run-rate results — and track record takes the two years you spent building the team and the cadence. You cannot sprint it in the final ninety days. Everyone tries. Buyers can smell a business that was tidied up for the sale and they price it accordingly, every single time.
How do I find out which phase I’m actually in?
Three moves. All three can be run in a weekend; the first takes about four minutes.
Move one — locate yourself on the clock. Get the date the deal closed, not the date you started. Count the days to today. Write the number down, then write which phase it puts you in.
Expect discomfort, because the reaction is close to universal: people are further along than they feel. Somebody who feels newly started turns out to be on Day 380 — which means Phase One is over, and whatever command was established is the command they have. You cannot manage a clock you refuse to read.
Move two — grade the phase you just finished. Honestly, alone, on paper, where nobody is watching and you do not have to defend it.
- Phase One, command: Can three people picked at random state the company goal in one sentence, without looking it up? Not the mission statement. The goal. Go walk the floor and ask three people this week. What comes back is your grade.
- Phase Two, cash: Is EBITDA in dollars higher than twelve months ago — and can you name the two specific decisions that moved it? Revenue does not count. “The market was better” does not count.
- Phase Three, the flywheel: If you were unreachable for thirty days, what breaks? Write the actual list. It is usually shorter and more specific than people expect — rarely “everything,” usually four decisions only you make.
Move three — pick one thing and date it. One, not five. Take the earliest failed phase, take its deliverable, and convert it into a single dated commitment inside thirty days. Not a project — a date and an outcome. Failed Phase One? A leadership meeting on the calendar this month that ends with one written goal and named owners. Failed Phase Two? The Right-to-Grow Ratio calculated off your actual P&L by the fifteenth. Failed Phase Three? The thirty-day break list, with one item permanently assigned to a named human being who is not you, by month end.
Then tell somebody — your CFO, your sponsor, your spouse. An undated intention is a wish, and a wish has never moved an income statement.
What does “you cannot skip a phase, only defer it” mean in practice?
It means the framework has no shortcuts, only interest payments.
A CEO on Day 500 whose organization cannot state the company goal does not have a growth problem. He has a Phase One problem, and he is paying interest on it. He is not doing Phase Three work; he is doing Phase One work, late, against a team that has already decided how seriously to take him. That is the crucial difference between doing the work on schedule and doing it later: the first time you get to announce a new era, and the second time you are correcting a failed one. Same four meetings, different room, different air in it.
The interest compounds in three specific ways. The work itself gets harder, because the organization has spent another year encoding the old behavior into systems, comp plans and habits. The clock gets shorter, because the deferred phase still takes as long as it always did but now has fewer days left to compound into a result a buyer can see. And your credibility gets thinner, because you spent the intervening quarters presenting a plan that did not produce the deliverable, which is precisely the conversation that starts around month nineteen.
So the practical rule is: whatever phase you fail is your real work. Not the phase the calendar says you are in — the phase you skipped. Grade backwards, find the earliest failure, and start there even if the date on the wall says you should be somewhere else. Nobody has ever recovered a hold by working harder on the phase they were supposed to be in while the deferred one accrued.
What does it look like when this is working?
It looks boring. That is the tell.
The scoreboard is the same five to ten numbers every week. The same people present their own numbers. Variances get causes and countermeasures, not adjectives. Nobody is surprised in a board meeting, because bad news travels fast in a company where it is safe to move it.
Theater looks exciting. Big kickoff, great deck, a new name for the program, real energy for six weeks. Then the cadence slips — the weekly gets moved for a customer visit, then moved again, then it is every other week — and by the second quarter nobody can state the goal, because it changed twice and neither change was announced.
Boring compounds. Exciting is what people do instead of the work.
The through-line: luck is the residue of design
Every CEO who delivers a premium exit will, over a drink, tell you they got lucky. The market turned. A competitor stumbled. A strategic buyer arrived with a reason to overpay. All true — and none of it the explanation.
The explanation is that when the luck arrived, the business was in condition to catch it. Clean segments. Real margin. A team that runs without the CEO in the room. That is not luck. That is residue, and somebody designed it on purpose, on a schedule, starting around Day 1.
The CEOs who were replaced will tell you they got unlucky. Also true, also not the explanation. Luck came for both. Only one was built to hold it.
You have a thousand days. Go find out which one you are standing on.
The ten episodes
| # | Episode | Phase | What it delivers |
|---|---|---|---|
| 1 | The Clock: Why 1,000 Days Decides Everything | All four | The map. The four phases, their deliverables, and the three-move self-diagnosis |
| 2 | Get a Goal: The One Number That Runs Your Company | Days 1–100 — Take control | Meeting One. One goal with a number, a unit, a date and a denominator |
| 3 | The 80/20 Engine: Where Your Profit Actually Lives | Days 101–365 — Earn the right to grow | The four-quad segmentation and the capacity diagnostic question |
| 4 | Earn the Right to Grow: The Two-Line Ratio | Days 101–365 — Earn the right to grow | The Right-to-Grow Ratio, its bands, and the two tracks it puts you on |
| 5 | The Complexity Tax: Quad 4 and the Dirty Dozen | Days 101–365 — Earn the right to grow | The twelve-source complexity audit and the four moves that retire the tail |
| 6 | Price First, Cost Last: The Five-Lever EBITDA Bridge | Days 101–365 — Earn the right to grow | The bridge in the right order, with a dollar figure and an owner per line |
| 7 | Zero-Up: Rebuilding the Budget From Zero | Days 101–365 — Earn the right to grow | The six-step rebuild, and capacity moved to where profit lives |
| 8 | The Team Lock: The Rule of Three | Days 366–730 — Scale what matters | Direction, Execution, Truth — the architecture that carries the company |
| 9 | The Cadence: How to Take Charge Without Taking Over | Days 366–730 — Scale what matters | Weekly, monthly, quarterly — the rhythm that makes the architecture load-bearing |
| 10 | Always Be Exiting: Building the Premium by Day 1,000 | Days 731–1,000 — Always be exiting | The four levers, the seven things buyers price, and the premium |
Frequently asked questions
What is the 1,000-Day Framework? The 1,000-Day Framework is a four-phase operating method for private-equity-backed CEOs, developed by Bill Canady. A hold typically runs five to seven years, but the window in which a CEO can still change the outcome is roughly the first 1,000 days. Those days divide into four phases, each with exactly one deliverable and one deadline: Days 1–100 take control, deliverable command; Days 101–365 earn the right to grow, deliverable margin, EBITDA and cash visibly turned; Days 366–730 scale what matters, deliverable a flywheel that runs without the CEO; Days 731–1,000 always be exiting, deliverable the premium.
Why 1,000 days and not the full hold period? Because a decision only counts if it has time to compound. Sponsors buy with an exit already modeled and dated, and by roughly Day 1,000 the diligence is running and the story is being packaged. The numbers a business is sold on are the numbers it has already made. Holds have stretched to their longest in about twenty-five years, but that extends the waiting, not the working window.
What is the deliverable of each phase? Command by Day 100 — anyone in the building can state the goal, the strategy and who owns what, verbatim. Margin, EBITDA and cash visibly turned by Day 365 — turned, not planned. A flywheel by Day 730 — growth that continues while the CEO is on a plane, a leadership team where each person owns and presents a number, and a weekly, monthly and quarterly cadence that runs regardless of the CEO’s energy. The premium by Day 1,000 — the multiple, worked deliberately.
Can you skip a phase in the 1,000-Day Framework? No. You can only defer it, and deferred phases get more expensive every quarter. A CEO on Day 500 whose organization cannot state the company goal is not doing Phase Three work — he is doing Phase One work, late, against a team that has already decided how seriously to take him. Whatever phase you fail is your real work, not the phase the calendar says you are in.
How do I know which phase I am in? Take the date the deal closed, not the date you started, and count the days to today. Then grade the phase you just finished: can three people picked at random state the company goal in one sentence; is EBITDA in dollars higher than twelve months ago and can you name the two decisions that moved it; if you were unreachable for thirty days, what breaks. Most CEOs are further along the clock than they feel.
What is the difference between growth and expansion? Growth is adding a dollar of revenue and keeping a larger share of it than you did last year. Expansion is adding a dollar and keeping a smaller share. They are indistinguishable on the top line and diverge on the bottom line roughly four quarters later — long enough for a CEO to have hired, invested and committed against the wrong one. Growing a business that has not earned the right to grow scales the dysfunction.
Why does the multiple re-rate last instead of first? Because margin and cash are the fastest levers a CEO owns, while everything that lifts the multiple is a form of track record — forecast accuracy over eight quarters, a bench with named successors, add-backs converted into realized results, a business that runs without the owner. Track record takes the two years spent building the team and the cadence, which is why it cannot be sprinted in the final ninety days. Buyers can tell the difference between a business that was built and one that was tidied.
Find out where you are on the clock.
The Board’s Number calculator reconstructs the underwriting behind your deal — entry EBITDA, entry multiple, target MOIC, hold period — and solves for the exit EBITDA you actually have to deliver. It takes about ten minutes and it produces the number this entire framework is organized around.
