Buy, build or partner: entering a new care format

Every new-format conversation I have sat in at executive-committee level started the same way. Not with the question of whether the group should be in day-care surgery, or dialysis, or fertility, or home care. It started with a deal. A banker had brought a chain. A founder wanted out. A unit head had met an operator at a conference. The route into the format was decided before anyone had decided the format was worth entering, and the rest of the discussion was spent rationalising that.

That is the first thing to fix. Buy, build or partner is the second decision, not the first. The first is whether the format has a place in the group’s strategy at all — and, if it does, what the group actually needs from it: demand, capacity, capability or a licence to operate. The route follows from that answer. When the order is reversed, you end up owning something because it was available, and defending it for years because you cannot admit it.

I have built the case for entering formats through each of the three routes, and watched the executive committee take each one. What follows is what I would weigh now, having seen what each route looks like eighteen months in.

What the three routes actually buy you

Strip the decks away and the trade-off is simple. Building costs the least capital up front and the most time, and gives you total control and total exposure. Buying costs the most capital, is the fastest to revenue, gives you control of an asset you did not shape, and hands you someone else’s problems on day one. Partnering costs the least capital and the least time, gives you the least control, and puts your brand on an operation you do not run.

Every group I know says it wants speed, control and low capital. You get two. The honest version of the executive-committee discussion is deciding which one you are giving up, and whether the group can live with that for the five years it takes for the format to matter.

Decide what you are actually buying

A format is not a building. It is a different business with different unit economics, and the most common error is assuming the hospital’s economics carry over.

A tertiary hospital earns on complexity — high-acuity cases, long stays, ICU days, the surgeon whose name fills a floor. A day-care surgery centre earns on throughput: short procedures, fast turnaround, a fixed cost base kept small. A dialysis chain earns on chair utilisation and consumable cost. Home care earns on route density and nurse productivity. Primary clinics earn — when they earn — on referral to the mothership and on volume nobody at the mothership would get out of bed for.

So before the route, answer four things. Does the group need the format’s demand, because its own patients are leaking to it? Does it need the capacity, because the flagship units are full of cases that should not be there? Does it need the capability, because nobody in the group knows how to run a throughput business? Or does it need the licence — the empanelments, the approvals, the regulatory position — that takes years to build from nothing?

If the answer is capability, building is the slowest and least reliable route, because you are asking a hospital culture to learn something it structurally does not value. If the answer is demand, partnering may be enough. If the answer is licence, you are buying, and the price is the price.

Build: the slow route, and what the group’s own habits do to it

Organic build looks cheapest and most controllable. It is the route the CFO instinctively prefers, because capital goes out in tranches and the group owns everything.

Three things go wrong. The first is overhead. A small format inside a large group inherits the group’s cost structure — corporate allocations, the shared-services charge, the same quality team, the same IT stack, the same procurement contracts priced for a hospital. A day-care centre that would be profitable as a standalone shows a loss inside the group because it is carrying a share of a corporate office designed for something ten times its size. I have watched a format get killed on a P&L that was mostly allocation. Ring-fence it before the first review, or the numbers will lie.

The second is people. The format needs its own head, and that head needs to be someone who has run a throughput business, not a hospital COO with a side project. The hospital’s best operations person will run the new format like a small hospital. That is precisely the wrong thing.

The third is time. Building means recruiting doctors without the flagship’s pull, winning TPA empanelments from scratch, and taking two or three years to reach the volume that makes the model work. The demand engine can shorten this — the pre-launch playbook that works for a hospital works for a format — but it cannot remove it. If the board expects the format to contribute in year two, build is the wrong route and the case should say so.

Buy: what you pay for and what you inherit

An acquisition buys time. It also buys a patient base, a set of empanelments, leases, licences, a doctor panel and — usually — a founder who is the brand.

The diligence that matters is not the one the bankers run. Ask who the referrals actually belong to. In a great many small chains, the referring doctors refer to the founder, the corporate contracts were signed on the founder’s relationship, and the star consultant has an understanding that is not in any contract. When the founder leaves at the end of the earn-out, the chain you bought is a set of leases.

Ask where the quality gaps are. A chain built for cash patients in Tier 2 cities may never have gone through NABH, may have consent processes that would not survive an audit, and may have an infection-control record nobody has looked at. Every one of these becomes your brand’s problem the day the deal closes, and the medical director will — rightly — refuse to put the group’s name on it until it is fixed. Budget that fix, in money and in months, into the price.

Then there is integration, which is the cost everyone underestimates. Systems, CRM, contact centre, the website, the brand transition, the pricing harmonisation, the doctors’ contracts, the payer contracts renegotiated under the group’s name. This is a year of work by people who already have jobs. The year after the deal is where acquisitions go quiet and the executive committee stops asking, because the number that was promised is now somebody’s integration plan.

A word on earn-outs. They protect you from paying for a business that evaporates. They also make the founder run the business for the earn-out number, not for the group — deferring maintenance, pushing volume, resisting the integration that would depress margins in the short term. Decide before signing which behaviours the earn-out will produce, and whether you can live with them.

Partner: asset-light, and what it actually costs

Partnering is the route that looks free. An operator brings the format and the operating model; the group brings brand, referrals, perhaps space, perhaps capital. Revenue share or management fee. Everyone is happy at the signing.

What you have given away is control over the thing that carries your name. When a patient at the partner-run centre has a bad outcome, the newspaper will not print the partner’s name. When their nurses are rude, the review lands on your listing. When their billing is aggressive, it is your group that gets the complaint on social media. Brand risk in a partnership is total and control is partial, and that asymmetry has to be written into the agreement: standards the partner must meet, the group’s right to audit, the medical director’s veto on clinical matters, and a termination right that does not require proving breach.

Then ask who owns the patient. If the partner holds the records, the CRM data and the relationship, then the demand you generated has built their business, and at the end of the term you have a brand with no patients and they have patients with no brand. I have seen this argued about only after the term ended. Settle it before.

And understand that a successful partnership creates a competitor. The operator that learns your market with your referrals will, at renewal, know exactly what you are worth to them. Build the renewal terms while you still have leverage, which is at signing.

Design the exit before you enter

The question a good board member will ask is not “what if this works” but “what if it does not”. Have the answer for each route, in writing, before the decision.

• Build. The exit is closure or sale. Closure means sunk capex, staff redundancies and a public admission. Sale means finding a buyer for a sub-scale format inside a big group, which is hard. Design the build so the assets are separable — a separate entity, separate leases, separate contracts — so that a sale is possible.
• Buy. The exit is divestment. Ask, before closing, who would buy it back, and at what discount. Structure the earn-out so that a poor year is not also a lawsuit.
• Partner. The exit is termination. What happens to the patients, the data, the branded signage, the staff on the partner’s payroll working in your space, and the non-compete? Every one of these should be a clause, not a negotiation.

An exit designed before entry is the difference between a format that was tried and a format that became a decade-long embarrassment nobody would close, because closing it meant admitting the original decision.

How I put the case to the executive committee

Three columns — build, buy, partner — and the same rows for each: capital over five years, time to first revenue, time to steady state, degree of control, brand exposure, integration cost in people-months, the exit, and the one thing that would make this route fail. Then a recommendation, and the reason the other two lost.

The rows that get the most argument are integration cost and brand exposure, because they are the ones most easily left out. Put a name against integration — whose team, for how long — and the CFO will take the case seriously. Put the medical director’s view on brand exposure in the paper, in their words, and the committee will not be surprised by it later.

The row I would add now, having got it wrong once, is demand. How many of the group’s existing patients need this format, in this city, this year? Not the market size. The group’s own patients, from its own data. The format that lives on the group’s referral base has a very different case from the one that must win a new market, and the executive committee deserves to know which it is being asked to fund.

If you’re starting this next quarter

• Write the one-page answer to “what does the group need from this format” before any target or partner is named. Get it agreed.
• Pull the group’s own referral and leakage data for the format. If you cannot, that is the first project.
• Build the three-column case with the same rows for each route, including integration cost with names against it and the exit.
• Get the medical director’s written view on clinical standards for each route before the paper goes to committee.
• Ring-fence the format’s P&L from corporate allocation for the first three years, whichever route you take, and agree that with the CFO now.
• Name the head of the format. If nobody in the group can run it, that tells you which route to take.

The route you choose will be remembered as a strategy. The exit you failed to design will be remembered as your name.

Questions people ask