Business turnaround strategy is one of those terms that everybody uses and almost nobody defines, which is a problem, because the four things it can mean require opposite actions. I have watched competent boards approve a turnaround plan while three people in the room held three different definitions of what they had just agreed to, and the resulting eighteen months were exactly as productive as that sounds.
So this is a definitional piece before it is an argumentative one. What the term covers, the four modes it can refer to, the criteria that separate them, and an honest account of which ones work in which situations. It is written for an operating executive or an owner rather than for a manufacturing audience specifically. If your situation is a plant, the operating sequence lives in how to turn around a manufacturing plant.
What a business turnaround strategy actually is
A turnaround strategy is a choice about which part of the business you are going to change, made under a time constraint. That is the whole definition, and the two halves of it both matter. The choice narrows to four options. The time constraint is what distinguishes a turnaround from ordinary strategic work, because it removes the option of doing all four properly.
Most of the confusion in this field comes from people using the word turnaround to describe an outcome rather than a choice. A recovery in performance is an outcome. It can be produced by any of the four modes below, by a market cycle, or by an acquisition, and treating the outcome as the strategy is how organizations end up copying the actions of a company whose situation had nothing in common with theirs.
The other reason the definition matters is governance. Boards approve turnarounds. If the board and the executive hold different modes in mind, the executive gets judged against a set of expectations that were never stated, usually at month nine, usually badly.
One more piece of the definition, because it is the piece that gets lost. A turnaround strategy is a choice about what you change, which makes it equally a choice about what you leave alone, and under a time constraint that second half does more work than the first. Every mode below implies a list of things you are explicitly not touching this year. Executives who cannot produce that list on request have not chosen a mode. They have chosen all four, and they will discover the consequence around month seven, when the organization runs out of capacity to absorb any of them properly.
What it is not: a restructuring
A restructuring changes the capital structure and a turnaround changes the operating model, and conflating them is the single most common category error in this space. Restructuring moves debt, renegotiates covenants, sells assets, and converts obligations. It is financial engineering in the neutral sense of the term, and it is frequently necessary.
What it does not do is change how the business runs on a Tuesday. A restructured business with an unchanged operating model has purchased time and nothing else, and the clock on the purchased time runs at the same rate as the clock that ran out the first time. I have seen this sequence twice at close range, and in both cases the second crisis arrived faster than the first because the balance sheet had less give in it.
The useful framing is that restructuring buys the calendar a turnaround spends. They run in parallel and they are not substitutes. If the only thing on your plan is a balance sheet action, you do not have a turnaround strategy, you have a deadline extension.
There is a practical test if you are unsure which one you are looking at. Ask what the business will be able to do at the end of it that it cannot do now. A restructuring answers that question with a number: less interest, more headroom, a covenant reset. A turnaround answers it with a capability: we will be able to see line output within the hour, or price against value rather than cost, or retire a product without a six-month argument. If the only available answer is financial, no operating change is underway, whatever the plan is called.
What it is not: a cost program
A cost reduction program is a lever available inside a turnaround, not a turnaround in itself, and the distinguishing feature is that cost programs have a target number while turnarounds have a target state. That difference sounds semantic until you watch how each one ends.
A cost program ends when the number is hit. Everybody goes back to work, the organization exhales, and the structural conditions that produced the cost base remain exactly as they were. Within two to three years the cost base reassembles itself, usually in slightly different places, and a second program gets commissioned by someone who describes the first one as insufficiently rigorous.
A cost program ends when the number is hit. A turnaround ends when the business can hold the number without you. Those are different finish lines and they take different amounts of time.
The tell is whether anything about how decisions get made has changed by the end. If the same people have the same authority over the same things and the only difference is the size of a budget line, you ran a cost program. That is a legitimate thing to have run. It is not what your board thinks it approved.
The distinction also determines who should own the work. Cost programs belong with finance, because the deliverable is a number and finance can verify it. Operating model changes belong with the line, because the deliverable is a behavior and only the line can install it. Businesses that hand the second thing to the function built for the first end up with an excellent tracking spreadsheet attached to nothing, and the tracking itself becomes the evidence that work is happening, right up until somebody asks what has actually changed.
The four modes and the criteria that separate them
There are four turnaround modes, and which one applies is determined by two questions rather than by preference. Does the business model still work, and how much calendar do you have. Those two axes produce the grid below, and nearly every failed turnaround I have looked at closely was a mode applied in the wrong quadrant.
The first axis is whether the underlying business model still works. A business whose model works is failing at execution: it wins work and cannot convert it, or it converts work and cannot price it. A business whose model no longer works is failing at position: the segment moved, the buyer changed, or the thing it is good at stopped being scarce.
The second axis is calendar, which in practice means cash runway. Under roughly two quarters of runway your options collapse to the ones that produce cash quickly. Past that, the options that produce durable value become available, and the difference in what you can attempt across that boundary is larger than most executives expect before they have been on the wrong side of it.
Two notes on reading the grid. It is a snapshot rather than a permanent position, so recheck it quarterly, because retrenchment moves you up the runway axis and repositioning moves you across the model axis by design. And both axes are answered with evidence rather than with judgment. Runway is arithmetic. The model question is answered by quote win rate and backlog, both of which you can pull this week. Executives who place themselves on this grid by intuition consistently place themselves one quadrant away from where their own numbers put them, and always in the more flattering direction.
Retrenchment
Retrenchment is the mode that produces cash and calendar, and it is the correct first move whenever runway is short regardless of which side of the model axis you sit on. It covers working capital release, discretionary spending, footprint consolidation, and the exit of activities that consume more than they return.
The mechanism is simple and the discipline is not. Nearly everything genuinely available to you in the first six weeks is money the business already owns and has parked somewhere unproductive: inventory sized against an old reliability assumption, receivables past terms because nobody owns the conversation, premium freight that exists because planning stopped working, capital in flight against assumptions that no longer hold. None of that requires taking anything away from anyone, which matters politically as much as financially.
The discipline is the boundary. Take every action that releases cash without permanently reducing capability, and take none that do. The moment retrenchment crosses into selling capability you will need in eight months, it has stopped buying calendar and started spending the recovery, and the person who discovers that will be your successor.
The other property worth knowing is that retrenchment has a natural end, and most organizations run past it. The first pass through working capital and discretionary spend produces the large majority of what is available. The second pass produces a fraction of the first at several times the political cost, and the third pass is where capability begins leaving the building disguised as efficiency. When the return on a pass drops sharply, stop, bank the calendar you bought, and move to the mode your diagnosis actually called for.
Repositioning
Repositioning changes what you sell and to whom, and it is the right mode when the model has stopped working rather than when execution has slipped. Segment exits, price architecture changes, channel changes, product line refocusing, and the deliberate abandonment of customers you have served for twenty years all live here.
It is the highest-value mode and the slowest, because a position change has to survive a full sales cycle before you can tell whether it worked. That timing property is what makes it so dangerous to attempt on short runway: you commit the organization to a direction, absorb the revenue dip that every repositioning produces, and then need enough calendar left to see the other side of it.
The test for whether you are in this mode is the fork between losing orders and losing to orders. If you are quoting and not winning, the problem is upstream of operations and repositioning is on the table. If you are winning and failing to convert, repositioning is a very expensive way to avoid an execution problem. I wrote the diagnostic separately: cost problem or demand problem.
Repositioning also carries a cost that rarely appears in the plan, which is the revenue trough between abandoning the old position and establishing the new one. That trough is not a risk to be managed away. It is a structural feature of the mode, because you stop serving the old customers before the new ones arrive. Size it before you commit, give the board the number, and hold enough runway to cross it. Repositionings that fail have usually failed in the trough rather than at the destination.
Replacement
Replacement changes who makes the decisions, and it is best understood as a precondition rather than as a strategy. It appears in the grid because in one specific circumstance it is genuinely load-bearing: when the position that needs changing was chosen by the people currently in charge of changing it.
The mechanism is not about competence. It is about sunk commitment. A leadership team that spent five years building toward a position will evaluate evidence against that position through a filter they cannot see, and the filter tightens as the evidence accumulates. That is a normal human property rather than a character flaw, and it is why boards reach for replacement when the required change is directional rather than operational.
Replacement is not about competence. It is about sunk commitment. Teams evaluate evidence against a position they chose through a filter they cannot see.
Where it fails is when it is used as a substitute for a diagnosis. Replacing an executive without knowing which of the other three modes is required produces a new executive who has to run the diagnosis you skipped, from a standing start, with less calendar than you had. That is not a strategy. That is a delay with a severance cost attached.
The related error is replacing at the wrong level. Boards reach for the chief executive because that is the lever they hold, but the sunk commitment blocking a directional change frequently sits one or two levels below, among the people who built the current position and still run it day to day. Changing the top while leaving that layer intact produces an executive with a mandate and no mechanism, and the organization absorbs the new direction roughly the way it absorbs weather.
Renewal
Renewal rebuilds the operating capability so that a model which still works actually performs, and it is the mode that gets discussed least and applies most often. Measurement systems, decision rights, planning capability, cadence, product development throughput, and the management layer’s ability to run any of it.
The reason it is underrepresented in the literature is that it makes for poor narrative. There is no bold exit, no dramatic segment pivot, no new leadership arriving with a mandate. There is a year of unglamorous work installing the things that should have been installed a decade ago, and at the end of it the same business sells the same products to the same customers and makes considerably more money doing it.
It is also the mode with the most durable result, because the thing you built is a capability rather than a decision. Retrenchment buys calendar that runs out. Repositioning creates a position that eventually gets copied. Renewal changes the rate at which the organization can do anything, and that compounds.
Renewal is also the mode most likely to be misread as inaction by anyone watching from outside the business. Six months in there is no announcement to point at and the financials have not turned, because capability changes move leading indicators first and lagging ones considerably later. That gap is where renewal efforts get abandoned in favor of something more visible. If you choose this mode, name the leading indicators in advance and report them monthly, or you will be asked to justify a year of work using a metric that is structurally not yet available.
Why three of the four get applied too early
Retrenchment, repositioning, and replacement all get reached for before the diagnosis is complete, and each one has a distinct reason, which is worth naming because knowing the reason is most of the defense.
Retrenchment gets applied early because it is the only mode that visibly does something inside thirty days, and a new executive or a nervous board wants evidence of action. The cost is that retrenchment applied to what is actually a demand problem removes the capability the recovery would have needed. That loop is self-sealing: you shrink, you cannot serve the demand when it returns, you lose the work, and you produce the evidence that justifies the original decision.
Repositioning gets applied early because it is the most intellectually satisfying mode and the one that makes the best board presentation. It also happens to be the mode that lets everyone avoid the conclusion that the business is simply not executing, which is a less flattering finding about the people in the room.
Replacement gets applied early because it is decisive and available. It requires no analysis, it demonstrates seriousness, and it transfers the diagnostic burden to somebody who is not yet in the room to object.
Renewal is the one mode nobody applies too early, and that asymmetry is informative. It gets chosen late or not at all, because selecting it requires the leadership team to conclude that the problem is the operating model they built and run. The other three locate the problem somewhere more comfortable: in the cost base, in the market, or in a person. If your instinct is pulling hard toward one of those three within the first month, that instinct is worth a week of scrutiny before you act on it.
How to sequence the modes
The modes are not alternatives to be chosen between so much as a sequence to be entered at the right point, and the order is fixed by dependency rather than by preference.
- Establish the runway. Cash on hand and weekly burn. This determines which modes are available to you at all, and it takes a day.
- Diagnose the model. Are you losing orders or losing to orders. Two weeks, and it determines which column of the grid you are in.
- Retrench only as far as the capability boundary. Release trapped cash. Stop when the next action would remove something the recovery needs.
- Decide replacement deliberately, if at all. Only where the required change is directional and the current team chose the direction.
- Run renewal or repositioning against the diagnosis. Left column is renewal, right column is repositioning. This is where the calendar you bought gets spent.
- Install the cadence that holds it. A recovery without a standing rhythm to sustain it unwinds on a schedule set by your departure date.
Two rules govern entry into the sequence. You can enter at any step, but you cannot skip step two, because every subsequent choice depends on which column you are in and there is no way to infer that from the financials alone. And once you have selected a mode at step five, hold it for at least two quarters unless the runway number changes materially. Switching modes on new information sounds like responsiveness from the executive suite and reads to the organization as an absence of direction, which costs you the management layer’s belief at precisely the point you need it most.
What actually works
Renewal has the highest success rate of the four in my experience, and it is chosen least often. The reason is not mysterious: it is the mode that assigns the problem to the operating model rather than to the market, the previous leadership, or the cost base, and that is a less comfortable place to put it.
The second thing that works, and it is a sequencing point rather than a mode, is retrenching to the capability boundary and stopping there. Executives who hold that line preserve the option value of every other mode. Executives who cross it have converted a turnaround into a managed decline and will not know it for about a year.
Third, and this one is underrated, is making the mode explicit in writing to the board before you start. One page: which mode, which evidence, what will be true at month six and month twelve, and what would cause you to change modes. That document is worth more than the plan it summarizes, because it prevents the month-nine conversation where you are assessed against a mode you never chose.
A fourth thing that works, and it costs nothing, is naming the capability boundary in advance. Write down, before you begin retrenching, the specific things you will not sell, cancel, or let go of regardless of pressure. Two or three items, named. The value of writing it early is that the pressure to cross the boundary arrives later, arrives with a good argument attached, and arrives when you are tired. A list drawn up under calm conditions is the only defense I have seen hold reliably.
What actually does not work
Running all four modes at once does not work, and it is the most common failure pattern I see in businesses with resources. It feels like urgency. What it produces is an organization where every function is simultaneously cutting, repositioning, absorbing new leadership, and rebuilding process, and where nobody can tell which of those four things any given instruction belongs to.
Sequential cost programs do not work either. If your business has run two cost reduction programs in five years and is about to run a third, the finding is not that the previous two lacked rigor. The finding is that the cost base is an output of the operating model and you have been treating it as an input.
And a turnaround with no defined end state does not work. If nobody can say what will be true when this is finished, then nothing will be finished, and the organization will remain in the elevated-effort posture indefinitely, which is a thing organizations can sustain for about eighteen months before the good people begin to leave.
Benchmarking your way into a mode does not work either. The instinct to look at what a comparable company did during its recovery is reasonable, and the output is almost always misleading, because the published account describes the actions and omits the two variables that determined them, which are runway and model viability at the moment the decision was made. Copying the action of a business that was standing in the other column is the most expensive form of research available to you.
How do you know which mode you are in?
Two questions place you on the grid: whether the business model still works, and how much cash runway you have. Answer the model question by testing whether you are losing orders or losing to orders, and answer the runway question in weeks rather than in reassuring language. Those two answers select the mode. Nothing else does.
If the two answers point to different modes, runway wins in the short term and the model answer governs what you do after. That is the entire sequencing rule, and it resolves most of the arguments this decision produces.
One caution on the model question. In a business with two genuinely different segments, or with several sites serving different buyers, the two axes will average into a reading that describes none of it. Run the placement separately for each segment and expect the answers to differ. A portfolio containing one repositioning candidate and one renewal candidate is a normal condition rather than a contradiction, and it is handled by sequencing the two rather than by forcing a single answer onto the whole company.
Can you run two modes at once?
You can run retrenchment alongside exactly one other mode, and you cannot run more than that. Retrenchment is compatible because it operates on working capital and discretionary spend rather than on direction, so it does not compete for the organization’s decision-making capacity in the way the other three do.
Repositioning and renewal together is the combination that reliably fails. Both consume the same scarce resource, which is your management layer’s ability to absorb and execute change, and running them concurrently produces two efforts at forty percent rather than one at full strength.
The exception people ask about is running renewal in one part of the business and repositioning in another. That is workable, and the constraint is not the strategy but the management layer’s attention. Each effort needs a leader whose primary accountability is that effort, and the two cannot share a leadership team below the executive level. Where the same six people are asked to carry both, the two efforts converge into one underpowered program inside a quarter, and neither reaches its own finish line.
How long does each mode take?
Retrenchment shows results in weeks, renewal in two to four quarters, repositioning across a full sales cycle plus a quarter, and replacement adds the new executive’s own diagnostic period to whichever mode follows it. Those are rough shapes rather than commitments, and the sales cycle variable is the one that moves most between industries.
The practical implication is that your runway determines your mode more than your preference does. If you have one quarter of cash and a position problem, repositioning is not available to you no matter how correct it is, and the honest strategy is retrenchment to buy the calendar that makes it available.
The other timing variable worth naming is the board’s patience, which is finite and almost never stated as a number. Ask for it explicitly at the start. A board expecting a visible turn by month six and an executive running renewal, which shows leading indicators first, are on a collision course that neither party has yet noticed. Getting the expected timeline written into the same document that names the mode removes most of the risk of the month-nine conversation.
Monday morning, write one page that names your mode, the evidence for it, and what would make you change it. Send it to your board before you send them a plan.
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 Global VP of Product Strategy and Marketing. 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.
Most mode errors are decision errors first. If your leadership team cannot agree which mode you are in, the underlying problem is usually how long a decision takes to clear. Book a Decision Velocity Audit and we will measure it before you commit to a strategy.
See also: Should I Fix It, Sell It, or Shut It Down?, How to Decide What to Fix First