Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Dan Balcauski (00:20):
Welcome to SaaS
Scaling Secrets, the podcast
that brings you the insidestories from the leaders of the
best scale up.
B2B SaaS Companies.
I'm your host, Dan Balcauski,founder of Product Tranquility.
Today I'm excited to speak withMike Meng.
Mike is the CEO and Co-founderof Stellar Health.
A rapidly growing healthcareSaaS revolutionizing how medical
practices deliver value-basedcare.
Before Stellar, Mike was aprincipal at Apex Partners,
(00:40):
where he deployed over 3 billionin capital and served on the
boards of several leadinghealthcare companies.
Mike's combination of deephealthcare investing experience
and entrepreneurship gives him aunique perspective on Scaling
B2B SaaS companies and regulatedmarkets.
Let's dive in.
Welcome Mike to SaaS ScalingSecrets.
Mike Meng (00:54):
Thanks for having me,
Dan.
Dan Balcauski (00:56):
I'm very excited
for our conversation today.
As we were chatting just beforewe hit record, I think you are
the first tech representative inthe healthcare space.
So very excited to tread someground that we haven't before.
Before we do that just you gavefolks a little bit of a taste of
you in the intro, but tell us alittle about your journey in the
(01:18):
SaaS world.
Mike Meng (01:19):
Yeah, so, investing
in a lot of different companies,
again, as you referred to in thehealthcare space, looked at a
lot of healthcare EMR companies,right, which is probably the
closest thing to true SaaS thatI can think of in the space.
Love the space.
I think it's helpful.
We've digitized a bunch ofinformation now, which is great.
And then, I had this problemwhere.
I felt like I was not doingenough to change healthcare.
(01:40):
I think it was great to invest,sit on boards, hopefully make
some money doing that.
But I think that it dawned uponme that I need to do a little
bit more, and the only way to doso is to get your hands dirty,
get in there, really try andchange the company and the
world.
It actually coincided with my,the birth of my first child, my,
my daughter.
And I think that was a changingmoment for me where I said, you
(02:01):
know what?
I life is great and all thatstuff, but I want to truly do
more for the world.
So that's what got me into it.
Dan Balcauski (02:08):
We all have
transformational moments in our
lives.
I think of him as a superherotransformation moment.
I'm Peter Parker's.
Normal high school student, Iget bit by a radioactive spider.
I wake up the next day.
I'm Spider-Man.
What moment was that for you inyour life?
Feel free if there was a, thebirth of your daughter, or maybe
it was some other moment for youon your journey.
Mike Meng (02:26):
There's so many.
I think personally I subscribefully to what you just said,
except I would also say I.
For a lot of people and maybe alot of superheroes.
It's a series of these things.
So I don't think it's one.
I think Peter Parker also hadlike maybe five more right after
being bit by the spider.
So my point here is I thinkcertainly the birth of my
daughter the big issue there Ithink is you really wanna leave
(02:47):
the world a better.
Place than you found it, Ithink.
And so when you have, you'recarrying your child your first
time, you're like, oh my gosh,is the world going to be better
in 30 years or worse?
Right.
And I think those are some ofthose moments I would point back
to though others.
I mean, I think everyentrepreneur that's ever changed
how we do things is amazing tome.
Like a early hero for me wasBill Gates, not because he was a
(03:10):
billionaire, not because of allthese things.
But if you are old enough toremember the green screen and
non gooey interfaces withcomputers, it made computers
only so accessible right toeveryone.
I.
And then this thing calledWindows and maybe it was off of
Apple, right?
We all know these things, butand Microsoft Office, all these
things.
But it made it absolutelyaccessible.
(03:32):
It made it that everyone needsto now use a computer, et
cetera.
We've obviously certain seenthat with the iPhone.
To me, that is the heroism inall this is you're changing how
the world does something in avery profound way.
That's what I think I it reallyis.
Dan Balcauski (03:47):
Yes.
I am old enough to remember thekaban prompt only version of
interacting with computers backin the day.
And yes, I agree.
The, there's only so much.
Of an audience you're gonnaattract with that sort of
interface.
So well, well, well noted there.
Well look I, I do wanna talkabout Stellar Health.
Before we kind of dive into themeat of the Scaling, the
(04:09):
company, et cetera, can you justgive us kind of a 32nd overview
of what is Stellar health andobviously plays in the
healthcare space.
But what is the mission, vision?
Why, what, how'd you end upthere today?
Mike Meng (04:19):
Yeah, start with a
little bit with the product real
quick.
Is that.
When we ask these doctors to dowhat we call value-based care
activities or value-based care,they're supposed to do all these
extra things that are better foryou, your patient, to keep them
healthy instead of worryingabout taking care of them when
they're sick.
The problem we have inhealthcare is we don't pay them
for that.
We actually just pay them fordoing more volume today.
(04:40):
So what Stellar does is wepartner with health insurance
companies, we get some moneyfrom them.
We pay that out to the doctorswho really make that happen.
So example, perfect example.
Dan, have you ever been on abreast cancer walk where you
raised some money for breastcancer or someone who had breast
cancer?
I.
Dan Balcauski (04:55):
Not me
personally, but I know people
who have.
Mike Meng (04:57):
So pretty popular
thing to raise money for.
Right?
And it is an important one.
The crazy stat is women over 50should be getting a mammogram
every other year.
Yet 35% of the women in thiscountry of that age do not do
so.
So it's kind of one of theselike sad things where you could
catch the breast cancer early.
That would really save lives andprevent costs, and we gotta do
more of that, right?
(05:17):
And so what Stellar does is wepay doctors whenever they get
things like that done.
And that's super important tokeeping people healthy and alive
and having a good quality oflife rather than in this, in the
hospital sick bed.
Dan Balcauski (05:30):
So helping to
incentivize some of the
proactive behaviors that can bemuch more cost effective than
the reactive ones when thereactive ones tend to be, I
guess, high dollar value, much
Mike Meng (05:45):
too late.
Like a little, too little, toolate, right?
Yeah.
Dan Balcauski (05:48):
Lower risk of
positive outcomes and more
expensive for that.
So a whole bunch of bad mix.
So going back really appreciateyou helping to move that there.
I guess, you, you did spend sometime on the investment side of
the fence, I guess, and nowyou're, leading, growing,
stellar health.
I guess what surprised you?
(06:10):
Most about being on the otherside, sitting at the CEO chair
that maybe you wouldn't haveexpected being in the in the
investor seat.
Mike Meng (06:17):
Yeah, there's so many
Dan here, so we could spend the
whole time on this alone, butwhat I would say is how much
people matter, right?
I always cared about people, andI'm kind of proud of the fact
that even in investing I wasviewed as a good manager and a
lot of the people who work forme wanted to work for me again.
So some of'em actually joined mehere at Stellar and whatnot, so
I'm very proud of that.
(06:38):
I think.
But when you have a company andyou have all these employees,
like just every employeematters, all these people are
real.
Like you can't just make thesedecisions and think that they
don't touch real people.
Like that is the most profoundpiece of this is how responsible
you are for these people as awhole.
I think there's further lessonslike how to actually actualize
(06:58):
the thing you're trying tobuild.
Another big one for me lately isif you've ever read like Paul
Graham essays on Maker Time.
Like, I increasingly clear mycalendar and want just blank
maker time.
'cause I think it's superimportant as you're building
something versus just beingscheduled back to back and being
someone who assesses something.
And I think that's actually adecent difference that I've even
(07:20):
myself evolved a little bit moreinto.
Dan Balcauski (07:22):
Well look, every
CEO faces Scaling challenges
along the way.
Is there sort of a particularchallenge in the course of
seller health that's reallystood out to you?
A maybe a crucible moment that'skind of made the company who,
what it is today?
Mike Meng (07:35):
Again, so many to
name.
I mean, just to tell a storyreal quick, early on, trying to
sell to health insurancecompanies and tech is just one
of the craziest things.
I remember when a lot of ourearly investors were like.
So Mike, let me get thisstraight.
You want to sell to doctors andthen really insurance companies,
(07:55):
and then you wanna hope that allthese people are gonna do all
these great things, do thoseactions that we just talked
about, and you're gonna get somebig, like result from this.
Like, how long is that gonnatake?
How long is the sales cycle?
And I, that was frankly nuts,right?
I think the average sales cyclein our industry is maybe
somewhere between nine and 15months.
(08:15):
So we went.
15 months without a sale.
Right.
At all.
And I think eventually we'revery proud.
A big, large national healthcarehealth insurance company gave us
a shot.
We proved it to them, and thenwe kept going from there.
But I think there are, therehave been those dark moments
where, you know, one of myco-founders at the time said
like, I don't know if this isgonna sell.
Like, I just don't think they'regonna buy someone, buy something
(08:36):
without credibility.
Like I, I mean, it's not that itdoesn't work, it's that no one's
gonna give us the shot at it.
And so, dark moments like thatcertainly exist.
I'd say the other thing that'sreally difficult in our space,
healthcare particular is howmuch of it is truly driven by
technology versus still peopleand processes is a very big
question.
Right?
(08:56):
I think the only pure SaaS inthis, in our world actually is
the hospital EMR system, right?
They're epic, Cerner's, Athena'sof the world.
Those are all real, but I wouldsay that a lot of the other
solutions that are technologybased.
Have not truly been SaaS.
And I think that's one of thechallenges we face this
challenge, frankly, which is youcan't just do this without any
(09:20):
people.
Like, it's not like people justinstall the software and start
picking it up.
And I think that is one of thechallenges.
There's going to be some peopleinvolved.
How do you make it so that it'sstill software driving the vast
majority of the outcomes ratherthan necessarily people?
Dan Balcauski (09:35):
Well, you dropped
a whole bunch of good sort of,
outlines and breadcrumbs therethat I don't know if we'll maybe
have time to follow up on all ofthem, but, so just within there,
healthcare is, obviously it's ajuicy market.
We're, I don't know, it's what,17% of the US GDP or something
like that
Mike Meng (09:51):
$4 trillion.
Dan Balcauski (09:52):
So it's giant.
But, it's also notoriouslydifficult.
It's highly regulated.
It's conservative has dominatedby large, incumbents, the epics
of the world, et cetera.
Right.
And can be very difficult as anupstart.
And you mentioned.
You know this kind of slog ofthe sales cycle.
I guess, what's been, I guess,most surprising to you about
(10:12):
getting doctors and medicalpractices to adopt, new
technology?
Like what have you kind oflearned on this journey?
Mike Meng (10:20):
I mean, you've hit
some of the nails on the head,
but I will expound.
Absolutely.
So first and foremost, there'sbeen.
A myriad of solutions thrown atthese doctors, in some cases
shoved down their throat, right?
So they didn't necessarily askfor it.
Obviously there were a lot ofsubsidies to do so, so that's
been a little tough.
I would also say that doctors ingeneral, and rightfully so, have
(10:40):
a lot of doctors in my familyare pretty wary of.
Of outsiders and business peopleit's a lot of like, what are you
going to sell me that's gonnarip me off now?
And that's a tough mindset tochange.
Right?
And on top of that, they'repretty burnt out.
They've been asked to do a lotof work, a lot more
administrative burden.
And that's one thing that wethink a lot about in our product
design, which is.
Like sometimes I don't even likewhen people think about zero
(11:03):
clicks or let's try and reducethe clicks.
Reduce the clicks.
I think of it as how can we dothis without clicks?
Like how can you just not haveto do some of this stuff?
Is one way I think about thisthat helps.
But I think it's just, it's atough one where we've asked too
much in terms of usage ofdifferent software and throwing
these things at them that a lotof our doctors don't want
another thing.
(11:23):
So that's one of the headwinds.
And then on the other side ofthe industry, on health
insurance.
It is insanely conservative,right?
And rightfully so, probably inmany ways, these are people who
you're playing with people'shealth and lives, right?
So you gotta be really carefuland respect that.
I would say I wish that theindustry saw a little bit more
whenever there was like somereal proof points and traction
(11:46):
though.
That they picked it up a littlebit faster and scaled it a
little bit more.
That's the area that I think wewant to prove out that I think
hasn't been there.
Which is that this these resultsare repeatable, they're
scalable, and they're systematicrather than just it worked here
in this one moment.
That's the big problem inhealthcare.
A lot of academic studies, yougo to, like any of the PubMed
(12:08):
and just read academicliterature, but the problem is
they always describe like oneinstance, one place where it
happened.
And I think that's the thingthat we're trying to tackle at
Stellar, which is how can youmake this nationwide everywhere?
It always works.
Dan Balcauski (12:22):
So a couple
things you mentioned there,
right?
I mean, a lot of them on mindsetthey've been, they've been
burned or feel like they've beenburned at least, right?
Both of those perception isreality.
So, so some high skepticism.
And they're already dealing witha lot of administrative burden
or more so maybe than theysigned up for it when they were
going through medical school andmaybe starting their.
(12:42):
Their practice whether that's,government enforced or, enforced
just by the realities of themarketplace.
So, but obviously you've hadsome success, you got past that
15 month sale.
I guess what approaches have youfound successful in building
trust with these folks despitethat mindset?
Mike Meng (13:00):
Yeah, I think a big
piece of this is, first of all,
credibility.
You do need it in healthcare.
Track record matters.
So like, one of the things is Ido tell my team like, we gotta
go nine and oh or 10 and oh,like, you can't just have like
a.
Eight and two record, and thinkyou're okay.
And so I think like we reallywanna make sure we're always
delivering everywhere.
That's a big piece of it.
We have credible investors,backers people on our board of
(13:24):
advisors.
That always helps too.
And then I think one big piecethat has helped us a lot is.
Being pretty transparent aboutthings.
So you won't see people atStellar.
This is maybe a cultural point,but you won't see us
misrepresent things.
You won't see us try and stretchthe truth.
I'm a big believer inintellectual honesty and just
telling it how it is or where itis.
This starts in the company, sowe always do it internally.
(13:46):
We tell people, here's where westand, here's how we're doing.
We share it pretty openly.
And I think that permeates outthere.
So you don't have the, doctorswondering.
What's going on?
Like, are you sure you're nottricking me?
And all this stuff.
Like one example is in oursystem we pay out these things
called stellar value units.
SVUs, every time you get one ofthose mammograms done I'm a big
(14:07):
believer in the like wizzywigconcept here of what you see is
what you get.
Meaning if there's a bug andsomething happens where what was
displayed was wrong and we kindof mess that up.
You're gonna see us eat it justlike a really solid four Seasons
hotel or a credit card that's onus.
We made the mistake, I'm goingto eat that.
I don't go back and try and likefight you on it, et cetera.
(14:28):
So kind of what the app shows isreally trusted.
I think that's a reallyimportant point of like
preserving that trust in thesanctity there.
And that's something that'sreally helped us as well.
Dan Balcauski (14:38):
I that reminds
me, I think there was an old I
think it was it Nordstrom's orMacy's?
I think it might've beenNordstrom's, where they had,
they will accept any returnpolicy and there's some
legendary stories about themaccepting returns for stuff that
they didn't even sell justbecause that was their
commitment to making sure thecustomer felt happy about the
situation.
So having that demonstratedability or willingness to make
sure the customer walks awayhappy, even if it.
(14:59):
Even if it wasn't your fault,right?
Mike Meng (15:01):
That's exactly right
and any time one of our
customers says, Hey, can I get areport on all the transactions
that were earned?
We have that ready on thefingertips, that's ready to go.
You want to audit it?
Like we're not, we don't want toplay the game of, well, let's
move things around.
Let's make some mistakes here.
This is pretty serious stuff, sowe're gonna be pretty
transparent about all thosethings.
Dan Balcauski (15:21):
And so I, I am
curious too,'cause we didn't
really touch on this before, butyou hinted at it where you do
have, you got the insurers theinsurance companies and then
you've got these the doctors andmedical practices.
And.
Then you guys are sort ofplaying between those two.
Is that, am I understanding thatcorrectly?
Because that's a little bitdifferent, right?
(15:42):
I mean, for most, most ask, CEOshave to think about one
customer, and maybe they've got,the marketing sort of IC user,
and then they've got their CMObuyer, right?
So they've got like thatdifferentiation, but you're
actually kind of straddlingmultiple organizations in your
go-to market.
Am I understanding thatcorrectly?
Mike Meng (16:00):
absolutely correct,
and maybe Dan, let's step back
for one second, right?
So if you want to servehealthcare, you have to
recognize that there's always anextra business partner.
At hand.
What I mean is most of us, youand me, we don't pay for our own
healthcare like usage directly.
We pay for health insurance,which then doles out that money
(16:20):
to these doctors.
And so there's an intermediarythat always exists here, and
it's an important one becausewhat it leads to is.
Every single situation inhealthcare real healthcare is a
three-Way program.
Like no matter what you want tosay or do, just like you have to
then get this big deductiblequestion or the bill like that
happens, and that's part ofhealthcare.
(16:42):
It's a big piece of it.
And so just remember thatthere's always these three
partners, and then what youdescribed is exactly right.
Therefore, we have to sitsomehow between those two.
In the end, our health insurancecompanies stand to gain if all
these quality things get doneand they get kind of more
dollars for having being higherquality and all these things.
And we then want to pass thatmoney to the doctors who are
(17:03):
making that happen.
The other interesting dynamicthat happens is the health
insurance companies, as much asmany of them have moved further
and further down towards thecare delivery space, which I'm
supportive of.
At the end of the day.
Most healthcare is stilldelivered by doctors like you
and me, or that you and me gosee.
And so, health insurancecompanies need to find a
mechanism to actually get them,to compel them to act in this
(17:28):
better way.
And so that's a core part ofthis is that health insurance
companies cannot do it on theirown.
They have to actually go rewardpeople or compel others doctors
to go do it.
And that's what we reallyenable.
We sit there particularly makingthat happen.
Dan Balcauski (17:43):
So a and I guess,
you know what, and help us
understand a little bit.
'cause this is a reallydifferent dynamic than I think a
lot of other companies in otherspaces have to deal with.
Like how, as you were kind ofgrowing stellar health at early
days, like how did this.
Make change how you thoughtabout your going to market.
Like, it's like, okay, we,because I guess one another
(18:07):
model, right?
You've got you think about liketwo-sided marketplaces where
I've got like Uber, right?
I've got the drivers and the theriders, right?
And I've got a match supply anddemand, right?
And so, you end up with thisthing called this Cold start
problem, where it's like, allright, do I go just try to get a
bunch of drivers and then, in acity, and then I try to worry
about getting a supply later or.
Or vice versa.
It's not exactly the same thingyou're in, but I can imagine
(18:28):
you've got some kind of similardynamics.
Help me understand how thataffected how you guys thought
about growing in this space.
Mike Meng (18:35):
And Dan, you hit the
nail on the head.
I think in many ways we are atwo-sided market.
We did face a cold star problem,so I'm actually now a incredible
student of all two-sided marketsever built.
I've read every book andacademic paper you could think
about.
You can find on this stuff andyou're absolutely right.
That is true.
And what's really odd andinteresting about our business
is we are a two-sided market.
And then the real challenge inScaling this is we need to
(18:59):
efficiently monetize on the factthat we have the two sides in
particular.
And that is the SaaS part ofthis.
Right.
And I think one of the things,just to answer, kind of some of
the thoughts earlier on some ofthe challenges, I'll come back
to one, which is sometimes ourhealth insurance companies.
Made us go too spread thin andwide.
(19:19):
So they were like, why don't yougo here?
Why don't you go there?
You can imagine being asked togo to Alaska and Hawaii as well
as where we are today.
It's like you're starting tospread your resourcing very thin
if it's not pure software.
And that's what I was referringto.
And so it actually led to I'dsay some profitability
challenges in those situations.
And we've now come back aroundto say, let's actually make sure
(19:41):
we exploit the solidrelationships we have between
payers and providers where we'realready connected and let's
actually add to that and reallyadd more SaaS to those
situations.
So kind of going deeper witheach provider rather than just
going so wide.
That is actually one of theproblems we faced.
And we've actually been, kind ofmore focused down, so there was
(20:03):
a period in which we grew reallyrapidly.
From 2000 providers doctors to10,000 in one year.
And, that spread our teamincredibly wide and thin.
And now we've kind of sloweddown the number of providers
we've adding.
We're still adding them, butjust a little bit more careful
about it while adding more SaaSto the existing base.
(20:25):
Instead.
It's actually what that resultsin is it's better for the
doctors.
They actually get more runningthrough the same stuff.
It's better for the insurancecompany'cause they get better
performance and results overall.
And then better for us becausewe're more effectively and
efficiently covering the fewernumber of doctors out there that
we work with.
Dan Balcauski (20:44):
That's so it's
fascinating.
So you have, so yeah, there maybe geographic dispersion, kinda
using your Alaska, Hawaiianalogy.
But there could also be otherkinds of services or other,
workflows, et cetera, thatmaybe, you're.
It, I used to be, had a productin a two-sided marketplace, and
there's a lot of work that I hadto do with product was a lot of
just digitizing what otherwisewere just Zendesk tickets.
(21:06):
So there's a lot of humans inthe loop.
So I can imagine you get, as youget spread out no matter how it
might be that, you you're havingto add people instead of adding,
Mike Meng (21:14):
That's exactly right.
You literally understood it.
Exactly.
And that's where, again, we'restill growing now, went up to
about 16,000 doctors that useit.
So we are adding.
But we're just a little bit morecautious about it and therefore
concentrating more of it on theexisting part of it is that's
also where like, the more youadd density on an existing
doctor, I.
(21:35):
The more mindshare you get oftheirs and they get more excited
'cause they're earning moremoney from doing this.
So it has all these like win-winwin benefits for doing so.
So that's the thing that we kindof watch for.
In, in two-sided markets speak,it's kind of like the power law,
the Pareto effect.
Here's the 20% that really drive80% of all the rides.
We think about that evermore andreally focus on the right
(21:55):
providers that that are reallyengaged and working pretty hard
with us.
Dan Balcauski (22:00):
There's so many
threads here.
You've mentioned kind of thismonetization aspect.
I do wanna touch on kind ofpricing'cause Yeah, it's,
healthcare is so weird where youhave these.
Crazy economic situations thatdon't really exist in many other
markets.
Right?
Like if I my doctor says, Hey,Dan, you have to get this test.
It's like, okay, I guess mydoctor's helping me get this
test, but it's the only serviceI ever.
(22:22):
Maybe I even go get the testbefore I even know what the
price is.
And that's completely differentthan any other Yeah, that's
different than any othermarketplace.
Right.
And I mean, I know there's beenattempts to try to solve that
and get transparencies forthings like, oh, like where can
I shop around and get an MRI,what's, what kind of the most
cost effective or other thingslike that, I guess.
Talk to me a little bit aboutwhat you've learned, because I'm
(22:44):
trying to also tie that, I guessbefore you were talking about
these these value units thatthese credits that you're
actually incentivizing on thehealthcare providers.
How, tell me what you've learnedand how did this come to be
where you're, it sounds likeyou're,'cause you're taking
money in and then also paying itout to these providers as well.
Help kind of tie this togetherfor me as with this weird
economic situation we've beendiscussing.
Mike Meng (23:06):
Sure.
So the good news is maybe wedon't run inside that normal
part of the, paying forhealthcare part of it.
We actually purposely by designwanted to operate outside of
that in this area called morevalue-based care payments.
So we kind of keep the, what youjust said to the normal payer
provider relationship.
(23:27):
Usually it's be, it's prettycontentious.
People fight over that point, soI don't want to get involved in
there.
Instead, this runs through whatwe call value-based care
dollars.
It's a separate point.
And to your exact point, what weare doing is we're paying out
anytime they're doing the rightthing overall for the patient.
These are, by the way,government defined star metrics
(23:49):
quality.
They're not made up.
They're kind of pretty wellestablished.
I think the thing we wish we didmore of is more of it as a
whole.
And so that's what we're reallyincentivizing these doctors to
do it.
And I think what it leads to is.
The doctors are actually moreengaged in doing this if they
see the immediate reward fordoing so, versus the way it's
(24:09):
kind of paid for today iscontractually, if we all hit our
stars the next year, you'll getthis big bonus check, which is
like kind of crazy, right?
Like, who's gonna wait for abonus 18, 24 months from now to
like get the feedback loop thatit worked.
What Stellar's really doing isit's giving you the feedback
loop sooner so you know that's,you want to do more of that
stuff.
(24:30):
And that's what gets the kind ofengagement level and kind of
results that we see here.
And then I'd say the reallyinteresting part is we get
measured in our industry quiteheavily on ROI, meaning it has
to drive better quality, right?
It has to drive thosemammograms, it has to, you have
to hit those things.
And so we're measured prettyclosely on these things.
We're very proud of that.
(24:51):
That's like a piece where Ithink relying on this amorphous
ROI over the next 10 years isnot something that's gonna win.
No one can put, on their fingeron that.
Whereas here, what we do is wefocus on like immediately
driving the quality stars up, etcetera.
And that really helps again, thestory and whatnot.
It's actually what our buyer, Ithink is buying ultimately, is
(25:11):
they're trying to make sure theybuy that engagement.
From the doctors that they dothese extra quality things,
which again, in turn leads tothat ROI.
Dan Balcauski (25:20):
Well, I'm
actually very curious about this
ROI conversation.
So, like how do you think, or,maybe this is just sort of
dictated on, on above or like,how do you think about
demonstrating ROI in a way thateveryone can trust?
Because you said like this iswhat the people are actually
buying.
Like, is there.
Like, is that, are there justways that like sort of
(25:40):
government mandates, everyonesort of has these ROI
measurements or are there waysthat you've sort of built the
company to help make that moreapparent?
Mike Meng (25:47):
Yeah, we follow like,
kind of again, a lot of the well
traveled paths of the governmentdefined points.
But like an example would be thegovernment does give a bonus if
you get to, let's call it fivestars on mammograms.
The example I've been using thatmight be you have to get out of
your population.
You need 93% of those women toget that mammogram done, right?
(26:09):
So it really matters to get 93%.
And maybe the population startsat 60%, meaning out there in the
wild, if you do nothing at 60%,our job is to move that 60 to
93% by prompting these actions.
And when we do government'shappy,'cause you hit their star
metric.
Insurance companies happy'causethey got rewarded for that.
(26:31):
And that is fairly measurable inour industry.
Dan Balcauski (26:34):
That's
fascinating.
'cause I mean, it's one of thesethings that, I think you have a,
a.
Well, some would call it ablessing, some would call it a
curse.
Like the government is sort ofhelping define these metrics,
but they are like, they're veryclear in terms of how the end
person gets value.
And I find that for so many SaaScompanies, that's just so fuzzy.
(26:55):
So I think that there's.
Even if you're not inhealthcare, I think there's so
much value of looking at, whatcompanies like you or, other
companies in the value-basedcare space are going through.
Because there really is aconcrete way to sort of think
about, the ultimate outcomesthat you're driving for
customers.
'cause it's like, oh, well we,we help you be more efficient.
Right?
And you're like, well what doesthat mean?
Like,
Mike Meng (27:14):
Well, and I think
you're right.
I think you're right, right.
So we use Slack in our companyand love the company, and we use
it.
I'm not sure I get exactly whatthe ROI is for us having me on
Slack.
Right.
I don't really fully understand.
I know we, I guess we have tobuy it'cause everyone's using
it.
Right.
Like, but like, I don't fullyget it.
In our industry it's, I'd sayvery dollars and cents.
I mean, it is like you get.
(27:36):
A certain extra bonus paymentfor hitting certain metrics.
These metrics, by the way, arealso defined from academic
literature.
So I mean, there's a lot ofsmart doctors who've studied
what works and what doesn'twork, and then help the
government define these things.
So, yeah, I think in our,because of that we have to live
or die by like the real resultsin the real world.
It's like an example.
Another example would be.
(27:57):
Smoking cessation, we gotta geta certain number of patients to
quit smoking.
Like that's just real.
Like you gotta get thosepatients off the smoking
cigarettes or off of theopioids.
You don't really get a choice tosay yeah, I think I helped in
general on this one.
Right?
Like it's literally counted inthe number of people still
smoking or the number of peoplestill on opioids.
Dan Balcauski (28:18):
Well, I, so one
other thing I wanted to touch on
in this world of healthcare, isinnovation kind of running into
a highly regulated industry?
Like how are you thinking about.
Being able to actually, make achange in this, in these
industry when, it seems like it,it's heavily locked down.
(28:41):
And, you have everything fromROI measurements dictated by the
government.
So, how have you sort of thoughtabout that, and I guess, what do
you see going forward for waysthat the industry could get
better about this?
Mike Meng (28:50):
Yeah, I mean, I wish
the government would look at
this a little bit more in termsof helping facilitate safe but
innovative ways to attack allthe problems we've talked about
today.
And what I mean by this is, I.
I respect.
Again, we're talking aboutpeople's health here.
We gotta be very cognizant andcareful about what we're
touching and changing here.
(29:12):
But on the other hand, we needthis innovation, right?
You referenced it, buthealthcare is becoming 20% of
our entire GDP.
It's$4 trillion.
We gotta contain it.
We gotta figure out a way sothat this doesn't get out of
control.
And I think the best ways we cando so are gonna come from
innovative entrepreneurs outthere, and it has to be probably
technology driven.
(29:32):
I always say this too.
Technology is actuallydeflationary for almost every
industry.
I.
Except for healthcare and likeeducation in which technology
tends to be inflationary, whichis kind of a problem.
And I think that's, it'spartially driven by we have to
be thoughtful about where we cantake out cost burden, whether
it's fraud, waste and abuse,ways to apply technology to that
(29:56):
ways to make sure we actuallyare getting the best care to the
right people whatever it may be.
And I think one of the bigthings, right, I have like, as
a, as an issue is.
I think we need to update someof the laws that we have around,
for example, hipaa, which Ilove.
hipaa.
HIPAA privacy, healthinformation privacy is so
important.
Protective health information.
(30:17):
PHI is super important.
We should all care as Americansabout privacy, but we should
also be thoughtful about how canwe use the data to come up with
more innovative ways to attackthe problem, and especially with
the advent of ai like.
I think we really need to thinka little bit harder about this
one.
Again, protect people's privacy.
(30:38):
Yet make those tools and abilityavailable to people.
That is, I think, where we needto play.
And again, it just it's a lawThat's, it was a great law.
It's a little outdated today.
We gotta update it so that webring it into the future.
If you follow the machinelearning AI stuff.
It's incredible what it canfigure out quickly.
(30:58):
Even on like contracts and lawright?
Or other domains.
Like we need to apply that hereon healthcare.
And let me be clear about whathere means.
I think on healthcare deliveryin general, I think there's a
lot of use of AI and likebiotech and the discovery of
drugs.
That's great.
And like we need to keep doingthat.
But I'm talking about just thegood old, when you go see a
(31:19):
doctor, how do you get paid?
All those different things.
There's a lot to be fixed therethat I think we need to attack.
Dan Balcauski (31:27):
Well, it just
just to give me a little bit
more, or audience a little bitmore concreteness around that,
like, so let's say, HIPAA gotmodified, got changed.
Like what are the applicationsyou see in, in healthcare
delivery that AI could reallyunlock?
That something that like thecurrent regulatory regime is
sort of preventing.
Mike Meng (31:45):
Yeah, I'll give you a
really simple one.
Today we allow kind of a, aninteresting thing that happens
is like two medical devices ortwo different drugs that compete
in say the same category, wekind of just go with like, they
kind of put out some studies,they kind of just say, yep, I do
it better.
And then they market the hellout to the doctors, right?
And then some of it on even ads.
(32:06):
And then the question for you asthe ultimate patient or
recipient is, which one shouldyou get and why?
Like, did you get that decision?
How did you make that decision?
How do we as a society make thatdecision?
And I think that's one where ifyou ran AI or machine learning
on all the outcomes that haveactually happened, when people
have installed this in like saya pacemaker into your heart,
(32:26):
like you would see one of'emactually is better than the
other one.
Let's go with that one more,right?
Like, I think those are justlike real ways that you kind of.
Get to decide these things.
And I think the data reallymatters on really long
timescales, really large amountsof data.
That's the kind of thing that Ithink is really interesting
about this stuff.
It's looking at the real worldevidence of like what happened
(32:49):
in the situation when peopletried one therapy versus the
other therapy.
Dan Balcauski (32:54):
Yeah.
And there's so much have I'vespent enough time on different
subreddits for, for peoplediscussing, different
medications.
You're like, oh God, I this isnot the way, we can't be, we
can't be relying on this.
There's gotta be a better wayto, to sort of surface like
what's the the best mix.
And yeah, I mean, you alreadymentioned, I mean, doctors are
already stressed.
They try to keep up to date,but, there's, I mean, how many,
(33:15):
medical papers and innovationsare happening every day.
There's no sort of perfectsource of information.
So, is there a way we couldsmartly unleash the AI to help
us
Mike Meng (33:25):
Well, there's an
answer to these things.
That's the thing, right?
Like it exists out there.
We just don't know the answeryet.
And so instead we ask all thesedoctors to keep up with
continuing medical education andreading all these papers and
trying to keep it in theirheads.
Like that just sounds kind ofinsane when you step back and
think about that.
I.
Dan Balcauski (33:41):
Well, Mike I
could talk to you all day, but I
do have to start wrapping it upto be respectful of your time
and our audience.
Are you ready for some rapidcloseout questions?
Mike Meng (33:48):
Yeah, let's do it.
Dan Balcauski (33:49):
Alright.
How do you think about definingsuccess for yourself today?
Mike Meng (33:55):
A call back to what I
said earlier.
I care that when I leave thisearth that we positively
impacted as many people.
Out there as possible.
And that's all the way from myown children to, the basketball
game, my plan and the weekends,which we like each other a lot
to hopefully changing an entireindustry.
So I think all our goals shouldbe trying to leave the better
(34:17):
the world, a better place thanwhere we found it, as much as we
can.
Dan Balcauski (34:21):
Amazing.
When you think about all thespectacular people you've had a
chance to work with, is thereanyone that just pops to mind?
Is that a disproportionateeffect of the way that you think
about building companies?
Mike Meng (34:31):
Yeah, I mean, look, I
think we're coming up on
Thanksgiving, right?
And I think gratitude issomething that's absolutely
important to me and the values Ihad.
I do want to thank, like myparents, my mom and dad,
they've.
Always supported anything Iwanted to do and made sure that
even in failure I always sawthat there was light at the end
of the tunnel and.
(34:52):
Get back on the horse, which Ithink in today's world is a
pretty hard lesson.
I worry about it with my ownchildren.
In terms of making sure theyinternalize this.
If you've ever seen, there'sthis great Angela Duckworth on
grit, right?
And you really want to teachyour kids grit.
And I believe in all thosethings and I think that's where
I think my parents reallyinstilled that in me and my
brother and I.
That's like a gift that I don'tknow.
(35:13):
I hope I live up to that with mykids.
Dan Balcauski (35:16):
Installing
resilience in your grit in your
kids.
I love it.
And thank you to your parentsfor helping put that into you.
So, look, being a CEO foundercan be challenging spiritually,
emotionally, physically.
Are there any habits you'vecultivated to help you stay at
the top of your game?
Mike Meng (35:33):
Sure.
And here's a couple things thatI will impart that I think are
important.
I am also a subscriber intomaybe Simon Sinek, the Why, but
we actually talked about this asa company, but in the darkest of
moments when you're not surewhere you're going understanding
what is it that motivates youand why you're doing it is just
absolutely critical.
So I leave that as an absolute,just figure that out for
(35:55):
yourself.
It's important.
It's really the intrinsicmotivation.
A second one is relationshipsmatter.
This is actually a stellarvalue.
But I've thought about this alot.
Why is it that relationshipsmatter?
They matter, obviously becauseof all the things you can think
of, of why they matter.
But I would also add thathealthcare, and I'd say a life
is a repeated game.
What that means is you're notjust here for one transaction.
(36:16):
You're not here for the nextyear's transaction.
The reality is you're here for alifelong of continuous
interactions.
And so you have to choosewhether you want to screw one
over and leave it that way, orif you believe what I believe
it's that these relationshipsmatter.
It's a long-term discussion.
Dan, you and me, even like, weend up not working on together
on something, but we stayed intouch and look where we are
(36:38):
here, right?
So I think these things matterand that's something I think to
live by.
And my final thing for anyonewho's looking to start a company
out there or a CEO is I'velearned the hard way that saying
no to doing things.
Is way more important to sayingyes.
That's how you stay focused.
Like eventually there's a myriadof things you can say yes to.
Yes to this, yes to that.
(36:59):
And actually more important islearning how to say no.
And again, learn that the hardway and really find that
important.
Dan Balcauski (37:05):
Some great
nuggets there.
And man, yeah we could haveprobably full episodes on each
one of those but we'll leavethem there.
For our listeners, this has beenfantastic.
Mike.
If folks wanna learn more aboutyou or Stellar Health where can
they do that?
Mike Meng (37:19):
Stellar health.com is
great and our team is looking
forward to partnering with youin every community to really
bring these value-based caredollars to make the healthcare
better.
Dan Balcauski (37:29):
Well, as an
American citizen who is paying
significant money in thathealthcare system, we are all
rooting for your continuedsuccess.
Everyone that wraps up thisepisode of SaaS Scaling Secrets,
thank you to Mike for sharinghis journey, insights, and
valuable tips for our listeners.
If you found this conversationas enlightening anxiety to
remember, subscribe If you don'tmiss out on future episodes.