Latest / The Acquisition Talk with Fexingo: Mergers, Buyouts, and Business Sales for Operators / Why Carve-Out Deals Are Harder Than Whole-Company Acquisitions
Transcript
- Lucas: Most people think a company sale is a company sale. You buy the entity, you get everything inside it. But a surprisingly large share of M&A doesn't work that way. It's a carve-out — you're buying just a division, a business unit, a set of assets that has never operated on its own. Luna: And that sounds like a minor logistical detail, but it's actually the source of a lot of failed deals. Lucas: Exactly. A carve-out is fundamentally harder than buying a standalone company because you have to separate something that was never designed to be separate. Think about shared services alone — a division might use the parent company's HR, its IT infrastructure, its payroll, even its physical office space. The buyer has to either replicate those services or negotiate a transition service agreement — a TSA — to keep using them temporarily. Luna: And the TSA itself can become a crutch that extends for years if the buyer doesn't stand up its own capabilities fast enough. Lucas: Right. And that's just the operational side. There's also the cultural challenge — employees in a carve-out often feel like they're being sold off, not acquired. They may lose the brand, the career ladder, the support system of a large parent. Retention becomes a real issue. So you're not just separating systems; you're separating people. Luna: Can we anchor this with a concrete deal? I think listeners would benefit from a real example. Lucas: Sure. Let's use IBM's sale of its healthcare data and analytics assets — what was called Watson Health, more or less — to Francisco Partners in 2022. That deal closed in early 2022, but the process started much earlier. IBM was selling a collection of products, data sets, and client relationships that had been built inside a giant technology company. None of it had ever operated independently. Luna: And the reported price was around $1 billion, which for a private equity firm like Francisco Partners is a sizable bet on a turnaround. Lucas: It was. And the complexity shows up immediately. First, there's the data itself — IBM had aggregated massive healthcare datasets, but those datasets were intertwined with other IBM businesses, like its cloud platform and its research division. The buyer needed to know: what exactly am I buying? Is it the software, the client contracts, the data licenses, or all of it? That's the scope definition problem. Luna: And that scope definition is often the first place deals fall apart. Sellers want to carve out only the crown jewels, but buyers want the full operating unit. Lucas: Exactly. In the IBM case, the assets included health imaging software, population health management tools, and a clinical data registry. But those products ran on IBM's internal infrastructure. So Francisco Partners had to negotiate a TSA for cloud hosting, for technical support, for even basic things like email and procurement. Every line item becomes a negotiation. Luna: And there's a ticking clock too. TSAs typically have a term — eighteen months, two years — after which the buyer must be fully self-sufficient. If they miss that deadline, it's either a costly extension or a scramble. Lucas: And the cost of standing up your own infrastructure can be enormous. One estimate I've seen is that for a carve-out of a business with, say, $500 million in revenue, the IT separation alone can cost $50 to $100 million. That's not a trivial number. It eats into the returns the buyer projected at deal signing. Luna: It's also a test of the buyer's operational capability. A financial sponsor like Francisco Partners might be great at underwriting deals, but do they have the operational team to manage a nine-figure IT separation? Lucas: That's the key question. And it's why we've seen a rise in what are called 'operational carve-out specialists' — consultancies like Alvarez & Marsal, or even dedicated teams within the big four accounting firms, that come in to manage the separation process. In the IBM deal, I know Francisco Partners brought in experienced healthcare tech operators to run the business post-close, not just financial executives. Luna: And this is where many carve-outs fail — not on the financial terms, but on execution. The legal work gets done, the price gets paid, and then the buyer realises they can't actually run the business without the parent. Lucas: Yes. Data from Bain & Company shows that carve-out deals underperform whole-company acquisitions by a measurable margin in the first two years post-close. Revenue growth tends to slow as customers get nervous about the transition, and costs spike as the buyer builds duplicate functions. The real value creation often doesn't show up until year three or four. Luna: And that's assuming the carve-out was clean. What if the seller didn't keep clean financial records for the division? That's another common problem — the business unit never had a standalone P&L. Lucas: Huge issue. In many large corporations, the division's financials are allocated costs from shared services — a percentage of the corporate legal budget, a slice of the CEO's compensation, an IT charge based on headcount. The buyer has to reconstruct what the actual cost structure would be as a standalone entity. That takes months of forensic accounting. Luna: And that reconstruction often reveals that the division was less profitable than it appeared under the parent's umbrella, because it was benefiting from scale that disappears once it's carved out. Lucas: Exactly. The classic example is procurement. A division of a Fortune 500 company gets discounted rates on everything from office supplies to cloud computing because the parent buys in bulk. As a standalone, those discounts vanish. So the buyer's margin projections get compressed. And that's before you even factor in the cost of the separation itself. Luna: Let me ask you this — are there any carve-outs that actually succeeded notably? Because the picture you're painting is pretty grim. Lucas: There are. One of the most studied is the 2014 carve-out of PayPal from eBay. That was actually a spin-off rather than a sale, but the separation challenges were similar. The two companies had deeply integrated payment and marketplace systems. eBay had to build a whole new payments infrastructure after the split. And PayPal had to create its own customer service, fraud detection, and risk management functions. It was painful for about a year, but then both companies thrived. Luna: Right, but that was a spin-off to shareholders, not a sale to a financial buyer. The incentives are different — eBay wanted PayPal to succeed because eBay shareholders became PayPal shareholders. Lucas: That's a crucial distinction. In a private equity carve-out, the buyer's timeline is typically five to seven years. They need to improve margins and grow EBITDA quickly to achieve a target return. That puts pressure on the separation timeline in a way that a spin-off doesn't. So the operational complexity is magnified. Luna: And the seller's incentives also matter. IBM, for example, had already written down Watson Health by billions. They were eager to get rid of it. That can lead to a less cooperative separation process — the seller doesn't want to spend a lot of money helping the buyer stand up a competitor. Lucas: That tension is real. The TSA is supposed to bridge it, but there's always a conflict: the seller wants to exit the business cleanly and stop spending on it; the buyer wants a long, cheap TSA. So you get fights over service levels, pricing, and scope. I've seen TSAs that run to hundreds of pages, covering everything from how quickly the seller must respond to an IT ticket to whether the buyer can hire the seller's employees. Luna: And speaking of employees — in the IBM deal, one of the big questions was whether key data scientists and engineers would stay. When you're selling a data-driven business, the talent is the asset. Lucas: Exactly. Retention packages, earnouts, equity in the new entity — those are common tools. But they're harder to design in a carve-out because the employees don't necessarily know the buyer. There's uncertainty. And if the seller is also laying off people elsewhere, morale can be terrible. Luna: There's a great case study out of MIT's Sloan School about a carve-out in the industrial sector where the buyer actually lost 40 percent of the division's engineers in the first six months. The deal never recovered. Lucas: That's the nightmare scenario. And it's why sophisticated buyers do what's called a 'clean room' process before close — a virtual data room and a set of protocols that let the buyer's team start planning the separation in detail without crossing antitrust or confidentiality lines. They map out every system, every contract, every employee, every customer. They create a day-one plan that goes into effect the moment the deal closes. Luna: And that planning takes months. It's not uncommon for a carve-out to have a six-to-nine-month pre-close integration planning phase, even before the regulatory approvals. Lucas: Absolutely. And that planning cost is real — it's not just lawyers and bankers; it's operations consultants, IT architects, HR specialists. For a $1 billion deal, the pre-close planning can easily cost $10 to $20 million. And that's all at risk if the deal falls through. Luna: It sounds like carve-outs are really only for buyers who have deep operational expertise and a long time horizon. I wonder if the podcast's audience, many of whom are operators themselves, can take any lessons from this. Lucas: I think so. If you're a business owner or a manager at a company considering buying a division, the lesson is: don't underestimate the separation cost. Triple-check the TSA. Visit the shared services in person. And prepare for the cultural shock. The financial model always looks better on paper than in reality, because paper doesn't capture the friction of untangling two organizations that grew together for decades. Luna: And for sellers, the lesson might be: if you're planning to carve out a division, start running it as a standalone business at least a year before you sell. Clean up the allocated costs, separate the IT systems, give the division its own P&L. You'll get a higher price because the buyer sees less risk. Lucas: That's a great point. Some of the best-prepared carve-outs I've seen were actually preceded by an internal spin-off — the parent creates a wholly owned subsidiary, transfers the assets, and runs it independently for six to twelve months. Then they sell the subsidiary. The buyer is essentially buying a standalone company, not a bundle of shared dependencies. Luna: That approach also helps with employee retention, because the division gets used to its own identity before the sale. Lucas: Right. And it makes the TSA simpler, because the division already has its own IT, its own HR, its own procurement. The separation cost drops significantly. Luna: So in a way, the best carve-out is the one that doesn't feel like a carve-out at all. Lucas: That's the ideal. And it's rare. Most sellers don't have the time or the will to do that prep. They want to sell quickly, and the buyer inherits the mess. But the buyers who succeed are the ones who go in with eyes wide open, budget for the separation, and staff it with operators, not just dealmakers. Luna: And if you're someone considering a divestiture or acquisition, these are the kinds of insights that can save you millions. It's why we dig into these mechanics episode after episode. Lucas: And if these conversations have helped you think more clearly about your own deals — whether you're buying, selling, or just advising — there's a small way you can help us keep this show ad-free and independent. It's not a big ask, but it makes a real difference to us. Luna: Yeah, we don't run ads. We want to keep the conversation focused on what actually matters in M&A. If you find value in that, you can support the show at buy me a coffee dot com slash fexingo. Lucas: And that support goes straight into research time, better guests, and avoiding any sponsor influence. So if that resonates, we'd be grateful. Luna: Alright, back to carve-outs. One more angle I want to touch on: how do you value a carved-out business differently from a standalone one? Lucas: Good question. The standard approach is to build a standalone P&L, but the valuation multiples are often lower than for comparable standalone companies. The discount can be 10 to 20 percent, reflecting the execution risk. A buyer might apply a higher discount rate in their DCF, or haircut the EBITDA multiple. The uncertainty is priced in. Luna: But some buyers see that discount as an opportunity — if they can execute the separation well, they can create value by buying at a carve-out multiple and eventually selling at a standalone multiple. Lucas: Exactly. That's the private equity playbook. Buy a division at 6 times EBITDA because it's messy and undermanaged, separate it, improve margins, and sell three years later at 10 times EBITDA as a pure-play company. The multiple expansion is the return driver. Luna: But it's not automatic. You need the operational chops. Lucas: You do. And that's why we're seeing more and more PE firms build in-house operating teams specifically for carve-outs. They're not just financial engineers anymore. They're hiring supply chain experts, IT integration specialists, even HR transformation leaders. The firms that do carve-outs well are the ones that treat it as an operational challenge, not just a financial one. Luna: And for the rest of us, it's a reminder that the deal doesn't end at closing. In many ways, that's when the real work begins. Lucas: Absolutely. Thanks for listening to The Acquisition Talk. We'll be back next week with another deep dive into the mechanics of M&A.