Every post-acquisition plan hits the same meeting. It’s usually the deal lead, sometimes the CEO themselves, who says the sentence that ends the integration before it starts: “Let’s not disrupt what’s working. Keep them running as their own business unit for now.”
Everyone nods. It sounds careful. It sounds like the disciplined call, the one that respects the team you just bought instead of steamrolling them with someone else’s process on week one.
It’s actually the decision not to finish the deal.
Two water heaters, one listing
A business unit, in the way it actually gets used after an acquisition, is two houses pushed together with a doorway cut through the shared wall. The listing says single-family. Walk the basement and you’ll find two electrical panels, two water heaters, two thermostats arguing with each other over the same hallway.
Nobody planned it that way on purpose. It happened because knocking out the wall, actually combining the systems, is disruptive, expensive, and political, and “keep them separate for now” was always available as the answer that requires nothing of anyone this quarter.
“For now” is the tell. Nobody schedules the follow-through. There’s no calendar invite eighteen months out that says “revisit whether we’re actually one company yet.” The business unit stops being a transition state and becomes the permanent shape of the org chart, because permanent is what happens to anything nobody’s accountable for un-doing.
The six processes don’t care about your org chart
Kinetic PE’s frame for this is the same six processes that run every software company: Lead to Cash, Procure to Pay, Application Lifecycle, Employee Lifecycle, Customer Lifecycle, Corporate Governance. They don’t ask permission to exist twice. If you have two business units, you almost certainly have two versions of at least four of them: two CRMs closing two pipelines, two billing systems producing two invoices for the same logo, two onboarding processes for the same open req, two board slides that get stitched together the night before the meeting instead of pulled from one source.
Each version works well enough on its own. That’s exactly the problem. Nothing forces the reconciliation, because nothing is visibly broken, until someone tries to answer a question that spans both units and discovers the honest answer requires three people, two spreadsheets, and a Slack thread that starts with “does anyone know how BU2 counts this.”
Bigger isn’t healthier
Revenue goes up when you buy a company. Headcount goes up. Logo count goes up. None of that is the business getting healthier. It’s addition, not improvement.
The thesis behind most of these deals assumes something more than addition: one back office instead of two, shared systems that get cheaper per unit the more volume runs through them, one sales motion touching both customer bases instead of two that never talk to each other. Those are the gains the deal was actually underwritten on.
None of that happens while the business unit wall stands. Keeping the units separate isn’t a neutral holding pattern. It’s the thing actively blocking the gains the price assumed. You can’t consolidate a back office that’s structurally split in two. Can’t share a system nobody agreed to share. Can’t run one sales motion across two CRMs that don’t talk to each other.
So the company gets bigger. It doesn’t get better. Bulk isn’t strength. It’s the same unaddressed inefficiencies the target had on day one, now carried in two copies instead of one, dressed up as growth.
What the buyer finds
This is the part that doesn’t show up until the next diligence room. A buyer’s operational team isn’t looking for one clean number. They’re looking for how many systems it took to produce that number. Two Lead to Cash processes stapled together with quarterly reconciliation isn’t one business with strong unit economics. It’s two businesses standing close together, wearing a shared logo.
That’s the same human-middleware problem this firm has written about before, just at the org-chart level instead of the function level. Whatever isn’t actually unified gets quietly staffed by people manually bridging the gap between two systems that were never supposed to stay two systems. That labor is real and it never shows up as its own line item. It just shows up as a person who has to exist for the numbers to tie out. Buyers price that. It shows up in the EBITDA bridge as a discount, not a footnote. The multiple assumes repeatability, and two business units pretending to be one company is the opposite of repeatable.
Count your systems
Here’s the test, and it takes about ten minutes: ask your CFO how many systems it takes to close the books this quarter. Not how many systems exist company-wide. How many it actually takes, today, to produce one number everyone agrees on.
If the honest answer isn’t “one,” you didn’t complete an acquisition. You completed a merger of convenience that’s still pending, dressed up as an org chart decision so nobody has to call it what it is.
Run the same question against Lead to Cash. Against Employee Lifecycle: do the two units even use the same HRIS, or is that also “for now”? If closing the books takes three spreadsheets hand-stitched together at midnight before the board meeting, you don’t have a business unit strategy. You have a business that’s still exactly as unintegrated as it was on day one, just with better PowerPoint.
Business units aren’t always wrong. Some acquisitions genuinely earn a separate operating unit: different market, different buyer, a real strategic reason to keep the P&L distinct. That’s a decision. What isn’t a decision is drifting into permanent separation because nobody scheduled the harder conversation.
The difference between the two is whether anyone can tell you when “for now” is supposed to end.