Latest / Becker’s Healthcare Digital Health + Health IT / Why Ambient AI Took Off in 2025 and What's Next in 2026 and Beyond

Transcript

Why Ambient AI Took Off in 2025 and What's Next in 2026 and Beyond · Becker’s Healthcare Digital Health + Health IT

• 0:00
  Hi, everyone. This is Lucas Voss with Becker's
• 0:02
  Healthcare. Thanks so much for tuning in to
• 0:04
  the Becker's Healthcare podcast series. It's fantastic
• 0:07
  to have you. Today, we're talking about why
• 0:09
  Ambient AI took off in 2025,
• 0:12
  as we all know, and what's next
• 0:15
  in 2026
• 0:16
  and beyond. And joining me for today's discussion,
• 0:18
  I'm so excited to have him, is doctor
• 0:20
  Thomas Kelly, CEO and cofounder of Heidi. Doctor
• 0:22
  Kelly, thanks so much for taking some time
• 0:24
  and being here today. It's great to have
• 0:26
  you.
• 0:27
  Oh, it is. Yeah. No worries. Excited to
• 0:28
  chat more. For those that might not know
• 0:30
  you yet, I do wanna kick us off
• 0:32
  with introductions here. If you just wanna give
• 0:34
  us a brief overview of, yourself and certainly
• 0:37
  your work in health care.
• 0:39
  Yeah. So I'm Tom. I'm a doctor by
• 0:41
  background.
• 0:42
  Australian born, if you can't tell from the
• 0:44
  accent.
• 0:45
  Founded Heidi after being a vascular surgery resident
• 0:48
  in 2021.
• 0:49
  Had a view that every doctor should practice
• 0:52
  with some sort of sort of AI partner
• 0:53
  in their care. I had sci fi visions
• 0:56
  of, like, you know, maybe kinda everything, like
• 0:58
  prescribing, ordering investigations for me, reviewing my results,
• 1:02
  doing my notes, basically freeing me from my
• 1:05
  medical record and letting me see more patients,
• 1:07
  particularly for me, operate more because that's what
• 1:09
  I love doing. And, yeah, I'm proud to
• 1:12
  lead,
• 1:13
  probably one of, like, the largest health care
• 1:15
  companies in the world. So we we're a
• 1:17
  global company,
• 1:19
  around more than 50% of doctors in across
• 1:22
  Australia, New Zealand, Canada, UK use Heidi every
• 1:25
  week. And then in The US, we have
• 1:27
  an amazing growing business.
• 1:30
  I'd say where we're most prominent is in
• 1:31
  outpatient specialties, so across, like, urology,
• 1:34
  like, GI Alliance,
• 1:36
  US Urology,
• 1:37
  a a bunch of other amazing partners, and
• 1:39
  then, a few large health systems and growing,
• 1:42
  every year. So like Beth Israel, Franciscan,
• 1:45
  Maine General,
• 1:47
  Well, it's great to have you, and I
• 1:49
  feel like, again, your sci fi vision that
• 1:51
  you just mentioned sort of became true in
• 1:53
  2025
• 1:54
  for a lot of folks, for a lot
• 1:56
  of providers in itself.
• 1:58
  It's moved incredibly fast. It's going to move
• 2:01
  incredibly fast in 2026
• 2:03
  and certainly beyond.
• 2:05
  From your perspective and your work in the
• 2:07
  space,
• 2:09
  what's driving that pace right now, and what
• 2:12
  makes ambient
• 2:13
  AI fundamentally
• 2:14
  different
• 2:15
  sort of from what we've seen in earlier
• 2:17
  health care health care tech waves?
• 2:20
  Yeah. I think overwhelmingly,
• 2:23
  what I'm seeing is
• 2:24
  that, you know, and it's like in every
• 2:26
  industry, not just health care. So the boards
• 2:28
  and CEOs are asking their teams, like, what
• 2:30
  are we doing about AI? Like, I'm here
• 2:32
  I'm using it. It's amazing. Like, it's it's
• 2:34
  saving me time. Like, what what's your AI
• 2:36
  strategy?
• 2:37
  And almost always,
• 2:38
  what I've seen is that the especially in
• 2:40
  larger organizations, the CIOs,
• 2:43
  I love them, but they don't have a
• 2:44
  lot of taste. So they just end up
• 2:46
  picking whatever is the most integrated option in
• 2:48
  their stack, and then they go back to
• 2:49
  their board and they say, hey. Look. We
• 2:51
  we're doing AI now, and we've chosen chosen
• 2:54
  an Ambient tool.
• 2:55
  And specifically with Ambient, the reason I think
• 2:57
  it's gone so fast is it's one of
• 3:00
  the few things that doctors actually love. So
• 3:03
  usually usually software implementation
• 3:05
  is like a doctor's worst nightmare, like sitting
• 3:07
  in classes with someone teaching them how to
• 3:09
  click click different places in the medical record.
• 3:11
  It's not not why I did the went
• 3:12
  into the job.
• 3:14
  And,
• 3:14
  basically, I think that that downwards pressure met
• 3:18
  doctors wanting this tool, and so the CIOs
• 3:21
  generally picked whichever scribe work best with that
• 3:24
  medical record. I think this year and probably
• 3:26
  next year is gonna be where the rubber
• 3:28
  meets the road on ROI. Like, how many
• 3:30
  of our my doctors are actually using this?
• 3:32
  Like,
• 3:33
  are there better tools? Like, how should I
• 3:35
  actually deeply think about, like, what AI means
• 3:37
  for the organization, especially as
• 3:40
  the capabilities
• 3:41
  grow?
• 3:42
  I think
• 3:43
  back to my sci fi vision, I think
• 3:45
  it'll become more and more about, like, an
• 3:47
  AI partner that actually participates in care. Like,
• 3:50
  we're already seeing in Utah, like, the Doctronic
• 3:53
  trial where where an AI is starting to
• 3:56
  basically, like,
• 3:57
  almost fully autonomously prescribe with some doctor sign
• 4:00
  off. And so I think the health systems,
• 4:02
  like, that's how I would conceptually try to
• 4:04
  plan how I work with AI. Like, where
• 4:06
  are the areas in the health system that
• 4:08
  are
• 4:09
  most costly or or hardest to staff or
• 4:12
  have the longest wait lists,
• 4:14
  and spend time with with companies like ours
• 4:17
  to try to figure out, okay, like you
• 4:19
  know, because we already have it. Like, my
• 4:20
  cardi can read the record, can look up
• 4:22
  evidence, can call patients, can so it's like
• 4:24
  there's so many things you can do. It
• 4:26
  just depends on what your objectives are.
• 4:30
  And, yeah, it'll be interesting to see. Like,
• 4:32
  I do think that
• 4:34
  your play scribing on its own
• 4:37
  will struggle to find durable value because it's
• 4:40
  it's getting easier and easier
• 4:42
  to generate amazing notes. It used to be
• 4:44
  a really hard problem. Like, I spent two
• 4:46
  years of my life trying to make this
• 4:48
  work.
• 4:49
  Mhmm. But now if you if you just
• 4:50
  pick up an amazing
• 4:52
  foundation model,
• 4:53
  you know, you get most of the way
• 4:55
  there.
• 4:56
  So it's gotta be about the depth of
• 4:58
  the product, I think, and how much providers
• 5:00
  love it.
• 5:01
  So one of the things that was happening
• 5:03
  in 2025
• 5:04
  specifically was the fact that a lot of
• 5:07
  organizations were moving sort of from the pilot
• 5:09
  phase into
• 5:10
  adoption, into trying to get this at to
• 5:14
  at scale, essentially.
• 5:16
  Yeah. What's what's needed from leaders to make
• 5:19
  that happen? What are the conversations that need
• 5:21
  to happen to be able to say, okay.
• 5:22
  We're moving from a pilot project that was
• 5:24
  sort of a nice to have to then
• 5:26
  now, okay. We're using this daily, weekly, as
• 5:28
  you mentioned with some of the organizations
• 5:30
  that you work with.
• 5:32
  Yeah. I think we you definitely have to
• 5:34
  spend the time to nail it in in
• 5:35
  initial kind of so we we do, like,
• 5:37
  a stage rollout,
• 5:38
  where you have
• 5:40
  an initial alpha group, which is friend like,
• 5:43
  friendlies, so people in each of the departments.
• 5:46
  I think the key is not to cherry
• 5:47
  pick specialties. You have to have absolutely every
• 5:50
  everyone represented that you want to ultimately use
• 5:53
  the tool.
• 5:54
  Then in that phase,
• 5:56
  doing, like, real kind of AI enablement,
• 5:59
  setup for each of those specialists. So we
• 6:01
  do that with our CS and implementation team.
• 6:03
  So that's where we go to hematology and
• 6:05
  we say, oh, okay. For hematology,
• 6:07
  the visit
• 6:09
  contains very little,
• 6:11
  useful information. Actually, the value of Heidi would
• 6:14
  be if we
• 6:15
  look at all of the historical blood results
• 6:17
  and summarize them in the note. Like, that's
• 6:19
  a valuable thing. And then so then we
• 6:22
  set up materials and templates that teach hematologists
• 6:25
  how to use HYDIA in that way. So
• 6:26
  you can use the context tab and select
• 6:29
  the past sessions and then use this template,
• 6:31
  and it will summarize all the blood results
• 6:33
  just the way you like. And I think,
• 6:35
  to be honest, I think that's where we
• 6:37
  stand out. So if if you're working with
• 6:39
  a partner, it's like, oh, incredibly easy to
• 6:41
  roll out. Like, time to value is four
• 6:43
  weeks. You just turn it on, and everyone's
• 6:45
  gonna be using it. I feel like it's
• 6:46
  a red flag. Like, it's it's likely
• 6:49
  that they're just gonna be rolling out, like,
• 6:51
  the same template and, like, limited personalization
• 6:55
  for every single specialty,
• 6:57
  which means that when you hit a, I
• 6:59
  don't know, a plastic surgeon,
• 7:01
  like, the SOAP note isn't is not really
• 7:05
  that useful to them. So if you're just
• 7:07
  trying to drive them to use, like, a
• 7:09
  baked in, SOAP note generator that does a
• 7:11
  good problems list and captures better,
• 7:14
  problems from the record in order to generate
• 7:16
  more revenue, They're probably just not gonna use
• 7:18
  it because they hate the notes, and they
• 7:20
  they don't actually get it it doesn't write
• 7:22
  like them. It doesn't contain it's like more
• 7:25
  note bloat. Like, it's it's a lot of,
• 7:27
  like, random stuff, like random issues and problems
• 7:30
  that they they're not actually managing
• 7:32
  just because that helps with, like, the RCM
• 7:33
  down the track. It's like, I'm just dealing
• 7:35
  with this right hand fracture. Like, get out
• 7:38
  of my hair.
• 7:39
  So I think I think it depends on
• 7:42
  every from our perspective, every specialty is actually
• 7:45
  quite different. Need to treat them differently.
• 7:47
  That's how you get so we we generally
• 7:49
  see, like, 80% activation
• 7:51
  from from sign up. So if someone
• 7:54
  if if you promote it internally, you set
• 7:56
  it free. If someone signs up, roughly four
• 7:58
  out of five will use it,
• 8:00
  every week, and something like 70% of them
• 8:03
  will use it every day.
• 8:04
  And the the thing that's hard is, I
• 8:06
  think,
• 8:07
  in the the PR and the marketing, it's
• 8:10
  sometimes hard to trust because if I talk
• 8:12
  to the systems and they're like, oh, yeah.
• 8:13
  It's only in primary care, but then they'll
• 8:16
  then they'll promote it as if it's, like,
• 8:17
  80% adoption in every every market. Yeah. Primary
• 8:20
  care, easy. Like, they they see everything.
• 8:23
  They do high, turnover medicine.
• 8:25
  This standard soap note is amazing for them.
• 8:28
  But if you think about, like, all the
• 8:29
  specialties,
• 8:30
  all of the different use cases, you have
• 8:33
  to have some ability to personalize, generate templates
• 8:35
  that
• 8:36
  live outside of the medical record. They still
• 8:39
  use data from the medical record, but there
• 8:41
  are many things that providers do that
• 8:44
  just require, like, a more flexible AI tool.
• 8:46
  So that's usually where we sit in. So
• 8:48
  if if a system is is churning off
• 8:50
  another tool or they haven't picked one yet
• 8:52
  and they have a lot of specialty practice
• 8:55
  or they they've tried
• 8:57
  the, you know, single template best at RCM
• 9:00
  tool and they haven't actually seen much,
• 9:03
  much adoption or much uplift, then then, yeah,
• 9:07
  I think we're a good option.
• 9:09
  Yeah.
• 9:10
  I wanna come back to the adoption piece
• 9:12
  and the and the usability part of it,
• 9:13
  which you've mentioned is is really crucial. Right?
• 9:15
  It needs to have that aspect to it
• 9:18
  for for it to work on a on
• 9:19
  a scale basis.
• 9:21
  For you personally, what does usable mean in
• 9:24
  day to day practice? And then certainly, we've
• 9:27
  talked about the workflow piece of this.
• 9:29
  How should leaders think about workflow fit versus
• 9:33
  then also the EHR integration
• 9:35
  when they're looking at evaluating
• 9:37
  ambient AI?
• 9:39
  Yeah. Yeah. I think for me so having
• 9:42
  been a doctor,
• 9:43
  the main thing we time we save is
• 9:46
  when
• 9:47
  the amount of time
• 9:48
  writing notes. So, basically, like, keyboard strokes,
• 9:51
  how many times do I have to edit
• 9:53
  change,
• 9:53
  how much of what the system writes am
• 9:56
  I actually accepting,
• 9:57
  and, of course, like, how often am I
• 9:59
  using it? Like, is it am I only
• 10:00
  cherry picking it for certain visits? Am I
• 10:02
  using it for every visit?
• 10:04
  And
• 10:05
  we we measure that on, like, what's our
• 10:06
  daily active usage? Like, what percentage of sessions
• 10:09
  are we capturing? And and we should be
• 10:12
  in most visits unless there's, you know, patient
• 10:15
  who hasn't consented or it's a really tricky
• 10:17
  tricky circumstance.
• 10:19
  So
• 10:20
  so for us, that has always meant that
• 10:22
  we don't really bend the knee and, like,
• 10:25
  put Heidi so integrated into the medical record
• 10:28
  that it is, like, unusably
• 10:30
  rigid. So, like,
• 10:32
  without naming names, I would say most of
• 10:34
  the options that are very, very, very tightly
• 10:37
  integrated with the record, like the default options,
• 10:39
  the product teams of those products, they live
• 10:43
  and die by the medical record. So they
• 10:45
  wake up in the morning and they think,
• 10:46
  if this tool is doing its job, it
• 10:48
  just fills the fields in front of the
• 10:51
  doctor. Like, that's how they view the world.
• 10:53
  Now don't get me wrong. That is valuable
• 10:55
  for sure. But I view the world more
• 10:58
  in the, like, consumer AI lens like Claude
• 11:01
  or Chativity or, actually, I think I view,
• 11:04
  like, some of the more direct to doctor
• 11:06
  competitors as as real competitors for us. So
• 11:09
  it's like, okay.
• 11:11
  Ignoring, like, the, CDI and the RCM
• 11:14
  flow that happens and all the reasons we
• 11:16
  have our medic record, in ten years,
• 11:19
  is a doctor actually gonna be sitting on
• 11:21
  a screen that has all these fields that
• 11:23
  require to be filled? Like, definitely not. I
• 11:25
  have failed if that is still the case.
• 11:27
  So my take is that that you have
• 11:31
  to build, like, a different UI. Like, for
• 11:33
  someone who hasn't hasn't,
• 11:35
  done engineering, there's this application called Cursor,
• 11:38
  which is what a lot of like, almost
• 11:41
  I don't know. Most engineers are now using
• 11:42
  this tool. And, basically, it's like a single
• 11:45
  chat field, and you can turn on voice.
• 11:47
  And you you speak to the system, you
• 11:49
  type what you want it to do, and
• 11:51
  it's fully abstracting away
• 11:53
  something like five to 10 different pieces of
• 11:55
  software for you. And they all look like
• 11:57
  databases. Like, they all look like the medical
• 11:58
  record. They have all these different things that
• 12:00
  you would have historically had to tick and
• 12:02
  click and write in this file and write
• 12:04
  another file and then so it's a very
• 12:06
  similar idea. And so Heidi is the same.
• 12:08
  Like, usability means that we build AI first.
• 12:11
  Like, what what not just MD and AI,
• 12:14
  but, like, what are all the things that
• 12:15
  will become possible with this tool? Yes. We
• 12:18
  have to integrate very deeply. Like, we have
• 12:20
  to be able to order. We have to
• 12:21
  be able to do problems list. We have
• 12:22
  to be able to write notes that generate
• 12:24
  more revenue, capture issues that we would have
• 12:26
  otherwise missed,
• 12:28
  all things we do. But but, fundamentally, like,
• 12:30
  we will not sacrifice making Heidi, like, some,
• 12:33
  like, ribbon
• 12:34
  inside of the medical record that does AI.
• 12:37
  Like, that's just not how it's gonna happen.
• 12:39
  It's gonna be a full screen
• 12:40
  experience that is, like, built for AI, that's
• 12:43
  fast,
• 12:44
  that does, like, agent calls really nicely, that
• 12:46
  shows you the thinking, that does, like, sourcing,
• 12:48
  that it's all, like, one thing, and that's
• 12:51
  the kind of, like, superintelligence
• 12:53
  AI care partner that I imagine every doctor
• 12:55
  brings to work.
• 12:57
  And so when when evaluating I know this
• 12:59
  sounds like evangelical and, you know, I'm I'm,
• 13:02
  I'm pitching on where it goes. I do
• 13:05
  think it matters. So the the things I
• 13:07
  would watch out for is if it's incredibly
• 13:10
  integrated in the medical record, then you should
• 13:12
  just buy the one in the medical record.
• 13:14
  Like, that's gonna exist,
• 13:15
  you know, whether it's Epic's or not, whoever.
• 13:17
  Like, they're gonna do that the best. I
• 13:19
  don't really think the companies whose value prop
• 13:22
  is there more integrated with Epic than Epic
• 13:24
  Zone AI Scribe are gonna be durable, really,
• 13:27
  unless they do other things.
• 13:29
  If that's failed, so if you tried that
• 13:31
  or or your doctors say, I don't really
• 13:33
  like like this approach. Like, I wanna be
• 13:35
  able to ask AI question. I wanna be
• 13:37
  able to look something up. I wanna be
• 13:38
  able to summarize my week. I wanna be
• 13:40
  able to,
• 13:41
  ask about the last three sessions and then,
• 13:43
  like, change the template because of this patient
• 13:46
  the last three times I spoke about the
• 13:47
  diabetes, but I don't want to summarize the
• 13:49
  whole year. Like, usability is about, like, I
• 13:52
  am an AI serving my doctor.
• 13:54
  And, yes, I ultimately send data back to
• 13:57
  the record, but I have a flexible interface
• 13:59
  that I can let the doctor do what
• 14:00
  they want and then ultimately send it back.
• 14:03
  So does that make sense? I don't know.
• 14:04
  I'm rambling a bit, but I think that's
• 14:06
  the that's the gist. That's your sci fi
• 14:08
  that's your sci fi vision. I think it
• 14:09
  does make sense. It's very important. It it
• 14:11
  creates that flexibility for users. It's not a
• 14:13
  rigid experience. It it it's it's tailored, and
• 14:16
  that's as you mentioned, I think that's the
• 14:18
  way it's it's going to go in the
• 14:19
  next ten years. It's not going to be
• 14:21
  rigid fields, as you've mentioned, that are going
• 14:23
  to be be filled out. It's a it's
• 14:24
  a flexible flexible process that needs to happen.
• 14:27
  Yeah. One interesting thing that you've mentioned, and
• 14:30
  I think this is very important, is
• 14:32
  that oftentimes
• 14:34
  these early wins, everything works, everything is great,
• 14:38
  can actually be a red flag for organizations
• 14:40
  in terms of, oh, this is everything's working.
• 14:42
  We have the early results. Yeah. We've seen
• 14:44
  we've seen, you know,
• 14:46
  after hours EHR time is is reduced, all
• 14:48
  of a sudden improved documentation, completeness, etcetera. It's
• 14:51
  all there. Right?
• 14:53
  When you're looking at results like these and
• 14:55
  sort of these when you think about the
• 14:57
  red flag, right,
• 14:58
  what should health care leaders ask right now
• 15:01
  to separate that, you know, early win sort
• 15:04
  of from repeatable
• 15:06
  durable impact for the next year?
• 15:09
  Yeah. I would ask so
• 15:11
  how many of the doctors across all specialties
• 15:14
  are using this tool, and how many of
• 15:15
  them like, what's what are the what's the
• 15:16
  actual denominator of the activation number? Like, is
• 15:19
  it everyone who's signed in, which I think
• 15:22
  is, like, the right because I, you know,
• 15:24
  I can only do so much to get
• 15:25
  someone to to actually sign up to the
• 15:27
  tool. But once they sign up, then then,
• 15:30
  yeah, like, if I'm activating most of the
• 15:31
  users, I've succeeded. So you'd wanna know that
• 15:34
  it's working across all specialties.
• 15:36
  You know, what's the,
• 15:38
  there's, like, UX, which is just, like, what's
• 15:40
  the user experience?
• 15:42
  Do they find it easy to use? Do
• 15:43
  they find it valuable? Would they be frustrated
• 15:46
  if if,
• 15:47
  it's taken away?
• 15:48
  You can look at objective measures.
• 15:51
  It's it's shocking to me seeing some of
• 15:53
  the studies that are coming out about, like,
• 15:56
  how little some some products are actually impacting
• 15:59
  both, like, revenue, time, and record.
• 16:02
  And they universally, doctors love it because they
• 16:05
  don't have to type as much, but, actually,
• 16:07
  that's not saving them as maybe it's not
• 16:10
  generating as much ROI as as maybe people
• 16:12
  thought months thought it would. The reason it's
• 16:15
  shocking for me is because I've only seen
• 16:17
  the studies that I've seen, and, again, they're,
• 16:19
  like, in that bucket of, like, very integrated
• 16:21
  world. And I think it makes sense because
• 16:23
  you still have to review every field. You
• 16:25
  have to, like, click all the drop downs.
• 16:27
  Like, you're stuck in the record. So you
• 16:28
  can only get so much of a speed
• 16:30
  up. Whereas in our,
• 16:32
  randomized,
• 16:34
  papers, like, in across the world, we see,
• 16:36
  like, overwhelmingly
• 16:37
  in improved, like, RBUs and time in record
• 16:40
  goes down massively. So that I think that's
• 16:42
  the story for 2026, 2027. That's it's up
• 16:45
  to me to tell it. But, basically, people
• 16:47
  who think that
• 16:49
  AI that is gonna change health care in
• 16:51
  the medical record is, like, a bad take,
• 16:53
  it's going to be that the records are
• 16:55
  gonna get increasingly
• 16:57
  almost, like, subjugated to the back and definitely
• 17:00
  used by admin and revenue,
• 17:02
  definitely exists. Like, they will always be there.
• 17:04
  Like, we will be great partners, but it's
• 17:06
  about this, like, AI layer that sits on
• 17:08
  top that partners with the doctor to mean
• 17:10
  that they do not need to use the
• 17:12
  record. Like, that is success.
• 17:14
  Success is I do not ever sit on
• 17:16
  a computer. Like, that's what a doctor wants.
• 17:18
  So you can't build that if you're you
• 17:21
  are the record. Like, that is like it's
• 17:22
  like you're you're trying to get rid of
• 17:24
  yourself. So I think I think that's why,
• 17:26
  like, companies like HYD will will exist is,
• 17:28
  like, to try to
• 17:30
  to actually, like, strengthen the record, like, put
• 17:32
  more data into it. Like, actually, like, increase
• 17:35
  its value to the org, but then to
• 17:37
  the doctors,
• 17:38
  breathe in to actually just practice and, like,
• 17:40
  get the satisfaction.
• 17:41
  So so, basically, I would say, look at
• 17:43
  those those objective metrics
• 17:45
  and then question, like, is there a good
• 17:47
  ROI case here? Like, am I getting the
• 17:49
  value out of the tool that maybe all
• 17:51
  the contracts maybe that I've committed to? And
• 17:53
  then just, yeah, raise raise your ambition and
• 17:56
  expectation for your partners and look at what's
• 17:58
  out there. Because I'd say for most cases,
• 18:01
  Heidi wouldn't have been one that they considered
• 18:03
  when their CEO said, hey. Choose an AI
• 18:05
  tool. At least in The US, everywhere else,
• 18:07
  yes, but in in The US, no, because
• 18:08
  we were just starting. And I think
• 18:11
  we have acquired a different perspective. So if
• 18:13
  you've had a bad time
• 18:14
  with a vendor,
• 18:16
  chances are you will have a bad time
• 18:18
  with the next vendor if the next one
• 18:20
  you choose is essentially the same thing. Like,
• 18:23
  we work super well with Epic. It's like
• 18:25
  that's not you already chose the one that
• 18:27
  works super well with Epic. Like, why is
• 18:29
  that gonna be any different? So I I
• 18:31
  think that's the way to think of it.
• 18:32
  Like,
• 18:33
  if if you're having success with the current
• 18:36
  version, then that's great. Like, I'm I'm happy.
• 18:38
  I just want doctors to have have good
• 18:40
  lives and and see more patients. So,
• 18:43
  yeah, I think just what is it? Always,
• 18:45
  like, measure
• 18:47
  think think about the metrics that matter to
• 18:50
  you and then talk to other partners and
• 18:52
  see see if there's a a better option
• 18:54
  out there.
• 18:55
  Be a little adventurous and buy into doctor
• 18:57
  Kelly's sci fi vision. I think that's the
• 18:59
  way to go. Yeah. Yeah. Doctor Kelly, thank
• 19:01
  you so much for your time. It's so
• 19:02
  fantastic to have you. It's a it's a
• 19:04
  great great conversation. Thank you.
• 19:06
  Thanks, Lucas. Appreciate it. And we also want
• 19:09
  to thank our podcast sponsor, Heidi. You can
• 19:10
  tune in to more podcasts from Becker's Healthcare
• 19:13
  by visiting our podcast page at beggershospitalreview.com.