Reading an acquisition target through its digital footprint

A hospital data room tells you what the seller has chosen to show: audited accounts, occupancy, licences, doctor contracts, payer empanelments, an equipment register. It is a picture of supply and of history. What it does not show is where the demand comes from, whether it is growing or draining away, how much of it belongs to three doctors who may not stay, and what the catchment actually thinks of the place. All of that is visible from outside, for free, in about a week, to anyone who knows where to look.

The first time I was asked to contribute to an acquisition assessment, the request was for a view on the target’s website. I gave a much longer answer than anyone wanted, and most of it turned out to matter more than the website. The digital footprint of a hospital is the demand side of the diligence, and it is the part the growth leader is uniquely placed to read.

This is the checklist I now run. It fits inside the diligence window, it needs no cooperation from the seller, and it produces findings that change price, structure and the post-close plan.

Why the footprint is worth reading

Three reasons. It is independent of the seller: reviews, listings and search data cannot be tidied for a buyer the way a management presentation can. It is longitudinal: the data room gives you three years of accounts, but search trends, review velocity and listing history run further back and update monthly. And it shows the future rather than the past: a hospital whose search share is falling and whose reviews have turned will show it in the footprint a year before it shows in revenue.

The footprint also reveals the things a seller is least likely to volunteer. Doctor dependence. Aggregator dependence. Paid-search dependence. A reputation incident that the accounts do not carry. Each of these is a valuation question, and each is visible from a browser.

Day one and two: entity and listings

Start with the most basic question: what is this hospital called, and does the internet agree?

Search the name, the address and the phone number separately. Count the maps listings. A well-run single-site hospital has one. A poorly-run one has three or four — an old name, a department listed as a separate business, a doctor’s clinic that shares the address — and each one is splitting reviews, confusing search and, after you buy, becoming your migration problem. Check whether the main listing is claimed and by whom; an unclaimed listing means nobody has been managing the front door, and a listing claimed by a former marketing agency means a recovery exercise before you can touch it.

Then the aggregators. Every doctor aggregator and hospital directory that matters in that city: is the hospital listed, is the profile current, are the doctors listed under this hospital or under their own clinics, and does the aggregator show booking or enquiry volume? A hospital heavily present on aggregators, with its doctors’ profiles pointing to the aggregator’s booking rather than the hospital’s, is a hospital paying commission on demand it should own. That is a margin finding, and it will not be in the data room.

Finally the domain and the site: who owns the domain, how old it is, what else is hosted on it, whether there is analytics you will inherit or a blank slate, and whether the site is doing any work at all in search. Note the phone numbers on the site and on every listing and check whether they match. Mismatched numbers mean enquiries are leaking to lines nobody manages.

Day two and three: reviews

Reviews are the closest thing to a patient-experience audit you can run without entering the building. Read them properly, not the score.

• Volume and velocity. How many reviews, over what period, and has the rate changed? A hospital that gained most of its reviews in one quarter has run a campaign or bought them; a hospital whose review rate has halved has lost either patients or the person who used to ask for reviews.
• Sentiment by department. Code a sample by specialty and by theme — billing, waiting, doctor behaviour, nursing, cleanliness, discharge. A strong cardiac line and a weak emergency department show up here before anywhere else.
• Named doctors. Count how many reviews name a specific doctor and which doctors. If a large share of the positive reviews are about two people, the hospital’s reputation is those two people’s reputation. Cross-reference against the doctor contracts in the data room.
• Response pattern. Does anyone reply, in what tone, and how fast? An unanswered stream of billing complaints tells you about the management culture you are buying.
• Suspicious bursts. Clusters of short five-star reviews with similar phrasing on the same dates. If they are there, the score is not the score, and the seller’s brand claims should be discounted accordingly.
• The incident. Search the hospital’s name with the words a patient would use when something has gone wrong, in English and the local language. One serious incident that reached the press or social channels can sit under a reasonable average score and still define the hospital in its catchment.

Day three and four: search share and demand trend

Pull search demand for the target’s catchment by specialty and procedure, in every relevant language, for as long a history as your tools allow. Then estimate the target’s share of it: branded search for the hospital name, doctor-named search for its doctors, and organic visibility for the generic procedure terms.

What you want is the trend. A hospital whose branded search has been flat for three years in a city where the category is growing is losing share, whatever the occupancy sheet says. A hospital whose doctor-named search exceeds its branded search is a doctor practice with a building attached. A hospital with no organic presence for the procedures it says it specialises in is either invisible or buying all its demand — and the paid-search footprint, visible from the auction and from the ad transparency tools, will tell you which.

Compare against the two or three nearest competitors. Share is only meaningful relative to someone. If the target is holding share against a strong set, that is worth something. If it is holding share because the competitors are weaker still, the market may be the problem.

Day four and five: doctor presence

This is the finding that most often changes a deal. In the Indian market, the demand you are buying is frequently attached to people, not to the institution, and the footprint shows you exactly how much.

For every senior doctor listed in the data room, check: their page on the hospital site and its traffic if you can estimate it; their aggregator profiles and which hospital or clinic those point to; their own site or social presence and whether it names the target hospital; and whether they appear on a competitor’s site as visiting faculty. A doctor whose aggregator profile books into their own clinic and whose social presence never mentions the hospital is not the hospital’s doctor in any demand sense. They will take their patients with them, and a change of ownership is when they decide.

Build a concentration table: the share of doctor-named search, of positive reviews and of listed specialties that sits with each of the top five doctors. If that table is steep, the deal is a doctor retention deal, and the earn-out, the contracts and the post-close plan should say so. This is where you work alongside the medical director on the buying side; the footprint gives you the list of who matters, and the clinical conversation decides what to do about it.

Day five: the front door, mystery-shopped

Call the hospital. Use the number on the listing, the number on the site and the number on the aggregator profile. Enquire about a common procedure in the local language and in English. Note how long it takes to be answered, whether the person can name a doctor and a price range, whether they offer an appointment, and whether anyone calls back. Then submit an enquiry on the site and count the hours until a response.

This tells you whether the target has a contact centre or a phone. It tells you whether a CRM exists in any meaningful sense. And it tells you what the conversion rate on the demand you are buying is likely to be, which is a number you will otherwise get from the seller’s management presentation, if at all.

What the footprint reveals that the data room does not

• Demand trend, running ahead of revenue by a year or more.
• Doctor dependence, quantified rather than asserted.
• Channel dependence — on aggregators, on paid search, on one referral source — that the accounts show only as a marketing line.
• Reputation trajectory and any incident that the catchment remembers.
• The state of the front door, and therefore the size of the post-close digital rebuild.
• Entity mess — duplicate listings, unclaimed profiles, agency-held accounts — that becomes your migration cost.

Each of these has a place in the deal. Demand trend and reputation go to price. Doctor dependence goes to structure — contracts, earn-outs, retention terms. Channel dependence and entity mess go to the post-close budget and the integration plan. Put them in those places explicitly. A finding that stays in a digital appendix changes nothing.

What the footprint cannot tell you

Be honest about the limits in the room. The footprint does not show payer mix, and a hospital with strong demand and a poor payer mix is a poor business. It does not show clinical quality except as patients perceive it, and patients perceive the waiting room more accurately than the operating theatre. It does not show what the doctors’ contracts actually say. And search share in a Tier 2 city with low digital penetration understates the walk-in and referral demand that never touches a screen. The footprint is the demand side of diligence, not the whole of it, and presenting it as more than that will cost you credibility with the finance and clinical people whose findings sit alongside yours.

The checklist, in a week

1. Days one and two: entity and listings. Count the maps pins, check claim status, audit aggregator presence and doctor profile routing, establish domain and account ownership, reconcile phone numbers.
2. Days two and three: reviews. Volume and velocity, sentiment by department, named-doctor share, response pattern, suspicious bursts, incident search in two languages.
3. Days three and four: search. Catchment demand by procedure, target’s branded, doctor-named and organic share, trend over the longest available history, paid-search footprint, comparison against the nearest competitors.
4. Days four and five: doctors. Presence audit for every senior doctor, concentration table for the top five, flagged for the medical director.
5. Day five: mystery shop. Every listed number, every enquiry route, both languages, time to response and quality of response.
6. Day five, afternoon: write it as three findings for price, three for structure and a costed post-close plan for the front door. Nothing else.

The seller controls the data room. Nobody controls the search results, and that is exactly why they are worth reading.

Questions people ask