Why Hospital Buyers Decide Before They Ever Meet You
by Lisa T. Miller, MHA
Dreamdata reran their B2B buyer journey research this year and quietly reset every benchmark commercial teams have been working against. Eighty-one percent of the buying cycle now happens before a buyer ever talks to a salesperson, up from seventy percent last year. The complexity inside that pre-sale window is accelerating just as fast.
The new numbers:
88 touchpoints in the average buying cycle, up from 76
4 channels involved, up from 3.7
10 stakeholders shaping the decision, up from 6.8
272 days from first impression to closed deal — nine months of you not being in the room (The Wise Marketer)
For anyone selling into hospitals, those figures should change how you think about pipeline. The hospital decision cycle was already the most layered environment in B2B. It has now gotten denser, longer, and almost entirely self-directed in the early phases.
By the time your CFO or VP of Supply Chain takes the call, the decision is largely formed. They have read your competitor’s case studies, asked a peer in their CHIME or HFMA network, watched a webinar from someone in your category, scrolled through your LinkedIn posts, and almost certainly run a few prompts through ChatGPT or Claude to stress test their thinking.
The conversation you thought was the start of the deal is closer to the middle. Often the end.
What 88 Touchpoints Actually Look Like Inside a Hospital
The Dreamdata benchmark is conservative for healthcare. A typical health system evaluation now moves through a clinical champion, an executive sponsor, supply chain, finance, IT, compliance, legal, a value analysis committee, an infection prevention review for anything touching the patient, and increasingly a chief AI or transformation officer. Ten stakeholders barely covers it for any meaningful capital purchase or service line decision.
Those stakeholders are not coordinating. They are gathering information independently, mostly through content, and arriving at the table with their own conclusions already formed. Your job between the first signal of interest and the first call is to make sure the version of you that lives in their heads is the version you would actually want representing the deal.
Most sellers have no idea what that version looks like, because they have never audited it.
What Hospital Leaders Are Carrying Right Now
Every C-suite conversation I have had this year keeps returning to the same set of pressures, and they are all converging at once.
Margins. New CMS Regulations. Workforce.
Seventy-two percent of hospital CFOs report operating margins of two percent or less, and Kaufman Hall’s January data shows the new year began with depressed margins, lower volumes, higher labor spend, and rising bad debt.
The CMS TEAM model went live January 1, 2026 — five years, mandatory, bundled payments across five surgical episodes for selected hospitals — and most participants are not operationally ready.
The HCAHPS overhaul is reshaping how patient experience translates into reimbursement under VBP. Workforce shortages persist that no signing bonus is solving. Cybersecurity exposure that the Change Healthcare attack made impossible to ignore. AI adoption pressure with no clear ROI playbook. Site of service economics moving volume out of inpatient and into ambulatory.
Each one of those is a board-level conversation. Most CFOs and COOs are juggling all seven at the same time while running the day-to-day. They have less time to take vendor calls than they did three years ago, and far more sources of information competing for the time they do have.
This is the environment in which “build a content engine” became the dominant marketing prescription across healthcare technology, devices, and services. It is also the environment in which most of those engines are quietly failing.
Why the Content Engine Model Keeps Failing
The current thinking is - buyers self-educate, so produce more for them to consume. Add posts. Add webinars. Add white papers. Add podcasts. Hire an agency. Spin up a content factory.
The factory optimizes for output, not impact. Cadence, quantity, pieces shipped per quarter, touches per account. All of it gets reported on dashboards. None of it consistently shows up in pipeline.
What hospital buyers want is the opposite of volume. They want deep nusanced insights - new innovative ways to help solve their challenges. They want a vendor who understands bundled payment structure better than their internal finance team does. They want frameworks they can take into their next executive meeting. They want a peer-level take on what hospitals three states over are doing about workforce, HCAHPS, or surgical bundles.
The most expensive content mistake healthcare commercial teams are making right now is producing more material that no hospital leader actually needs.
What High-Value Content Looks Like to a Hospital Buyer
Think about content the way a product team thinks about a software release.
Every feature is built for user impact. Every release teaches the team something that gets folded back into the next one. The product gets smarter over time.
Content for hospital buyers should work the same way. Each piece should earn its place in a CFO’s week. It should tell them something they did not already know, give them a structure they can use, or sharpen them for their next executive conversation.
Three qualities separate content that compounds from content that disappears.
It speaks at peer level. Hospital C-suites do not need vendors explaining HRRP or VBP to them. They need vendors who can engage with the second and third-order implications of policy changes most analysts have not yet surfaced. If your content could have been written by a marketing intern with a search engine, it will be read like one.
It carries a point of view. The most shared content among hospital leaders right now is sharp, opinionated work from people willing to take positions on TEAM readiness, HCAHPS strategy, and where capital should and should not be deployed. Generic best-practice content disappears the moment it lands.
It builds infrastructure rather than calendar fill. Each piece should leave something structural behind that makes the next conversation easier. A framework someone references in a board meeting. A model that gets forwarded internally. A point of view that travels through the CHIME, HFMA, ACHE, and AONL networks where your buyers actually live.
The Question No One Is Auditing: What Are Your Buyers Asking AI About You?
There is a layer of the eighty-one percent that almost no commercial team is measuring, and it is the layer where the most consequential conversations are now happening. Your buyers are not just reading content. They are asking ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews to do the work for them.
The data on this is no longer speculative. Thirty-two percent of B2B buyers now use generative AI tools as much as traditional search when researching vendors.
The behavioral shift underneath those numbers is even more striking. The average Google user runs about four searches a day. The average AI user submits in the neighborhood of thirty-four prompts a day. Your buyers are not asking the AI one question about your category. They are asking it twenty, in conversation, with follow-ups, refinements, and cross-checks. By the time they finish a session, they have something close to a private analyst report. Your name is either in it or it is not.
In healthcare, the prompts have already evolved past “best vendors for X.” Hospital CFOs, supply chain leaders, value analysis chairs, and chief AI officers are asking questions that tie directly to the regulatory and financial pressures they are actually managing:
· Which medical device companies help hospitals reduce costs under TEAM bundled payments?
· What vendors have a defensible ROI story for the LEJR episode under TEAM?
· Who actually understands the HCAHPS digital transition and can help us protect VBP revenue?
· Which RPM platforms are being used by TEAM-mandated hospitals for the thirty-day post-discharge window?
· What companies should a system CFO talk to about workforce technology that does not require a five-year payback?
· Which AI vendors do health systems regret buying, and why?
If you do not know whether your name comes up in those answers, you do not know what is happening in the most important sixty seconds of your buying cycle. And almost no one in healthcare commercial is auditing this yet.
Why This Is Structurally Different From SEO
The instinct most marketing teams have is to treat AI search like another SEO channel. It is not. The mechanics are different in ways that change the playbook.
The answer is the result. In Google, you compete for clicks against ten blue links. In ChatGPT, there are no ten blue links. There is one synthesized answer with a small handful of cited sources. Perplexity averages 7.5 cited domains per answer; ChatGPT averages 4.1. If you are not in those four to seven slots, you are not in the consideration set. There is no page two.
Citations behave differently across engines. Perplexity shows numbered, ranked sources inline on every answer. ChatGPT cites sparingly and sometimes not at all. Claude hedges its recommendations and names brands more cautiously. Google AI Overviews collapse their citations behind a link count. Each engine selects sources from a different mix of signals, which means a brand can be visible in one and invisible in another. Cross-engine averages hide what is actually happening.
Recognition and source authority are two different battles. A buyer can hear your name mentioned in a ChatGPT answer without your domain being cited as the source. The reverse is also true. Winning one does not solve the other. Most teams are not even tracking the difference.
The signals that drive citations are not the signals that drive Google rankings. AI engines reward content that is structurally chunkable, FAQ-shaped, comparison-shaped, schema-marked, and that lives on domains the model already trusts as authoritative in your category. A blog post designed to rank for a keyword is often the wrong shape to be cited in an answer.
Personalization quietly distorts what teams see. When a marketing team prompts ChatGPT from their work account to check whether their brand shows up, they are seeing a personalized answer shaped by their own search history. The buyer asking the same question from a hospital network is seeing something else entirely. If you are auditing without a clean room, you are auditing a flattering mirror.
How to Actually Figure Out What Your Buyers Are Asking
This is the part most teams skip. They jump to optimization before they have done the diagnostic work. The right sequence is reverse first, then audit, then build.
Reverse-engineer the prompt set. Sit down with your sales leaders, your sharpest reps, and one or two friendly customers if you have them. Build a list of fifty to one hundred prompts a real hospital buyer would actually type — not the prompts your marketing team wishes they would type. Group them by buyer role (CFO, COO, VP Supply Chain, value analysis chair, chief AI officer, clinical champion), by buying stage (problem-aware, solution-aware, vendor-aware, decision-stage), and by the pressure that triggered the search (TEAM readiness, HCAHPS exposure, workforce, cybersecurity, capital allocation). The prompt set is the audit instrument. Get this right and everything else follows.
Run those prompts across all four engines in clean conditions. ChatGPT in a temporary chat with memory off. Perplexity in a private window. Claude in a fresh conversation. Gemini and Google AI Overviews from a logged-out browser. Run each prompt the same way every time, log the answer, log the sources cited, log whether your brand was named, log which competitors were named, log which categories of source the engine pulled from. This is tedious and it is also the highest-leverage research a healthcare commercial team can do this year.
Score every answer on three layers. First, brand recognition — were you named at all? Second, source authority — was your domain cited as the source the answer leaned on? Third, narrative accuracy — when you were named, was the thing it said about you the thing you actually want hospital buyers to believe? A surprising amount of the time, the AI is naming brands but mischaracterizing them. That is a fixable problem only if you know it is happening.
Map competitive citations. The domains that show up next to yours, and the domains that show up instead of yours, are telling you what the model treats as authoritative in your category. Almost always it is a mix you would not have predicted — a regulatory primary source, an analyst piece you forgot existed, a peer-reviewed study, one specific competitor’s resource page, occasionally a Reddit thread or a niche newsletter. That mix is the actual editorial board for your category right now. Your job is to be on it.
Look at the gaps as a content strategy, not a vanity metric. The most useful output of this audit is not a score. It is a list of prompts where you are invisible, prompts where you are mentioned but mis-described, and prompts where a competitor owns the answer cleanly. Each of those is a content brief. A real one. With a defined buyer, a defined intent, and a measurable target.
Re-run the audit on a cycle. AI engines update their indexes constantly. Citation positions move. Competitors publish. Models change behavior. A one-time audit is a snapshot of a moving target. The teams that win this are running a quarterly cycle at minimum, and the most aggressive ones are running monthly.
What This Means for Healthcare Commercial Teams Right Now
The hospital buyer asking ChatGPT which RPM platforms are TEAM-ready in May 2026 is making a decision that will be largely formed before any vendor knows the question was asked. That is the new front of the buying cycle. It is also where the biggest gap currently exists between the teams who are paying attention and the teams who are not.
Most healthcare commercial teams are still operating as if the funnel begins at the form fill. It does not. It begins at the prompt. The teams that build a disciplined audit cadence on what their buyers are asking AI engines, and a content infrastructure designed to be cited rather than ranked, will own a structural advantage in this market for the next several years. The window on that advantage is open right now and closing fast.
Three Things You Can Do This Week
Here is what to actually do between now and Friday.
1. Run the Stakeholder Mirror Test on your last three closed-lost deals.
Pull the last three deals you lost in the final stages. For each one, list every stakeholder you know was involved. Then list every piece of your content, every LinkedIn post, every email, every webinar that stakeholder plausibly saw before the decision. If the answer for any stakeholder is “nothing,” that is your gap. If the answer is “two LinkedIn posts about our product features,” that is a worse gap. The goal is not attribution. The goal is to see, honestly, what version of you was in the room when you were not.
2. Write one piece this week that no one else in your category could have written.
Not a product post. Not a “five tips” post. One piece — eight hundred words is plenty — that takes a defensible position on something a hospital CFO is wrestling with right now. TEAM target price methodology. The HCAHPS digital transition. The cybersecurity insurance cliff. Site of service economics in your specialty. The test: if a competitor’s marketing team could have written it, do not publish it. If a CFO could forward it to their CEO without embarrassment, you are close.
3. Build one reusable piece of infrastructure, not another deliverable.
Pick one frame, one model, or one diagnostic that your buyers can use without you in the room. A four-question TEAM readiness self-assessment. A one-page worksheet that maps a service line’s HCAHPS exposure under the new methodology. A capital allocation lens for AI investments tied to specific reimbursement levers. Build it once. Put your name on it. Let it travel. The piece of paper that ends up on a CFO’s desk three months from now is worth more than thirty posts that get a polite like.
What I Can Offer That Most People Cannot
Over 25 years inside hospital operations means I read commercial content the way a CFO reads it — looking for whether the writer actually understands how the building works.
Here is what I am opening up to a small group of healthcare commercial teams this quarter:
The Hospital Buyer Content Audit. I will personally review up to ninety days of your outbound content — LinkedIn, email sequences, white papers, webinars, sales decks — and tell you, line by line, what a hospital CFO, COO, VP of Supply Chain, and value analysis chair would actually do with each piece. Read it. Forward it. Roll their eyes. Delete it. You will get a marked-up file, a one-page summary of what is working, what is invisible, and what is actively hurting you, and three specific pieces of infrastructure I would build first if I were running your commercial team.
This is not a content audit from a marketing agency. It is a buyer-side audit from someone who has sat across the table from your buyers for a quarter century.
The Hospital Buyer AI Search Audit. This is the companion offering, and for most teams it is the more urgent one right now. I will build a healthcare-specific prompt set tailored to your category and your buyer roles — CFO, COO, VP Supply Chain, value analysis chair, chief AI officer, clinical champion — and run it across ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews under clean conditions. You will get a written report covering: where your brand is named and where it is invisible, where your domain is cited as a source and where competitors own the citation slots, where the AI is mis-describing what you do, which prompts are owned cleanly by a single competitor and why, the actual editorial board of your category (the domains the models trust most in your space right now), and a prioritized list of the specific content pieces and structural fixes I would build first to start moving the needle. This is the diagnostic almost no one in healthcare commercial has run yet, and the teams that run it first are the ones who will own the next several years of pre-sale visibility.
Both audits can be run separately or stacked. Stacked is the version I would push you toward, because the AI search audit tells you what your buyers are actually asking, and the content audit tells you whether what you are publishing answers it.
If you are interested in either one, message me directly on LinkedIn at https://www.linkedin.com/in/lisamiller at lisa@lisatmiller.com.
I am taking a limited number this quarter.
The Question Worth Sitting With This Week
If you sell into hospitals, the hardest question to sit with right now is whether the content you are producing would meaningfully sharpen the thinking of the executive you are trying to reach.
Not whether it is on brand. Not whether it shipped on schedule. Whether it would actually move them.
Most of what gets produced fails that test. The commercial teams that pass it are the ones building the next decade of pipeline.
If you want a deeper read on how this specifically plays out for medical device, life sciences, and healthcare technology companies, I wrote a longer companion piece on this earlier this year that is worth reading alongside this one: How Medical Device and Life Sciences Companies Can Use AI Search to Win Hospital Contracts in 2026.
— Lisa