A hundred days. That is the horizon nearly every integration plan is written to, and in an industrial acquisition it covers roughly the first sixth of the work. The rest of the sequence, the part that runs from month four to month eighteen, is where the value in the deal either shows up or quietly does not.
The standard post-merger integration playbook was not built for manufacturers. It was built in and for businesses whose value sits in contracts, systems, and people who can be recombined on a slide: professional services, software, financial services, distribution. In those businesses, integrating early is usually right, because the assets move when the org chart moves.
An industrial business does not work that way. Its value sits in physical flow, in a plant’s learned ability to build a specific thing at a specific rate, and in what a handful of people know that has never been written down. None of that moves when the org chart moves. Some of it breaks. So the sequence has to be different, and the difference is the subject of this article. The owner’s version of the wider first year is in how to run year one after an acquisition.
What does a hundred-day integration plan actually cover?
Control, reporting, and the decisions that cannot wait. That is a real and necessary scope. What it is not is the integration, because almost nothing physical can be safely changed inside a hundred days in a business you have owned for three months.
The confusion comes from treating the hundred-day plan as a compressed version of the whole job rather than its first phase. A compressed version tries to do a bit of everything early: some purchasing consolidation, a first pass at sales coverage, an ERP decision, a footprint study. Each of those is defensible on its own. Together, in month two, they commit the acquirer to a set of choices about a business nobody yet understands.
The better frame is that the hundred days exist to buy you the right to decide well later. You are establishing what you own, taking control of cash and safety, and building a single set of numbers. That is enough work for the period, and doing it properly is what makes months four through eighteen productive rather than corrective.
There is a practical test for whether your hundred-day plan has the right scope. Count how many of its line items would still be correct if the business turned out to be quite different from the one described in diligence. Control, reporting, safety, and decision rights all survive that test, because they are true regardless of what you find. A footprint study or a coverage change does not survive it, which is the signal that it belongs to a later phase.
Watch a related video: Integration: Why No Framework Works Without It.
Why does the standard playbook misfire in an industrial business?
Because it assumes value moves with structure. In a manufacturer, value sits in physical flow and in knowledge that lives in people’s heads, and reorganizing either one before you understand it destroys more than it captures.
Think about what a typical synergy case contains. Consolidated purchasing. Combined sales coverage. Shared back office. A single ERP. A rationalized footprint. In a services business, four of those five can be executed within a year at modest risk, because the work follows the people and the people follow the reporting line.
In a manufacturer, two of the five touch the thing that actually produces the money. Combined sales coverage changes who calls on which customer, which changes what gets sold, before anyone knows which products each side sells well. A rationalized footprint moves production between plants, and a plant’s capability is not portable in the way a spreadsheet treats it. Yield, scrap, changeover time, and the informal fixes that make a line run are learned locally over years.
The standard playbook also treats speed as a virtue in itself, on the reasoning that integration fatigue sets in and windows close. That reasoning is sound for the reversible moves and dangerous for the rest.
None of this is an argument for slowness as a virtue. The reversible half of the integration should move faster than most acquirers manage, because there is no reason to leave duplicated insurance, audit, treasury, and benefits administration running for a year. The argument is for separating the two halves and giving each the pace it deserves, rather than applying one tempo to a list that contains both.
What has to be true on day one?
People get paid, product ships, the plants are safe, and everyone knows who decides what. Day one is an operational continuity exercise, not a change exercise, and every hour spent on anything else is borrowed from those four things.
Payroll is first and it is not a formality. A missed or late first payroll in an acquired plant does more damage to trust than any communication plan can repair, and the people it damages trust with are the ones whose knowledge you most need.
Shipping continuity is second. Customers should not be able to tell from their delivery that anything happened. That means order entry, scheduling, and outbound logistics stay exactly as they are, whatever the eventual plan is.
Safety is third and is the one area where immediate change is appropriate. If the acquired site’s practices fall short of yours, that is not an integration decision to be sequenced. Fix it now.
Fourth is decision rights, stated plainly and in writing: what the site decides alone, what comes to the acquirer, and how fast the acquirer answers. Most integrations slow down in month two for the simple reason that nobody wrote this down in week one.
One more day-one item is worth naming because it costs nothing and is routinely skipped. Tell the acquired organization what will not change, specifically and in writing. People assume the worst in the absence of information, and the assumptions that form in the first week are remarkably durable. A short, concrete list of what is staying the same for at least the next six months buys more calm than any amount of vision language.
Keep the day-one communication short and operational. Long documents about vision and culture land badly on a plant floor on the morning everything changed. What people want to know is whether they are paid, whether their shift is the same, who their supervisor is, and who to call with a problem. Answer those four and the rest can wait for week two.
What happens in week one?
You start listening and you start counting. Week one is where an acquirer either builds or destroys the credibility it will need for every hard decision in the following seventeen months, and the difference is mostly whether anyone senior is physically present.
Go to the plants. All of them, in the first two weeks if the geography allows. Not for a presentation, for a walk. The walk is where you learn which operations run hot, where product queues, and which supervisors know their numbers without looking them up.
Start the counting at the same time. What the constraint operation actually produces per shift, on two different shifts. What the order book contains, read line by line rather than in summary. What the quote win rate is, and on what dimension the losses turn.
What you do not do in week one is announce anything about structure, footprint, or brand. Every one of those announcements converts the entire organization’s attention from running the business to protecting position, and you will need that attention intact for the diagnostic work.
Bring a notebook rather than a template. An integration checklist arriving in week one tells the site that the answers are already written, which ends the flow of information you came for. The people on the floor will tell an acquirer a great deal in the first fortnight, while the relationship is still undefined, and much less after the first structural announcement lands.
Two weeks is also long enough to form a first view of where the acquired business is genuinely strong. Write that view down before anything is announced, because it is the version least contaminated by what people later work out you want to hear, and it is a useful check on the diligence you inherited.
What belongs in the first thirty days?
One set of numbers, an established baseline, and a named constraint. By day thirty you should be able to state what the acquired business is physically capable of and where it stops being able to say yes, using evidence you gathered rather than evidence you inherited.
One set of numbers is the unglamorous prerequisite for everything else. Two businesses will define revenue recognition, cost of goods, capacity utilization, and on-time delivery slightly differently, and until those definitions are reconciled, every comparison between the two sides is an argument rather than an analysis.
Do it in the acquired company’s own systems where possible, even if that is slower. A parallel reporting structure built by the acquirer produces numbers that only the acquirer trusts, and the site quietly keeps running on its own set.
The baseline matters for a second reason. Every synergy you eventually claim will be measured against it, and a baseline built after the first changes have started is not a baseline. It is a negotiation.
Naming the constraint is the part that most distinguishes a good thirty-day output from a thorough one. A thorough output describes the business. A good one identifies the single operation, lead time, or commercial limit that governs everything else, and states the evidence behind it. That sentence is what the next fourteen months of integration work should be pointed at, and without it the plan defaults to whatever the synergy case said.
Expect the reconciliation to be contentious and budget time for it. On-time delivery is the usual flashpoint, because one side measures against the original promise date and the other against the most recent revised date, and the gap between those two definitions can be twenty points. Settle it explicitly rather than letting each site keep reporting its own number.
What should not be integrated in the first ninety days?
Anything that changes how the customer buys, how the product is built, or where it is built. Those three categories contain most of the synergy case and all of the irreversible damage, and none of them should move before the diagnosis is finished.
The reason is not caution for its own sake. It is that the acquired business is frequently good at something the acquirer did not buy it for and does not yet see. That capability is usually embedded in exactly the areas the synergy case wants to consolidate: a service line the acquirer reads as legacy, a customer relationship that looks small on revenue and is large on margin, a production method that looks inefficient and is actually the reason a difficult product works.
The acquired business is frequently good at something the acquirer did not buy it for and cannot yet see, and it is usually embedded in whatever the synergy case wants to consolidate first.
There is one exception. If the acquired business is consuming cash fast enough to threaten the deal, purchasing and working capital move immediately, alongside the diagnosis rather than after it. Label that honestly as a cash decision, not as integration, so it does not become the precedent for moving everything else early.
The practical way to hold the line is to write the deferred list down, with dates, and to share it with the people pushing for speed. A deferral that is documented and scheduled reads as sequencing. A deferral that is never mentioned reads as avoidance, and it invites exactly the kind of unilateral early move the sequence is meant to prevent.
How do you sequence by reversibility?
Rank every integration move by how expensive it would be to undo, then execute in that order, cheapest to reverse first. Reversibility is a better sequencing criterion than synergy size because it fails safely when your understanding of the business turns out to be incomplete.
Shared treasury, consolidated insurance, combined audit, and common payroll processing are all effectively free to unwind and invisible to customers and to the floor. They go first. Purchasing consolidation on commodity inputs is nearly as reversible, provided you do not exit long-standing supplier relationships that carry technical support you have not yet valued.
At the other end sit plant closures, product transfers between sites, ERP cutovers, and brand consolidation. Each of those is expensive or impossible to reverse, each depends on understanding you will not have for months, and each is where the standard playbook tends to want speed.
The detailed comparison of the three orderings people actually choose between is in what to integrate first after a manufacturing acquisition. The short version is that the back office earns its place first on this criterion and on every other one.
Reversibility also gives you a fair way to settle disputes between functions. When two teams disagree about what goes first, the question stops being whose synergy is larger and becomes which move would be cheaper to undo. That is an answerable question, it does not depend on anyone’s forecast, and it tends to produce the same ordering regardless of who is in the room.
Build the list once, early, and keep it visible. Three columns are enough: the move, how long it would take to reverse, and who would notice if it went wrong. Most integration arguments in months two through six are really disagreements about one of those three columns, and having them written down converts a political conversation into a factual one.
When do you touch the customer-facing side?
After you know what each side sells well and why, which in an industrial business takes a full quarter of order-book work. Commercial integration done early looks efficient and quietly moves revenue from the accounts that were producing it.
The move that causes the most damage is reassigning coverage. An industrial salesperson’s value is frequently specific: they know a plant engineer at a particular customer, they know which specification that customer actually cares about, and they know the history of the last failure. Reassign the account to somebody covering a broader territory and the relationship does not transfer, it thins.
Pricing is the second. Harmonizing two price lists is a legitimate synergy and it is also the fastest way to lose the accounts that were priced well for a reason. Before harmonizing, find out what each price list was doing: which products it was steering customers toward, and whether the steering was working.
The sensible window for commercial integration is months four to twelve, after the order books on both sides have been read properly and after the operating side has been stabilized. Even then, move coverage last and move systems, quoting, and reporting first.
There is one commercial move worth making early, and it is informational rather than structural. Have both sales organizations submit their lost quotes for the past year, with the reason each was lost. That exercise costs nothing, changes nobody’s coverage, and routinely surfaces the real cross-selling opportunity, which is usually narrower and more specific than the one in the deal model.
Sequence within the commercial phase as carefully as you sequence the phases themselves. Quoting systems and pipeline reporting first, pricing analysis second, price changes third, and coverage changes last. Each step in that order gives you information that improves the next one, and only the last step is difficult to reverse.
When do you touch the plants?
Last, and only with a specific product-level case rather than a capacity-level one. Footprint decisions are the highest-value and highest-risk moves in an industrial integration, and they are the ones most often made on the wrong evidence.
The wrong evidence is aggregate utilization. Two plants at sixty percent utilization look like one plant at a hundred and twenty percent, and on a spreadsheet that is a closure. On the floor it frequently is not, because the sixty percent at each site is made of different products on differently configured lines, and the receiving plant cannot run half of what it is being asked to absorb without tooling, qualification, and a learning curve nobody costed.
The right evidence is product by product: which specific parts move, onto which specific lines, with what tooling, at what yield during the learning period, and with what qualification requirement from the customer. That analysis takes months and it is worth every week, because a transfer that fails costs the customer relationship as well as the money.
When a plant does move, the useful move is often not the one in the synergy case. In a shopping cart business inside Berkshire Hathaway, a new-cart operation in Virginia closed and the site became the East Coast remanufacturing hub. The footprint changed, the site stayed, and the capability that mattered was preserved rather than transferred.
When the analysis does support a transfer, run it as a pilot before a commitment. Move one product family, qualify it properly, and watch yield and on-time delivery through a full quarter. The pilot costs a few months and it converts the biggest irreversible decision in the integration into a reversible one, which is worth far more than the time it takes.
Customer qualification deserves a specific mention because it is the requirement most often missed. In many industrial markets a customer has approved a part at a named site, and moving production triggers a requalification that can take months and can be refused. Check the qualification position before the transfer analysis, not after it.
What about systems and ERP?
Later than the systems team wants and earlier than the plants want. An ERP cutover consumes the management attention that months four through twelve need for operating decisions, and it is close to irreversible once started.
The practical answer in most industrial integrations is to leave the acquired site on its own system through the first year, connect the two for reporting purposes only, and plan the cutover for after the operating changes have been made rather than during them.
That ordering annoys everyone. Finance wants a single system for reporting, and the argument is fair. The counter is that a reporting bridge can be built in weeks while a cutover takes a year and owns the calendar of every senior person in the acquired business for the duration.
If the acquired system is genuinely failing, that changes the answer. A system that cannot support the operating plan is itself a constraint, and it moves up the list accordingly.
Whatever the timing, decide the target state early even if you execute it late. Ambiguity about which system wins produces a year of parallel investment, as each side keeps building on a platform that may be retired. Naming the destination in month three and scheduling the journey for month fifteen costs nothing and stops that waste immediately.
The reporting bridge is worth describing because it is what makes the delay tolerable. It is a monthly mapping from the acquired company’s chart of accounts into yours, built once and maintained by one person. It is imperfect, it requires judgment at the edges, and it delivers ninety percent of what a single system would deliver for a fraction of the disruption during the year you most need the attention elsewhere.
How do you keep the people who hold the knowledge?
Identify them in the first month, by asking the floor rather than reading the org chart, and give them a reason to stay that is specific to them. In an industrial business, the knowledge risk is concentrated in a small number of people, and none of them is necessarily senior.
The scheduler who knows which combinations of orders run well together. The maintenance lead who knows which machine drifts and how to hear it. The application engineer who knows why the difficult customer’s product is specified the way it is. Those three can be worth more to an integration than a vice president, and retention packages rarely reach them.
Find them by walking and asking. Who do people call when something goes wrong at two in the morning? Who trains the new operators? Whose absence changes the schedule? The answers converge quickly and they will not match the salary bands.
Then be concrete. Most of these people do not leave over money, they leave over uncertainty and over the sense that an acquirer has decided their site does not matter. Telling them specifically what the plan is, and that they are in it, does most of the work.
The same logic applies to the first departures. When someone with deep knowledge does leave, treat it as an urgent operational event rather than a human resources one. Get the knowledge documented, get them to train a successor if they are willing, and find out honestly why they went. The reason is usually informative about what the integration is doing wrong while there is still time to change it.
What does month six look like?
Back office consolidated, one set of numbers running, commercial work underway, and the first operating changes visible in leading measures. By month six the integration should have produced something the business can feel, and it should have come from the reversible half of the list.
The leading measures matter more than the EBITDA line at this stage. On-time delivery, quoted lead time, win rate on the deals where delivery decides, scrap and yield at the constraint operation. Those move months before profit does, and they tell you whether the diagnosis from month one was right.
Month six is also the honest checkpoint on the synergy case. You now know enough to say which of the original synergies are real, which are smaller than claimed, and which turned out to be sitting on top of a capability worth keeping. Revising the case at month six is a sign the process is working.
Month six is also the right moment to ask the acquired organization how the integration feels from inside. Not in a survey, in conversation. What has been clumsy, what has been slower than it should be, and what decisions are stuck. The answers arrive faster than any reporting line will surface them, and at month six there is still time to act on all of them.
If the leading measures have not moved by month six, resist the instinct to add workstreams. The far more likely explanation is that the diagnosis from month one was wrong, or that the work is sitting behind a decision nobody has made. Both of those are cheaper to fix than a new initiative, and both are found by asking rather than by adding.
What does month twelve look like?
Commercial integration largely finished, footprint decisions analyzed and either committed or explicitly deferred, and the capacity or lead time investments in progress. Year one ends with the irreversible decisions understood rather than executed.
That last distinction is the one acquirers most often get wrong in both directions. Some execute the irreversible moves in month four on a thesis nobody has tested. Others defer them indefinitely because the analysis is uncomfortable, and the deal quietly becomes two companies with a shared owner.
By month twelve you should have the product-level footprint analysis in hand and a decision written down, even if the decision is to keep both sites and specialize them. Specialization is frequently the better answer and it almost never appears in an original synergy case, because a synergy case counts sites rather than capabilities.
Specializing two sites rather than closing one deserves more consideration than it usually gets in an industrial deal. It preserves the local capability that took years to build, it removes duplicated product ranges, and it frequently captures most of the cost benefit of a closure without the transfer risk. It is also much easier to reverse if the demand picture changes.
Where you defer, say so explicitly and attach a date and a trigger. A footprint decision held until the capacity investment is commissioned is a sequencing choice. The same decision held with no date attached becomes an open item, and open items in integrations quietly turn into permanent duplication.
What should be finished by month eighteen?
Every reversible integration, the commercial model, and the physical changes you committed to at month twelve. Eighteen months is the realistic completion horizon for an industrial integration, and plans written to a shorter one usually finish late and loudly.
What eighteen months gets you that twelve does not is a full cycle of evidence on the physical changes. A product transfer completed in month thirteen is producing trustworthy yield data by month eighteen, not before. A capacity addition commissioned in month ten is filling by month eighteen. Those are the checkpoints that tell you whether the deal is working.
If something is still open at month eighteen, name it and give it an owner rather than letting it run. Open integration items have a way of becoming permanent features, and the cost shows up as two of everything for the rest of the hold.
Write a short closing memo at month eighteen: what was integrated, what was deliberately left separate and why, and what the combined business can now do that neither side could. That document is worth having when the next acquisition arrives, because the second integration in a company is usually run by people referring to their memory of the first one.
Close the integration formally when it is done. Stand down the workstreams, end the separate reporting, and move whatever remains into the normal operating cadence. Integrations that are never formally closed keep consuming management attention through meetings nobody can quite cancel, long after the work that justified them has finished.
How do you know the integration worked?
The combined business can do something neither side could do alone, and you can name it. Cost synergies are the easy test and the weaker one, because cost comes out whether or not the two businesses ever genuinely combined.
The stronger test is capability. A product the acquired plant could not build at volume and now can. A customer neither side could serve on lead time and now both can. A price list that steers demand toward what the combined footprint makes best. Those are the results that show two businesses actually became one, and they only arrive if the sequence protected the capabilities long enough to find them.
Run one more check at month eighteen, honestly. Ask the people who were at the acquired business on day one what got better and what got worse. The list of what got worse is where your next twelve months of work is, and it is almost always shorter and more specific than anyone expects.
Then set the measures that will carry past the integration. On-time delivery, quoted lead time, win rate on the deals where delivery decides, and flow-through on any revenue growth. Those four tell you whether the combined business is genuinely better than its parts, and they keep telling you long after the integration workstreams have been closed down.
Be honest about the answer even when it is mixed. Most industrial integrations capture part of the synergy case, miss part of it, and find value that was never in the model at all. A clear-eyed account of all three is what makes the next deal better, and an account that simply confirms the original case is usually the least informative document produced in the whole eighteen months.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B+ shareholder value across five turnarounds at Berkshire Hathaway, Illinois Tool Works, and Whirlpool. He is 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. His newest book is 10-Minute Transformation (February 23, 2027).

