- Day 1: The Triage Protocol
- Days 1 to 30: The Foundation
- Days 31 to 60: The Structural Execution
- Days 61 to 90: Deployment and the Decision
- The Scorecard: Closing the Cycle on Your Terms
- Day 91: The Next Cycle
- About the Stagnation Assassin
Diagnosis without sequence is trivia.
You can know your business is a Growth Trap, know the trajectory, know all five structural upgrades by name, and still lose, because knowing what to do and knowing what to do first, on which day, with which authority, reading which signals, are entirely different competencies. The first one comes from a book, or from Harvard Business Review’s take on how to turn around nearly anything. The second one is a calendar, and this page is that calendar.
Here is what you need before you start: a three-letter signature and a named type. If you do not have them, this page cannot help you yet. Go run the ATM Test, which takes ten minutes, then match your signature in the 12 Business Types, which takes five more. Come back holding your name. Everything below assumes you have one.
And here is what this page is honest about: it carries the sequence and the judgment calls. The daily cadence, which pull happens which morning, which deliverable lands which Friday, lives in the 90-Day Companion Workbook, the tear-out operating document this playbook runs against. The judgment calls on this page are worth far more when the cadence underneath them is already handled.
Ninety days, four phases: the Day 1 triage, the thirty-day foundation, the thirty-day structural execution, and the thirty-day deployment that closes on a decision. Start the clock.
Day 1: The Triage Protocol
The 90-day playbook is three playbooks, not one. Day 1 classifies the operation into one of three execution paths: Emergency Triage for active failure, Stabilization for capped stagnation, and Position Hardening for operations that already rebuilt. Your three-letter signature and named business type set the path, and the mapping is fixed.
This is the structural distinction most transformation methodologies miss. The conventional approach publishes one generic 90-day plan and assumes every operation starts from roughly the same place: thirty days of listening, thirty days of planning, thirty days of early wins. Run that generic plan on an operation in active failure and you will spend your listening tour watching the runway burn. Run it on an operation that has already rebuilt and you will bore a healthy business into losing its edge. The same ninety days in front of three different operations should produce three different opening moves.
So the first decision of Day 1 is not an action. It is a classification, into one of three execution paths.
Path A: Emergency Triage. For operations in active failure mode: Money Pit territory at meaningful severity, where the trailing data will catch up to the structural slope within roughly four quarters, and the operator who waits will face the failure mode instead of acting before it lands. This path runs the prescription layer at maximum tempo. The first thirty days deploy leak-plugging against the largest dollar leak. The next thirty execute the unilateral pricing and portfolio moves that require no structural commitment. The final thirty scope and authorize the structural moves.
Path B: Stabilization. For operations not in failure but in stagnation that caps the structural slope: acceptable trailing financials, no structural advantage being built. The work is converting a steady-state position into an advancing one before a material headwind arrives to convert it into something worse. This path runs at deliberate tempo: thirty days of diagnostic depth, thirty executing the named type’s primary upgrade, thirty confirming the signature migration is underway.
Path C: Position Hardening. For operations that have already captured the rebuild and are operating in the High band on multiple metrics. Nothing needs fixing, which is exactly the danger; these operations need the strategic-navigation layer to hold the signature beyond the current operator’s tenure. The Aggression Gap diagnostic and the Inheritance Standard audit run in the first thirty days, the hardening practices install in the next thirty, and the long-horizon allocation discipline sets in the final thirty.
The mapping from type to path is fixed:
| Execution Path | Types |
|---|---|
| A: Emergency Triage | Growth Trap (H/L/L), Acquisition Hangover (H/L/L), Busy Pit (M/L/L), Slow Bleed (L/L/M), Phantom ATM (H/M/M)* |
| B: Stabilization | Cruiser (M/M/M), Margin Fortress (L/M/M), Cash Vampire (M/M/L), Founder Bottleneck (H/L/M), Capital Glutton (H/M/L) |
| C: Position Hardening | Sprinter (H/H/H), Hidden Compounder (M/H/H) |
*The Phantom ATM is the one type whose path is set by trajectory rather than current signature. Read on its signature alone, H/M/M lands in Stabilization. But its 12-to-18-month trajectory toward a full Growth Trap pulls it into Emergency Triage, because the structural intervention has to happen before the trajectory completes. Nothing looks broken. That is why it goes first, not last.
The path selection also runs through one more filter the type table cannot see: your actual authority to act. The playbook assumes decision rights over pricing, portfolio, and structure. If you hold some of those rights but not all, the sequence modifies rather than dies, and the modification is its own discipline: Running the Playbook From the Middle. The full triage architecture, the decision tree, the severity calibration, and the mis-triage failure modes (running Emergency Triage on a Slow Bleed burns authority on urgency the organization cannot see; running the standard tempo on a Cash Vampire is too slow to matter) are in The Triage Protocol deep-dive.
Path chosen, Day 1 is over. The foundation month begins.
Days 1 to 30: The Foundation
The first thirty days produce exactly three deliverables: a diagnosis you trust, the leaks you can plug without anyone’s permission, and a scoped view of the structural work that permission will be required for. Four judgment calls determine whether that foundation is sound or merely complete, because a checklist can be finished and still be wrong.
The Companion Workbook carries the daily cadence. What follows are the judgment calls.
Judgment call one: knowing when the data is good enough to band. The diagnostic does not run on clean SKU-level attribution. It runs on directional accuracy at the segment level, because the three velocity metrics produce band assignments, and a band assignment is robust to noise that would ruin a unit-cost analysis. Revenue Velocity is a two-point-wide band; you do not need revenue accurate to the dollar to know whether you are beating the industry by five points or losing to it by four. The operator who spends the first three weeks reconciling the data warehouse has spent three weeks letting the trajectory compound while producing precision the diagnostic never asked for. The test is simple: if the same number, perturbed by the size of your data uncertainty, would land in a different band, get better data on that one input. If it would land in the same band, stop pulling and start calculating. Most inputs clear this test on the first pull.
Judgment call two: building the industry comparator honestly. Revenue Velocity is your organic growth minus the industry’s, and the second number rarely has a clean published source that maps to your segment, your geography, and your definition of organic. The defensible move: average the organic growth of your three to five closest publicly traded competitors, measured the same way you measure your own. The trap is asymmetric and worth staring at. Pick a comparator that is too narrow, a single struggling competitor, a subsegment growing slower than your true market, and you flatter your own RV, reading High where the truth is Moderate.
A flattered RV is the most dangerous error in the entire diagnostic, because High Revenue Velocity is the signature element that makes a Growth Trap look like a Sprinter. A conservative RV under-celebrates a real win at worst. A generous RV hides the exact failure mode this methodology exists to catch.
When in doubt, bias toward the broader, faster market definition.
Judgment call three: refusing to band a lying Flow-Through. When revenue barely moved year over year, FT’s denominator is small, and a small denominator makes the ratio jump on noise: flat revenue plus a modest operating-profit swing can print 300 percent or negative 200, and neither number means what the band table implies. Worse, a declining-revenue year with declining operating profit divides negative by negative and prints a high positive FT that is pure arithmetic artifact; banding it as overhead leverage is a serious misdiagnosis. If revenue moved less than a few points, treat single-year FT as suspect and weight the three-year trend.
Judgment call four: plugging the unilateral leaks and reading the pushback as data. By the second week the customer-product matrix exists (the full method is customer profitability analysis), and the bottom of the portfolio has a short list of combinations destroying value. The ones you can act on without anyone’s sign-off, minimum order quantities on long-tail accounts, discontinuation notices on dead SKUs, repricing on grandfathered specialty configurations, are the first wave. They are reversible at low cost, which means they clear the 70 percent confidence bar of the 70% Rule, and you do not need to wait.
The judgment is not in deciding to plug them. It is in reading the response, because pushback comes in two kinds and they demand opposite handling. The first kind is friction: the buyer is annoyed at a change to a long-standing arrangement and is testing whether you fold. Friction is noise. Hold the move. The second kind is information: the buyer reveals a value, a volume, or a strategic role in your portfolio that the matrix did not capture. Information gets weighed, and occasionally it reverses a move, which is fine, because reversibility is why these moves went first.
The month closes with the third deliverable: the scoped view of the structural work ahead, sized and sequenced, ready to be authorized. If you are running from the middle without full authority, this scoped view is your product, aimed upward.
Days 31 to 60: The Structural Execution
The second thirty days run four workstreams in parallel: the pricing test goes live, the streamline phase removes complexity, the automation roadmap gets built, and the redeployment plan gets staffed. Data starts coming back from your own moves, and three of the four judgment calls are about reading ambiguous early signals correctly.
The workbook carries the week-by-week sequence. This month’s character is different from the first, and the difference is the reason its judgment calls matter more: this is when data starts coming back from your moves. Any generic 90-day plan can schedule actions. Reading the returns is where transformations are won and lost.
Judgment call one: reading an ambiguous elasticity test. The pricing test (the full 30-day method is in Running a Pricing Test Without Blowing Up the Account; this page owns its calendar, that page owns its mechanics) produces clean answers more often than operators expect, because it runs on configurations the audit identified as underpriced, and underpricing means the market was already willing to pay more. The judgment call is the ambiguous case: acceptance drops, modestly, and you must decide whether that is rejection of the price or normal sampling noise on a small test population.
Three discriminations resolve it. First, separate the acceptance-rate signal from the margin-per-deal signal: a test that loses ten percent of deals but lifts margin twenty-five percent on the deals it keeps is net-positive, and the drop is not failure, it is the test working, shedding the most price-sensitive sliver while the rest pays more. Read only the acceptance rate and you will kill a winning move. Second, read the drop’s shape: lost deals that cluster in one segment, one salesperson, one configuration are a local signal, so hold the move with a carve-out; lost deals scattered evenly across the population mean the price found a genuine ceiling, so reset to an intermediate point. Third, resist deciding early. Five business days shows direction, not always magnitude. A genuinely ambiguous first week, small drop, mixed shape, margin lift roughly offsetting volume loss, gets a second week. Ambiguity is information that says you do not yet have enough information, and forcing a verdict on a coin-flip read is gambling, not testing.
Judgment call two: telling vestigial complexity from load-bearing complexity. The streamline phase removes operational complexity before any capital gets deployed against the constraint, on the premise that most operations carry fifteen to thirty percent complexity that exists for no current reason: approval steps nobody can justify, buffers sized to a demand pattern that no longer exists, reporting cadences that consume hours and produce no decision. The judgment is that vestigial and load-bearing complexity look identical from the org chart. An approval step seven people sign might be six signatures of theater and one that catches a real category of error. An oversized-looking inventory buffer might be quietly absorbing a supply variance that, removed, resurfaces as missed customer commitments. The streamline failure mode is not cutting too little. It is cutting a buffer that was doing invisible work and discovering the work only when it stops getting done. The protecting discipline is one question, asked of every piece of complexity on the cut list: what failure was this built to prevent, and is that failure still possible? Nobody can name the failure: vestigial, cut it. Somebody names it and it is still live: load-bearing, and removing it means first removing the failure mode it guards.
Judgment calls three and four follow the same signal-reading pattern across the automation roadmap and the redeployment staffing: the roadmap gets built this month but deploys next month, and the discipline of what never gets automated first is its own page, When to Automate During a Turnaround. The staffing call is about sequencing people onto the redeployment before the redeployment exists, which is a bet, and it gets sized like one: 70 percent confidence, reversible assignments first.
Days 61 to 90: Deployment and the Decision
The final thirty days deploy the first structural automation, activate the first redeployment, and re-run the three velocity metrics against the Week 1 baseline. A flat secondary diagnostic has three possible causes demanding opposite responses: lag, moves that were too small, or a wrong diagnosis, and the leading indicators separate them.
The workbook carries the deployment checklist and the measurement protocol. This month has two judgment calls, and the second is the single highest-value judgment in the back half of the playbook.
The first call: reading a rough deployment. A newly deployed system that is rough but converging is fine; roughness is what the first two weeks of any structural change look like. A system that is degrading the customer-facing result is the one the rollback line was written for. Know which one you are watching before you are watching it, by defining the rollback trigger during deployment planning rather than during the incident.
The call this playbook exists for: when the secondary diagnostic shows no migration. Week eleven re-runs the three velocity metrics against the Week 1 baseline. The expected result is migration: Profit Velocity out of Low, Flow-Through climbing, the signature stepping toward its target. The chapter the workbook cannot write is the one where the migration is not there.
No migration at the ninety-day mark is not automatically failure, and the first job is diagnosing why, because there are three very different causes demanding opposite responses.
Cause one: lag, not absence. A pricing test that rolled out in Week 6 and an automation that went live in Week 9 have not had time to propagate into a trailing twelve-month metric by Week 11. PV moves before FT; FT moves before the trailing operating margin reflects either. If the leading indicators are moving, quote acceptance at the new pricing, throughput on the new line, the freed-capacity numbers, but the banded signature has not shifted, the migration is in flight and the trailing metric is doing what trailing metrics do. The response is to hold the course and measure the leading edge. This is the most common cause of a flat secondary diagnostic, and the operator who panics here kills the transformation one quarter before it would have printed.
Cause two: the moves were too small. The unilateral leaks got plugged, the pricing test ran, and the dollar magnitude of the executed moves was never large enough to shift a banded metric. The tell: the leading indicators are also flat. Not lagging. Flat. This is a scoping error, not a strategy error, and the response is not to wait. It is to go back to the customer-product matrix and the constraint audit and execute the larger structural move the first cycle scoped and deferred, usually the one that required authority or capital.
Cause three: the diagnosis was wrong. The prescription executed faithfully and the signature did not move because the operation was never the type the first diagnostic named. The classic case: an operation diagnosed as a Growth Trap that was actually an Acquisition Hangover, same H/L/L signature, different root cause, where organic leak-plugging could never fix an integration problem. The tell: the moves executed cleanly, produced their local results, and both the signature and the leading indicators on the targeted layer stayed flat. The response is re-diagnosis against a different type.
Lag says hold. Too-small says go bigger. Wrong-diagnosis says start over. Read lag as wrong-diagnosis and you throw away a working transformation. Read wrong-diagnosis as lag and you burn another ninety days on a prescription that was never going to work.
The discrimination among the three is everything. The leading indicators are the separator: if the layer you targeted is moving at the leading edge, you have lag or under-scoping; if it is flat at the leading edge, the diagnosis is in question.
The Scorecard: Closing the Cycle on Your Terms
The close-out scorecard is one page in four fixed blocks: a header with starting and ending signature, a metrics block with bands and migration, a moves block, and a forward block with the Week 12 decision. Pre-registering the expected trailing lag in month one turns the month-three conversation into confirmation.
A transformation the stakeholder set cannot see is a transformation the operator will be asked to defend on the stakeholders’ terms, and their terms are trailing. Ninety days of structural work will not yet have fully landed in the trailing P&L, which means the close-out communication either puts the progression on the diagnostic’s terms or surrenders the narrative to numbers that have not caught up yet.
The scorecard is one page, and the constraint is not aesthetic: executive audiences read the first page closely and everything after it with declining attention. Four blocks, in fixed order, verdict to evidence to forward motion. The header: cycle dates, starting type and signature, ending type and signature; an executive who reads only the header holds the whole trajectory, because H/L/L to H/M/M tells the story before the supporting blocks. The metrics block: the three velocity readings at Week 1 baseline and Week 11 secondary diagnostic with their bands and migration, plus the two financial lines every audience tracks independent of any framework, operating-margin change in basis points and cash-flow impact in dollars. Five numbers, and they are what a skeptical reader checks first. The moves block: the structural work executed, sequenced by upgrade, one line each, the operational evidence that the metrics did not move by accident. The forward block: the Week 12 decision, the next cycle’s path, type, priority upgrades, and structural commitments with timelines.
The credibility mechanics matter as much as the format, and the biggest one is pre-registration: state the expected trailing lag in month one, in writing, so the month-three conversation is a confirmation instead of an excuse. The full architecture, the filled composite example, and the language discipline that separates a structural progression from a project status report are in The 90-Day Scorecard deep-dive.
Day 91: The Next Cycle
Day 91 closes on a choice with three branches. Migration underway means run the next cycle on the same path. Migration complete into a new type means re-triage against the new name. No migration means re-diagnose. The 90-day cycle is the permanent operating rhythm of a business that reads its own slope.
The cycle closes on a choice, not a finish line, and the choice has three branches that map to the three secondary-diagnostic reads. Migration underway: run the next cycle on the same path, executing the structural moves the first cycle authorized. Migration complete into a new type: re-triage against the new name, because a Growth Trap that has stepped to H/M/M is a different patient with a different priority. No migration and the diagnosis in question: re-diagnose, and be grateful the methodology caught it at Day 90 instead of Day 400.
Then the clock restarts, because the 90-day cycle is not a rescue protocol you graduate from. It is the operating rhythm of a business that reads its own slope: diagnose, triage, execute, measure, decide, again. The operators who compound are not the ones who ran one great cycle. They are the ones who never stopped counting.
The sequence is on this page. The cadence is in the workbook: put your signature at the top of the expanded 90-Day Companion Workbook and start Day 1 on Monday.
The 90-day playbook is Section 4 of Ten Minute Transformation (Koehler Books, February 2027), which carries all four phases in full chapter depth, every judgment call with its complete reasoning, and the tear-out Companion Workbook as Appendix C.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B+ in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as VP of Global Product Strategy. Known as The Stagnation Assassin, he is the author of two published books: The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.
You do not need a consultant and a six-month engagement to find out which of the three execution paths you are on. You need your signature and fifteen minutes. Book a 15-minute Triage Protocol read and start your ninety days running the right playbook, not the generic one.