Latest / Medsolve Dynamics / The Alternative Does Not Have to Win (AMAC Summer Seminar 2026)

Transcript

The Alternative Does Not Have to Win (AMAC Summer Seminar 2026) · Medsolve Dynamics

• Yoel Bakas: Thank you, Jim, and good morning everyone. If you were here with me in the spring, welcome back. If you weren't, ⁓ maybe one thing to know about me before we start, I spent 20 years on the vendor side of this business, ⁓ at Varian, at Mavion, at VRay, and most recently at GE, ⁓ before ⁓ leaving and starting my own company, MedSolf Dynamics. And I was a person that they send when the deal is big enough to matter. And for today, you know, billing and coding and reimbursement and all the education pieces are, you know, there's great people for that. And many of those are speaking at this meeting. That's not what I'm going to be talking about today. I'm going to be talking about negotiation with vendors, and that's a whole different beast. And that part I lived. So That's what I'm gonna be focused on today. And today I'm on your side of the table and I'm gonna show you how the other side actually runs. Let me start with likely the question that many of you walked in with this month. And this month it's not hypothetical because CMS dropped two proposed rules in twelve days, two different payment systems working in opposite directions, and a lot of the reads land in the same place, which is a paragraph about the long term sustainability of physician practice. And that's true. And it's something that the societies are right to fight. It's also not a thing that you can do anything about on Monday morning. So let me tell you instead what this hour is about and what it isn't. I'm not here to tell you the sky is falling. You'll have plenty of people doing that, and your inbox is likely full of them. And everything I'm gonna show you here is about what you actually can do. not what's being done to you. And there's a lot you can control right now, and quite frankly, more than you're using. And that's the whole hour. The question underneath the noise is something like this. If you're a hospital outpatient, conventional treatment next year might look better, except it's funded by a 340 B cut that half the people writing about expect will get walked back. And if you're freestanding, a different rule applies and you're looking at a cut. And if you've been building a Theranostics program, That's based on the 340B spread, that margin might be gone. That's the same specialty, the same patience, the same machines. Two payment systems working in opposite directions, 12 days apart. And which one you're sitting in is not something you chose this year. And unfortunately, none of it is final. Your vendors are already quoting you like it is final. And your budget your budget locks before any of this gets done. So what do you do? This really isn't a trend that you can plan against. It's more of a scatterplot. And underneath it are two clocks. One is the rule, and the rule can still move. It's at the moment proposed and not final, and it won't resolve on your timeline. The other clock is your budget, and you have to lock that now. And it doesn't wait for the rule. So the work today really isn't in predicting the rule. Nobody in this room can predict that. I certainly can't. The work is in knowing whether the capital decision on your desk survives it with either version of it. And in the next 90 days, most of you are gonna sit across from somebody that wants to sell you something. Because everyone at the table needs this rule to mean something. You just need your numbers to actually be right. Your vendor, on the other hand, means move now. Your consultant needs it to mean hire me. And look, I'm a consultant, so apply that to me too. Everything I tell you here in the next hour has to hold to the same test. And if it doesn't survive your own numbers, don't take it. If you're with me In the spring, you'll remember four lenses that we applied to all deals. I'm not gonna reteach them today, but we are gonna see them work to answer this exact question against a rule that dropped two weeks ago in a deal that already closed. So Here's a premise for the whole hour. The 2027 conversion factor is 32.8409. Four decimal places. Here's what you can do about it. Nothing. Not one of them. Now, if I asked you what the escalator is on your service contract, if I asked you if it's three percent, four percent. How many of you know that off the top of your head? And here's the crazy part. That one you actually can change. And that's the whole hour right there. Whether those two rules help or hurt you, one thing does not change. The only number you can actually control is what it costs to run your program. Not the payment rate, what you spend to deliver care. The payment rate is set in Baltimore, and you don't have the option to touch it. Your cost structure is set across the table. And that one is yours. Some of you may go up next year. Some of you will likely lose the Theronostics margin. No one has the final number. And again, you have to figure out your budget for next year now. Is not stability. That's volatility. And that's exactly why you can't build your margin on a number that somebody in Baltimore controls. So the question was never what does CMS do? It's whether your program still pencils in at 80 cents on the dollar. If you were here with me in the spring, you'll recognize that question. Hang on to it today, because you're gonna see it earn its key. We'll walk we'll walk through a program and how we applied that same question on a deal that actually happened. So the four lenses is a reminder. Every decision should get held up to the same four lenses. The first one is reimbursement alignment. Does it actually hold at 80 cents on the dollar? The second one is operational burden. Can you staff it, supervise it, schedule it without breaking something else that's already running? The next one is compliance exposure. And other speakers here will cover the compliance rules themselves. I'm one step earlier than that. I'm in the purchase that creates the compliance obligation in the first place. Because by the time it's a documentation problem, the decision that cost it was made months before. And the last is contribution margin. Does it make you money? At your volume, not the vendor's projected volume. Those are the four lenses. And before you can use them, you have to understand what you're using them against. Because there's a conversation happening across the table, and it's far more organized than you think. See, the incentives. We're never aligned. Your rep has quotas. They have things to sell. You go into it expecting one thing, and it becomes something very different. And it's not the numbers are horrible, it's that they're not aligned with what you expect them to be. And in order to understand this, you have to understand what's going on with that rep. Your sales rep has a quota that's due every week of the quarter. A quarter that ends before yours does. They have Wall Street targets that they have to hit. They have a committed deal that they told their manager was coming in, and if that deal slips, someone else on their team has to make that up. None of that is about your margin. And all of it is pressure. That pressure is your leverage if you know it's there. So when you're sitting across from a sales organization, your rep isn't improvising. It's a four-step sequence, and it's the same one every time. So let me give you the actual words so you'll hear them coming. The first step is to make you doubt what you have. That sounds something like this. The new platform has better throughput. It has better functionality. You don't want to be on technology that's a decade old. You want to make sure you have the latest and greatest. It's future-proofed. You're protecting your investment. And that fleet of Linux you have, you know, the ones that we're treating patients fine this morning, it's all of a sudden the risk. Step two is to make you want their new thing. Ours is the only thing that protects the investment because we always iterate. We'll keep you one step ahead of where this is all going. Of course, they have no idea where this is all going. Nobody does. And then it's you know things like it's integrated with your workflow. We've been in the business the longest, we have the deepest, strongest, best solutions for everything. Step three is making now the only time. It's the end of the quarter. The price is gonna go up. This deal expires. They won't be able to extend it. And this happens the worst at the end of their fiscal year. And it's all designed to force your hand. Step four is partnership. And this one's tricky because it sounds like a really good thing. Lately though, partnership means a discount off of list price. They talk about the relationship and then they don't address a single thing that the relationship would actually include. Watch the things that you don't get asked. What matters to you? What your program should actually have. What training your team actually needs? What education they could be running for you. What response time should look like on a service contract? Whether a faster installation or an earlier delivery would matter to you. None of that. It's all price and discount. They leave out what partnership actually means, which is understanding what benefits you instead of what benefits them. Now, let me tell you what that four-step thing actually is. Because it has a name on the other side of the table. It's called a sales methodology. It's a licensed, structured training process. And every major manufacturer has one. There are slides for it. There are scripts. There's role play. And there's a manager. That's checking whether the rep is executing the steps in order. I know this because I used to train people on it. So when you're in that meeting, you're not having a conversation. You're being walked through a process. And most programs never know they're inside it. In this year's version, the rule you walked in worried about. It's just the newest line in step one. Make you doubt what you already have. It's being handed to every rep this quarter by CMS. And step three, now is the only time, it's also at play. And look, I think vendors know that at some point pricing will have to come down. So the urgency you're feeling isn't about your budget cycle. It's about locking you in at today's number while they still can. Everything so far is about how they sell you the machine. The capital conversation. But there's a second conversation, and it usually happens the same day, in the same room, with the same people, and that's the service contract. And almost nobody gives it the same attention. Here's how to catch that one in a single number. The same flat service rate quoted in a one-year-old machine and a 20-year-old one. The new one barely has any part risk, no downtime history. The old one carries every failure mode the platform ever had. And you're getting the same rate on both. Think about what that means. That means the rate has nothing to do with what it actually costs to service that machine. It's not built up from parts or labor or travel or failure history. It's a percentage of what you paid. A price with no link to cost isn't a price. It's an anchor. And an anchor has no defense. I'm gonna walk you through a deal, a real deal, that we helped the radiation oncology and a hospital navigate, a joint venture. This is a radiation oncology group that was trying to do a joint venture with a hospital. A new program, major capital commitment, and the projections came back showing a 17-year return on investment. 17 years in radiation oncology at 17 years. That project is dead. It's not challenging, it's dead. Nobody funds a 17-year return on investment on a radiation oncology program. And here's what made this maddening. The demand was there. This was not a market problem. The patience were real. The referrals were real. The need was real. The program was dead on paper anyway. So we started looking, and everything was wrong. The vault cost was wrong, the capital number was wrong, the construction estimate was wrong. Nothing was correct. And here's the part I want you to sit with, because it's the whole reason this matters. You have two parties going into a joint venture. In theory, Both want the same thing. Same goals, same patience, but you look at the numbers and it doesn't look like that at all. The return on investment is 17 years. The pricing favors one side. The construction numbers are strange. It looks like one party is trying to undermine the other. And here's the sad part. I don't think they were. I don't think anybody in that deal was trying to undermine anybody. What I think happened is this. There was a construction cup company that builds hospitals. They don't build vaults. They're not bad people, just the wrong expertise. There was a physicist with the absolute best of intentions. Making sure that the machine was as complete as it could possibly be. Because they know how this works. You get the toys and the bells and whistles now, or there's no funding for them later. And then things get added to the configuration. And you have a vendor who's trying to make sure the hospital has everything it could possibly need without ever really understanding what the program's actual clinical needs were, what they'd actually use, and what made sense for the patients in front of them. So nobody really was the villain. Everybody was doing their job. They were just doing it with the wrong expertise. And nobody was holding the whole thing at once. In that 17 years, that wasn't a vendor lowballing them. That was the hospital's own projection. Their people, their model, clean, confident, right there on the slide. And nobody had flinched at it. So the job becomes juggling two organizations and finding a path that makes both of them work. Because at the end of the day, you're serving patients. And patients matter more than anything else. The math just had to make sense. So let me show you what was hiding in that number. The first thing is the vendor opens the same way they always do, with one number. The LINAC, the service contract, the software platform, all bundled together. ⁓ and by the way, it expires a quarter end. And before I tell you what happened, imagine what that's like if you're buying something. What would you do? Imagine you have two houses. One's made by a Valley company and one's made by a European company. And those two houses that you can buy are very nice and they're very expensive, and they give you one number, and you're trying to buy this house, and all they tell you is, hey, the house is one price. How many people have ever bought a house with one number and just a discount without ever doing any research on anything else? Yet that's what happens. The reality is most programs do one of two things. They either push back on the total and take whatever discount they can get, or they treat this artificial deadline as real and sign the deal before they actually negotiate it. Both of those accept the bend the vendor's frame. That is that this is one number, and that all you're negotiating is some sort of discount. And that single number is the point. One price is easy to approve. Seven prices invite questions. So the move is to break the bundle apart before you talk about price at all. Because inside that bundle, I'm gonna walk you through where the actual money was. Let's take one line out of a bundle. Service contract. On a $4.1 million dollar contract, a service contract runs a little over five percent a year. So call it $215,000 in one year. It's not super aggressive, that's about standard. So here's my question. In five years, how much do you pay? For those of you that just multiplied 215 times 5, you'd be wrong. The actual answer is not 1,075,000. The answer is 1.19 million. And that's a difference of $113,000. That $113,000 is the escalator. It's 5% a year that's compounded. On a contract that nobody opened and nobody approved that $113,000. Nobody had a meeting about it. It just happened. And that's one machine. So let me do the rest for you. Three Linux in your department, that's about $3.6 million over five years. Five Linux, that's roughly six million dollars. And in each of those, about a tenth. of it, its escalator that no one ever put in a budget. It's happening in your department right now, today. And the thirteen the $113,000 comes from basically four clauses in your service contract. They're all there, they're all doing their job quietly while no one watches the calendar. The first one is the Evergreen Clause. That's the one that starts the chain. And it works something like this: the contract auto-renews, unless you give written notice, usually 90 days out. And nobody owns that date. So the term rolls, and you never had a meeting. Once it rolls, the escalator comes along with it. So go back to that $215,000 contract with a 5% escalator. Nobody flagged the notice date, so it renews on its own, and that's $10,750 more in year two that nobody approved. Nobody said yes to it. Nobody said anything at all. Do that five years running, and you get to the $113,000. That's the whole mechanism. Evergreen rolls the term, the escalator rides in on the renewal, and it works because no one is watching that date. Now multiply that by every single contract in your department that's running on autopilot. The next one is uptime with no teeth. And this one's twofold, and both halves really matter. First, most programs don't track uptime with any real specificity. And almost nobody sits down with their vendor regularly to hold them to it. Those meetings are where you established what the contract is worth now and going forward. If you don't have them, You don't have a ground to stand on. The second, and this is a big one, is when uptime is calculated differently than the way you treat. Scheduled maintenance is sometimes excluded. The measurement window may not match your treatment day, or they classify customer-caused downtime as excluded. But here's the thing. If that machine was down during clinical hours, your patients don't care that the contract calls it excluded time. So make sure uptime to you means the same thing that uptimes means to your vendor. And if it doesn't, you have a definition, not a guarantee. The next one, and I used to love to do this, is to bundle software into the Linux service contracts or capital purchase. That platform comes with the Linux. And three years later, it means reopening the whole agreement to evaluate the software. On long-term agreements, especially, it's super important to have enough KPIs built in over the life of that contract that you can get out. Know what you're going to measure because the vendor will not perform. It's not out of malice. Nobody performs a hundred percent perfectly for seven years. But when they don't, you want to be able to say to them, you're not meeting the KPIs, and so we want credits or we want out. So let me walk you through four questions that we walked with the JV team. All these are done before the negotiation starts, and it's what sets up the negotiation. The first one is what are you being asked to buy? And separately, what do you actually need to run this program? It's two different lists that nobody had put side by side. And here's how to build it in your own program. It starts before anybody sends you a quote. You interview your clinical team. Not a survey, a real conversation. You ask things like, what are we trying to treat? What do you need to treat it? What are you actually trying to accomplish clinically? Then you talk to your physicists before the configuration, not after, and you ask them if they have the right QA equipment, the right physics controls to deliver what the clinicians just told you they need. And then you take that configuration and you break it apart piece by piece. And you figure out what's included in your capital contract versus what you're buying separately. You ask your vendor if they can pass through certain items at a discount, or whether it's better for you to be buying those separately. All of it starts in the same place: a deep understanding of what you're trying to treat, what your volumes are, what your patient population actually looks like. So you're not overbuilding for something you're never going to treat. When we did that and we removed the spend that didn't match the clinical need, we found over $3 million in savings. And then there's the configuration itself. Not whether you buy the machine, but what's on it. And think about it this way: if you're never going to do radio surgery, do you really need a headframe? Do you need the higher isocentric accuracy or a standard isocentric accuracy that okay. And if there are just a few patients a year that you can send off somewhere else, are you willing to refer those out? That's not a clinical compromise. That's a clinical decision made on purpose, with the number right in front of you. And it takes knowing your clinical goals and having a long enough vision to say, this makes sense, or it doesn't. In this particular example, that was another $900,000 to $1.8 million in savings. The next one is construction. And it's does the construction scope reflect what you're actually trying to do clinically, your clinical workflow, or is it some architect's assumptions? And in this particular example, here's what we found: the vault was being priced inside of an exclusive relationship. Essentially, one bidder. And someone on the program side looked at it and asked one thing. They said, I want more parties bidding on this. I want fair market value assessment of what this would actually cost or should actually cost. And if you won't give me either, you eat the difference. That was the whole intervention. That was the ask. Refining that scope without moving the timeline saved an additional $1.3 to $1.5 million. Think about that. Someone in that program asked for a second bid, and that was worth over a million dollars. Question three is does the revenue projection hold at eighty cents on the dollar? Or does it only work at the vendor's number? You recognize that question, it's the one I left you with in the spring. The operating assumptions in that plan were built on the vendor's number, the vendor's projections, not the program's actual volume, not the program's Paramix, somebody else's number, in somebody else's number, describing somebody else's business. And when we got in there, two things were wrong. Chargers were not getting captured correctly, and the workflows didn't match how they were actually going to run patients through that clinic. And you have to understand both at once. You have to understand what's being billed and how, and how patients actually move through the building and where the charge capture was supposed to happen. And I'll be honest with you about who can do that. That's not a spreadsheet job. That's something you figure out from the inside. It's not something you can figure out from outside. That takes somebody who's been an administrator. Who knows how patients should move through a clinic and how that should be billed because they've done it. The people in this room and online are the ones who can see that. It's a peer to peer conversation and there's no substitute for it. In this particular example, workflow and billing adjustments led to about a million dollars in operating revenue that the original plan never counted on. The fourth question, and you now see what happened here, it took a 17-year ROI to under three. None of it came from coding, it came from how the decision was structured before anyone signed. And the fourth question is not anything about the vendor. It's about you. Not what's the price. What are you bringing to the table? Because Look at what the first three questions actually did. They didn't change the vendor's price list, not one line of it. They changed what the program knew when they walked in. And that's the actual job. Before you ever sit down, know what you're trying to treat. And just as importantly, know what you're not trying to treat. Know your numbers. your volumes, your payer mix, your case mix. Not roughly, actually. Know what your clinicians need. Because you asked them, not because a vendor told you. Know where you want this program to be in 10 years, in 15. Because the contract you're about to sign runs through most of that. And the machine you're buying will outlive three of your budget cycles. If you don't have a vision that long, the vendor's vision will fill that gap. And most important of all, bring the data, the actual analytics. Not a feeling, not a memory, not what somebody said in a meeting last year. That's what you bring to the table. And if you bring all of that, you're not really negotiating anymore. You're just comparing the number to the one you already know is right. And I want to be precise about what we did, because that's the whole point of this talk. We did not renegotiate on their behalf. We restructured what the program was bringing to the table. so that when they did negotiate, the numbers were already different. Nobody out negotiated anybody. Nobody was aggressive. They just walked in with a different set of facts than the ones they had been handed. If you want the whole walkthrough of all four questions, how we found each number, shoot me an email. It's right on the right there on the screen, and I'll send it to you. There's no form, no sign-up, just shoot me an email, Yol at medsolvdynamics.com and I'll send it to you. Now, this same sequence is showing up somewhere new. And that's artificial intelligence. Most of the AI being sold into radiation oncology right now. Improves workflow. It does not improve reimbursement. It is a cost center dressed up as an investment. All four lenses apply here, but two of them are where the vendor can't answer. And remember the talk from yesterday. CPT has set three or four dozen new AI codes. None in radiation oncology. Today they're in cardiology, ophthalmology, pathology. Not one is in radiation oncology. Right now, every one of them is category three. So for what they're selling you, there is no billing offset. It's not an investment, it's an expense. So run the math before you sign. The second one is compliance. If it touches the clinical workflow, it touches your documentation. So ask them who owns the output. Ask it in the meeting and see how they answer. The demos are always slick. The performers are often silent. And here's the other one that's live right now. It's Theranostics. I've been saying the same thing on calls since the spring, and I'll say it here. If you can get 340B pricing, Theranostics is really interesting. If you can't get 340B pricing, it's a lot less interesting. It's not a knock on it. The clinical evidence is super strong. ACRO is pointing programs there, the revenue is real. The clinical indications are growing. It's just that the economics ran through the 340B spread. Well, CMS just proposed taking away the if you can get it. In plain terms, what they have proposed is this. The 340B that was the margin, the rule proposes turning that into a pass-through. So what do you do now? The discipline is simple. Before your budget locks and before you build it, run the plan at three different prices. Run it at today's price, the rules fall back, and the rule as proposed. If it only pencils in at today's price, you don't have a plan. You have a bet. Nobody's going to tear up a program on a proposed rule, and nobody should be signing capital before checking whether the plan survives it. And to be super clear, a program that runs those three prices and decides not to build is just as good an outcome as one that builds on numbers that hold. So it all starts with what you now know. The data and the knowledge is exactly where you start. Without it, you don't have the leverage to go have these conversations. With it, the conversation changes drastically to this is where we are and this is what we have. And there are three levers you already hold. The first one is this the vendor's clock is not your clock. And it's one of your best lovers if you know how to read it. Remember, reps carry a quota every quarter. The end of any quarter moves reps. The one that really moves them, it's the end of their fiscal year. That's when their annual number gets counted. That's when leadership is staring at the gap. That's when a deal the closest on the 30th and a deal the closest on the first lands in two different years. That is the most motivated a sales organization ever gets. And here's the part that most programs miss. A vendor's fiscal year often isn't the calendar year. It can end in January or June or September. So the pressure you want to use isn't always in December. Find out when your vendor's fiscal year actually closes. That date is your leverage, and usually they're not going to volunteer it. The next one is bundling, and we talked earlier about how much I liked bundling when I was on the sales side. And this one's a bit of a paradox, so stay with me. The vendor bundles to keep you from seeing line items. The first thing you need to do is break it apart, separate out the Linux, the service, and the software, and find out what each piece is actually worth. That's what we just did. But then You put it all back together. Because vendors move on price when the total deal size justifies it. So you don't want to negotiate the machine in March, the service contract in June, and the software in September. Because that means three smaller conversations and three smaller discounts. So what you want is to have one conversation where the whole number is on the table. That's where the movement is. And if you can bundle across multiple years or multiple machines, even better. Always break it apart to understand it, but bundle it back up to negotiate it. The next one is competitive tension. Always get competitive bids. In the spring, I talked about one of my favorite purchasing guys who always had a competitive bid on his table. I would walk in representing the big V, and he always had an Elekta quote right on the table. Every single time. And look, I knew they were never gonna buy Elekta. I did. But It put pressure, it put pressure on the sales team. So even when you already know what vendor you want, especially then, make sure you have competitive bids. The alternative does not have to win. It just has to exist. It's the thing that gives you somewhere to stand. And you just watch what one additional bid did on that vault construction cost. And don't ever undervalue how much power you are actually sitting on. The reality is vendors need these deals. Their leadership gets measured on whether they hit their number. And if they miss the number, that's somebody's job. The person across that table is under more pressure than you are. They're just Much better at not showing it. The best outcomes don't go to the most aggressive. They go to the most prepared. And look, I know a lot of you in the seminar are not the person who signs the contract. That's fine. Go back to that vault for a second. The person who saved that program a million dollars didn't sign anything. All he did was ask a question. Can I get two more bids? That's the job. You don't have to run the negotiation. You just have to put the right question on the table before somebody signs. So here's a whole hour in one question. Does it pencil at 80 cents on the dollar? Not the vendor's projection, not Baltimore's November guess, your volume, your payer mix, your schedule. In the spring, I handed you this hypothetical question. Today, you watched it do real work. It's one of the four questions that took a dead project to under three years. The same question. And it's the one that holds up no matter what CMS does in November. If you can't answer it, you don't have a purchasing decision. You have a risk you haven't priced. And I know you can't run the math from your seats right now. So here's a version you can run on Monday in one afternoon. Pull your original ProFarma out of the capital file, put it next to the actual 12 months. Of performance. Mark every assumption that missed by more than 10%. That marked-up page is your agenda for two conversations: the one with the CFO about what your last performer got wrong, and the one with your vendor about the contract you're renewing next. And to be clear, this isn't a pitch. I'm not selling you anything from up here. I write a newsletter called the Oncology Operator. It's free. This week's issue walks the exact deal I just showed you in writing. It's called The Machine Was Right, The Deal Wasn't. If it's ever useful to run your own numbers before your budget, that's what the newsletter is for. It's at oncologyoperator.com. If you're here, you can scan the QR code. ⁓ or you can go and sign in or sign up. And if AI is what you want to hear about, Heather Turner and I are doing a talk at SROA in September. Jim talked about SROA already. We'll be doing a talk there on AI coming up next. So that's the whole framework put to work. And look, cancer isn't slowing down. Patients aren't slowing down. People still need to be treated, and that hasn't changed, and it won't. Everybody in the joint venture wanted to serve those patients. Nobody was the villain. The math just needed to make sense. That's what this whole conversation is about. Not negotiating harder, making the math make sense. So the program. still is still standing to treat them. I spent 20 years on the vendor side, so you don't have to guess what the other side is doing. Thank you.