How a Device Startup Wins Its First Hospital (Illustrative Case Study)

1. Case Studies

By Saroj Jha · 7 min read

A fictional device company has a cleared dressing, one surgeon champion and no buying path. An illustrative twelve-week plan to win one named hospital, with claims, evidence and price that committee will ask for.

Example Co. is a fictional medical device company, and this is an illustrative sample engagement: the numbers are modelled projections, not client results. It shows how AAJ would approach a problem many device founders have, a cleared product, one enthusiastic surgeon and a supply chain office that never replies.

In my view, a small device company should not try to “sell to the hospital”. It should win one named committee at one named hospital, with the claims, evidence and price that committee will ask for, and write down how long each step took. In this example the plan targets a 300-bed community hospital in a six-hospital system, because a win there is also the first step into the other five.

Snapshot

The situation

Example Co. had a 510(k)-cleared dressing, a surgeon at a community hospital who liked it, and a list of supply chain directors who had not answered its cold emails. The team of 11 was working on five kinds of buyer at once. It could not say which contract covered the dressing at its champion’s hospital, who would decide, what that group would ask for, or when.

Phase 1: Diagnose (weeks 1 and 2)

The aim is to find out how this one hospital buys, from the people who do the buying.

• Which contract. Ask the champion and the materials management office which group purchasing organization the hospital uses for dressings, and which distributor delivers. A 2026 research letter in JAMA Health Forum found that, in 2024, three group purchasing organizations covered 84.1 percent of 4,760 hospitals, so the answer is usually one of a few names, but the hospital has to confirm it^[1]. GAO’s 2014 report quotes a trade association estimate that about 96 to 98 percent of U.S. hospitals buy through such contracts^[2]. A group contract is a route to the hospital, not an order for the product.
• Who decides. The surgeon is the champion, not the decision-maker. Map the value analysis committee, the perioperative nurse leader, infection prevention and supply chain, and note who the system, not the hospital, controls.
• What the hospital uses now. The incumbent dressing, its unit price and the volume of closures a year. Without these the value case is a guess.
• How long it takes. No public source gives a typical committee cycle, so the plan records the real dates from the first conversation.

Diagnosis: the product was not the problem. The team did not know the buying path at one site, so it could not prepare what that site would ask for.

Phase 2: Design (weeks 3 to 6)

The aim is to prepare the claims, the evidence and the offer before anyone asks.

1. Claims that match the clearance. The FDA decides a device’s intended use from labeling, advertising and the statements of the company and its representatives, not only from the clearance letter^[3]. The website, deck and sales team are checked against the cleared indication, and the status is written as “FDA cleared”, since clearance is not approval^[4].
2. A short evidence plan. Bench data and the surgeon’s own case series are labelled as what they are. A comparative study is costed before it is promised, because the committee will ask whether the dressing lowers dressing changes, dehiscence or time in the operating room, and a claim of that kind needs substantiation.
3. A value analysis pack. One page on the problem, the clearance, the evidence, the unit price against the incumbent and what the hospital would measure in a trial. The hospital’s own numbers go in the comparison.
4. An evaluation agreement. A 60-day trial in the operating room with a stated unit price, a named owner, the measures agreed in advance and a decision date.

Phase 3: Execute (weeks 7 to 12)

The aim is to take one hospital from trial to a standing order, and record the dates.

The dressing is a low-cost, single-use item, so the hospital’s question is less whether it can afford the dressing than what it has to change and who answers if something goes wrong. In my view the pack should answer that second question first.

AAJ resources used: the MedTech Marketing Playbook (Steps 2, 3, 5, 8 and 9), the MedTech Claims Checker, the Value Analysis Pack Template and the Evaluation Agreement Template.

Projected results, 12 months after launch

These are projections, not historical results. They use a price of $38 a dressing and an assumed 1,200 closures a year at the first hospital, so the first site is worth $45,600 a year. If all six hospitals adopted at a similar rate the total would be about $273,600 a year. On a real project the volumes would come from the hospital’s own records.

What has to go right

This is AAJ’s arithmetic on the example. The first hospital’s $45,600 a year is 1,200 × $38. The company needs about 10 such sites to reach the goal of ten approved hospitals, and at that size they would bring in about $456,000 a year, which would not fund a team of 11. In my view that is the real finding: at $38 a unit the plan works only if the six-hospital system adopts together, or if the clinic reorders and the price move the total up. A one-site win is a proof point, not a business.

What carries over to any device company

• Win one named committee first. Learn the buying path at one site, and the next site is faster.
• Keep claims inside the clearance. The website and the sales deck are part of the intended use.
• Bring the hospital’s numbers. The value case is the hospital’s unit price and volume, not yours.
• Write down the dates. No public source tells you how long a committee takes; your own record will.

Method and limits

Example Co. is fictional and every figure about it is illustrative; $38 a dressing, 1,200 closures a year and a six-hospital system are assumptions. The outside sources describe the market and the rules in general; none shows what a particular hospital will do. This is general information, not legal or regulatory advice.