April 14, 2025 • 43 mins
Technology in workers' compensation claims isn't about replacing humans but augmenting them to work more effectively. Sam Neer from Berkeley Alternative Markets Technology shares how technology is transforming claims handling while keeping adjusters at the center of the process.
• Machine learning identifies potentially serious claims early by analyzing first reports of injury
• Technology excels at identifying patterns across millions of claim notes that humans could never process
• Digital payments and texting capabilities are improving claimant communication experiences
• "Human in the loop" models keep adjusters involved while reducing mundane tasks
• Cloud computing is making systems faster and more reliable with better uptime
• Technology teams need to spend time with claims staff to understand their daily processes
• Effective communication during system outages builds trust and transparency
• Balancing tactical needs with strategic vision helps prioritize technology development
• Incremental improvements often yield better results than waiting for perfect solutions
I encourage you to like, share and follow the podcast, as well as to leave a five-star review so we have an opportunity to help others find this podcast and enjoy the content we've created. Remember to do right, think differently and don't forget to care.
Season 9 is brought to you by Berkley Industrial Comp. This rebroadcasted episode is hosted by Greg Hamlin and guest co-host Matt Yehling, Directory of Claims at Midwest Employers Casualty.
Visit the Berkley Industrial Comp blog for more!
Got questions? Send them to marketing@berkleyindustrial.com
For music inquiries, contact Cameron Runyan at camrunyan9@gmail.com
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:13):
Hello everybody and welcome to Adjusted.
I'm your host, greg Hamlin,coming at you from Sweet Home,
Alabama, and Berkeley IndustrialComp.
And I'm excited to share withyou this rebroadcast.
One of the hot topics in theindustry has been technology and
how technology is transformingthe insurance industry,
(00:34):
especially on the claim side,and in this episode we had the
opportunity to talk to Sam Neer,and Sam has a lot of experience
when it comes to the claim sideand technology, as he manages
quite a bit of this as a projectmanager for us, and so I really
enjoyed this dialogue as wetalked about how we go through
that process, whattransformation looks like, what
(00:58):
opportunities could still beahead of us and what companies
are doing to leverage that onthe claim side.
So I hope you enjoy thisepisode.
As always, I encourage you tolike, share and follow the
podcast, as well as to leave afive-star review so we have an
opportunity to help others findthis podcast and enjoy the
(01:23):
content we've created.
And, as always, I remind you todo right, think differently and
don't forget to care.
Enjoy.
Welcome to Adjusted.
I'm your host, greg Hamlin,coming at you from beautiful
Birmingham, alabama and BerkeleyIndustrial Comp.
And with me is my co-host forthe day, matt Yelling.
Speaker 2 (01:42):
Hello everyone.
This is Matthew Yelling, comingfrom St Louis, Missouri, along
the banks of the mightyMississippi.
Speaker 1 (01:49):
So glad to have Matt with us.
Today.
We actually have a specialguest with us.
That is with Berkeley.
Actually, it goes by theacronym BAM Tech, but Berkeley
Alternative Markets Technology.
Our special guest for the dayis Sam Neer, who is our group
product manager at BerkeleyAlternative Markets Tech.
So, sam, could you say hello toeverybody.
Speaker 3 (02:11):
Hey everybody, Very excited to be here.
Greg and Matt, and again I'mbased out of Raleigh, North
Carolina, where it's a littlebit warmer not as warm as
Birmingham, Alabama right now,but still there.
And I will say, no one wants tosay Berkeley Alternative
Markets Technology over and over, so we might just call it BAM
Tech just for all of our sanity,if that's okay.
Speaker 1 (02:30):
I think that's wise.
I think that's wise.
So the topic today this issomething we've never really
tackled in the nearly threeyears that I've been doing
adjusted is claims andtechnology and the role that it
plays in helping us do our jobs.
So I wanted to bring Sam intoday to talk to us a little bit
about that, since he is part ofthe Berkeley company that
(02:51):
manages our technology to helpus be successful.
Before we dive deep into thetopic, Sam, I wanted to ask you
a little bit about yourself, soI'm sure that you knew you were
going to be in workers'compensation, right?
Speaker 3 (03:05):
Greg, you knew it it's like.
And coming up as a small boy Iwould say to my dad with a glint
in my eye one day, dad, I cando work in workers' compensation
and a quippy line I'll have iseither the way I ended up in the
industry is either I found myway into making great technology
products with greatorganizations and really
affecting change, or I followthe goal of becoming the most
(03:25):
stable, uninteresting version ofmyself.
One of the two, and again maybeit's a little bit of column.
Speaker 1 (03:30):
A, maybe it's a little bit of column B hey there
is a lot of stability inworkers' compensation insurance,
so that's the one positiveright, it's something to be
accounted for.
Speaker 3 (03:41):
So, with the 30-second background again, I've
been working in digitaltechnology for the last 12 years
in product management, Startedoff in PR and analytics, then
moved to telehealth right at theheight of the pandemic Would
not recommend trying to jumpinto telehealth at that time and
I've only been in the insuranceindustry for a couple years now
.
But at the same time I'vereally just had a baptism by
fire but learned a lot and Ithink insurance is a great area
(04:03):
in SureTech that's having a lotof new frontier, a lot of old
legacy systems upgrading.
So it's a great time to bealive.
And again, my specialization isclaims.
I'm beginning to dabble alittle more on the policy and
underwriting side, but thetrends are out there that are
impacting all of us and reallyexcited to talk those through.
Speaker 1 (04:21):
That's awesome.
Well, I know the headlinesright now.
When you hear about technology,you hear about the mass layoffs
of companies like Twitter andFacebook and everybody else
Google.
So there really is something towhat you as you were joking.
There is some stability thatcomes with insurance and workers
compensation.
It might not be as glamorous,that's definitely not the
(04:42):
Silicon Valley dream, right butthere's definitely opportunities
for stability.
Speaker 3 (04:46):
And glamour is in the eyes of a beholder, right, and
I will say that the impact andthe scale of which we're being
able to operate here at Berkeleyis enormous, right, and so
that's what I do.
Enjoy that, and you getting tobuild technology that powers
really a lot of what happens inthe workplace is really awesome,
so I'm thrilled.
Okay, I tell my wife I'm thecoolest guy ever, and she
doesn't believe me yet, butshe'll get there soon.
Speaker 2 (05:08):
So when we talk about technology and claims claims
have been around for a whilewhat do we mean?
What's the role of technologyin claims?
Speaker 3 (05:19):
I think that's a great question.
Again, I'm going to start offwith a buzzword, because
everyone needs a cool buzzwordto begin, but I'm going to say
the technology in claims.
The word of the day isaugmentation, right?
So with this is you know, theclaimant experience is a very
human one, right?
You're in the moment, you're inthe accident.
How are you basically gettingthrough that process?
And that's why, at least howI'm viewing this is how can
(05:40):
technology assist in thatprocess to augment the adjuster
and the humans who are very muchinteracting with that claimant
at that moment and thatexperience, and also with the
policyholders and also withagents.
Again, there's a whole spectrumright here.
But with that is, some peoplejust try to say can we automate
it?
Can you just talk to a computer?
Can you never interact?
And yes, there's going to bethe small paper cuts or the
(06:00):
report.
Only there's going to be somesubset that can be automated.
But the way that I've beenviewing it, at least my time
here at Berkeley, is how can weuse the best of technology to
augment the best of humans toreally attack these problems?
So that's where I believe thatmost places will say just throw
a computer or machine learningmodel at it.
But in reality, how can weleverage technology to leverage
(06:23):
humans is the way that at leastI've been trying to approach it.
Speaker 2 (06:26):
Can you give an example a recent example maybe
of how you're using technologyto augment something in the
claims industry?
Speaker 3 (06:35):
Way too many, matt.
And again, this is what keepsme on my toes, right.
So with that, for instance,sometimes again we will use
machine learning or what somepeople call AI.
So ML is the buzzword, machinelearning which is sometimes
people call it AI, it's allthese different cool terms,
right?
Just saying hey, the Skynet isdoing something in the
background.
(06:55):
So from there we, at least someof our Berkeley time will say
hey, when a new claim comes in,based off the first report of
injury.
How do we get some of theseearly details to identify
whether it's going to be alarge-scale claim or just the
paper cut right.
And again, when it's thesmaller ones, we're saying hey,
an adjuster doesn't need to do17 of the normal steps.
Again, all the regulatory andmust-have requirements we're
(07:17):
always doing to meet the laws ofthe land, but some of the steps
of the process, all thefollow-ups.
We can reduce some of that onthat, when the machine learning
or AI or Skynet identifiessaying hey, you don't need to do
everything on this one, right,and we don't get it 100%, but
directly.
That saves some time.
So we just do the human steps,the augmentation where they need
to plug in and reduce some ofthat early.
(07:38):
So that's a good example wherewe're saying, hey, technology
will identify some things.
Example where we're saying, hey, technology will identify some
things that reduce some of thosesteps, but we still have the
human doing the must-have stepsin that process.
Speaker 1 (07:49):
I think it's really key one of the things you're
highlighting there and thatwe're not in a state where we're
looking for technology toreplace human beings, but if we
could redeploy some of thoseefforts so we have time to do
the things humans are reallygood at.
That's one of the things thatcan make us special if we allow
computers to help us andtechnology to help us improve
the processes.
(08:09):
Sam, what are some of themisconceptions you see when you
start to hear about howtechnology can improve claim
processes?
You're dealing with a lot ofclaims people who probably don't
all have high levels oftechnology background, and so
what are some of thosemisconceptions you sometimes
hear, where there's some, youknow, maybe they don't align
with what's actually, whatactually can be done.
Speaker 3 (08:30):
Well, yeah, I think that's a great question, greg,
and, but I will say it's.
There's two ends of thatspectrum, of those
misconceptions, right.
One is that technology, hey,anything that I don't like doing
right now, technology can solve.
It can solve world hunger justwith a single flitch.
Sam, here's an idea.
Just go make it happen, waveyour magic technology wand and
it's just like my parents, whenI go home for Christmas, think
(08:52):
that I, because I work intechnology, can solve all of
their router problems andprinter problems.
Some people think that, hey,technology can solve anything
you want there, just like theprinters, the TV's broken, can
you fix the Roku it you wantthere, just like the printers,
the TV's broken, can you fix theRoku it's like I can't do
everything, folks.
So that's one end of thespectrum and the other one is
(09:12):
that misconceptions is thattechnology can't do some of the
cool things that humanshistorically could do, and some
of those which, when youactually talk about it, makes
sense Like, hey, we wanttechnology to catch mistakes or
errors, right, oh yeah, ofcourse I would like that.
Or could it reduce mundane,repeatable tasks?
So you're doing like think,work and not just click, work,
right.
(09:32):
But the two ends of thosespectrums are always fun, and
especially insurance.
Some it tends to gravitatetowards oh, technology isn't
going to be able to do anything.
So that's the fun conversation,and I'll actually flip the
question back to you.
How have you all seen, likewhat are some of the
misconceptions you've found, orworking in this industry longer
than I have that you know thatcome up a lot as well.
Is there any different onesbesides from those two extremes?
Speaker 1 (09:56):
All right, Well, I was just going to say the one
I've seen the most in myexperience has been that often
technology can do things, butwhat's not being calculated is
the amount of time and effortand resources that would have to
be deployed to get there.
And so often I feel likeadjusters, managers, supervisors
(10:17):
have the want list, and thewant list might save them five
minutes, but it might take 10years to accomplish it today.
So then the prioritizationbecomes the issue.
It's not always that it can'thappen.
It probably could, but is itreally worth having that happen,
for what it's going to cost?
Speaker 2 (10:36):
Yeah, my example would be the first notice of
loss and automation of that andtriggering, maybe, keywords.
You gave the example ofhighlighting a more serious
claim, so you trigger it onwords like amputation, traumatic
brain injury and those go to amore experienced, high-level
(10:57):
claims person, versus you havesomething where it's like
contusion, laceration, and thosethings get assigned to a newer
person or somebody maybe withless experience.
So I mean, I've seen someexamples of that through
automation and augmentation Iguess you know the word we're
using.
(11:17):
So that would be an exampleI've seen of where it's been
applied in the industry versushaving it go basically across
the spectrum.
A new claim comes in and alwaysgoes to Sandra because Sandra
reviews every new claim.
It's like, well, if we canidentify those things and you
might identify 80 percent of thestuff and then Sandra just gets
(11:41):
20% and then that time thatSandra's freed up is now able to
go and do some other tasks,versus all she's doing is
reviewing new losses coming in.
So that's a recent example I'veseen.
Speaker 3 (11:53):
I love that example, man, because I will say it's
like it's a combination ofrouting rules saying hey, and
we're picking out words.
But this is where technology,like we say, can assist is
sometimes when you have to writethe exact if-then rules and
then you forgot a derivationright.
Some people call things verysimilarly.
There's also this idea intechnology called stemming,
where you know you do plurals orING or things like that, right,
and that's where you can say,hey, technology with natural
(12:15):
language processing, called NLP,come up with all variations of
this type of word or evenrecommend words.
And that's what it's gettingcrazy right now.
It's saying here I'll justrecommend these 20 other words.
Should these route to this typeof adjuster who's going to
handle this type of catastrophicclaim, for instance, or this
cat claim?
So that's where it can use thecombination of an if-then rule
set.
That's kind of basic but it'sgood in technology.
(12:37):
But then can technology figureout a few more examples around
that to help escalate that.
I love that example and again,it's one we're even looking to
utilize here to some degree.
And how can we expand andenhance that as well?
Speaker 2 (12:49):
And it's small things , and then you can continue to
build on them too.
Right, it's nice to have thehome run, but a lot of times
it's the little singles.
And how do you gain littleefficiencies, the 1%?
You know better improvement,and then you know that becomes
exponential improvement overtime, right?
So Sandra now has you know, 80%of her time freed up to do more
(13:09):
meaningful other tasks.
You know, we're not trying toeliminate her role.
We're trying to elevate herrole into like okay, let's, you
know, you're better utilized inthis capacity versus you know
reading.
Yeah, I mean, that's anotherexample.
I mean and the difficulty Iwork with a lot of attorneys and
nurses.
You know the difficulty I get alittle pushback from them is
(13:31):
I'm in the show me state, I'm inSt Louis, I'm in Missouri, they
want to see everything.
So that's some of thedifficulties I experience.
On the kind of the reverse end,like no, let me review the
document, give me the document.
Speaker 3 (13:43):
And one of the fun things about the changes in
technology is sometimes underthe hood.
It is not easy to decouple withmachine learning and AI and
insert buzzwords I mean, I jokeSkynet from the Terminator back
in the day, right, but to thatend is, it can't always show you
that way.
So that's where, sometimes,again, the idea of again
buzzword, back againaugmentation, where you could
say the computer or ML or AIrecommended this.
(14:06):
Do you want to take action onit?
So it's not basically makingthe choice all the way, but it's
allowing those to say hey, oh,I want to double look at the
document, the show me side, Iwant to.
Okay, let me just double checkto make sure the computer got it
right.
And if it does that 100 out of100 times or 99 out of 100,
maybe you start trusting it.
So that's the idea.
Is you don't, like you said, weall want to.
Using a bad sports analogy, weall like seeing the long pass
(14:28):
bomb to Randy Moss at the end,you know, with a touchdown.
I don't know why I'm going tothe Vikings in the 90s, but in
reality the teams that justchunk it Chunk it forward like
five yards at a time.
Five yards at a time.
The small incremental gains caneventually really get you there
, and it's not as glamorous asthe star play, but that's really
what's going to move the needlein the long run.
Speaker 1 (14:48):
Okay, so one of the questions I had, sam, is what is
technology good at in the claimspace?
Speaker 3 (14:54):
Yeah, and that's where I think the unglamorous
side sometimes of technology isthe piping right.
Technology is a goodthoroughfare of information,
connecting different systems,facilitating seamless and easy
communication interactions.
A couple of ideas that come tomind that, especially in the
claims space, especially as weas adjusters need to be more
easily contactable, or basicallyreaching out to claimants.
(15:15):
Things like digital paymentsDon't go say my old school
checked my grandma like mygrandma did.
Can I get something in?
You know?
Venmo?
Right, we've heard that.
Or texting, right, like whichkid under the age of 25 wants to
have a phone call?
They dread it.
That's like is that apunishment, right?
Am I being grounded because Ihave to answer the phone?
So, everything from the pipingto also thinking about where you
(15:37):
can identify patterns, right,technology could say I'm seeing
a lot of this problem.
Let me just say this is aproblem.
What should we do with it?
So we can't always say the whatto do with it.
That's the million-dollarquestion.
For instance, one of ourpartners at a similar company,
midwestern Employers Casualties,have this idea of identifying
problematic claim notes.
Hey, we've seen a bunch ofthese claim notes.
(15:57):
This one is flagged forattention.
Or another operating unit atBerkeley is also saying we've
seen claims like this.
You might want to go checkthose out.
So these are the ways thattechnology can find those
patterns, to then surface it,and then at that point the
adjuster or the human can takeit there.
And so, again, the reticence issometimes that technology is not
(16:19):
as good as humans in everythingit's like.
Well, that's majority clue.
True, we have a brain, we havethe ability to think, but it can
also look at 10 million claimnotes and come up with some
groupings of problems and say,hey, I've seen similar ones that
a human would never be able todo.
And then data validation andanything that we said earlier.
It's not think time, it's justchecking a box, because
(16:41):
computers, they do make errors.
Trust me, I've got a backlogfull of bugs to try to tackle,
but at the same time it's muchmore consistent than humans at
that part.
But trying to ask your computer, or even the newfangled chat
GPT, I tried.
You know that's the buzzwordnowadays.
I tried to ask it to write ahaiku about claims adjustment
and it failed miserably.
So anything that wantsperspective, but it did actually
(17:03):
have a good song to claims.
This is one aside about claims.
I asked it to do a song aboutclaims adjustments to the tune
of Baby Got Back and it did.
Says I like small claims and Icannot lie and it actually went
there.
So it started.
It's getting closer, but thehuge nuance is not.
There is where it needs to be.
Speaker 2 (17:20):
Great points huge nuance you know is not.
There is where it needs to be.
Great points, yeah.
So when you're working with,obviously you know individual
adjusters, you're working withinjured employees, you're
working with differentcomponents, what are some of the
challenges in managing thoseexpectations of the claim staff
when it comes to technology andwhen it comes to, you know,
enhancements and things like youknow Greg alluded to earlier,
(17:41):
like, and you alluded to whenyou're like fix all my stuff.
Like, what are some of the?
What are those expectations?
Like, how do you, how do younormalize?
Speaker 3 (17:51):
those.
That's great, man.
It's like it's the idea of Iwant it all and I want it now
Right.
Like that's the problem thatwe're dealing with right here is
like there's so much that wewant to fix.
Greg, you made a great pointearlier saying, hey, we don't
always necessarily know the size.
This will save me five minutes,but it's going to take Sam six
months to build Right.
So some of the misconceptionsthat I want to when thinking
(18:13):
about this question it's areally good question is that one
?
You hear about it elsewhere.
Well, we can do it right away.
Hey, I've heard about my friendwho works at a similar
insurance company.
They have this cool button thatbrews some coffee and does X, y
, z for my claim process.
Why can't I do that?
So the misconception is peopledon't forget all the piping, all
the underlying architecturethat goes up to make that.
Another is the response times.
(18:35):
We think that everything andevery system is supposed to be
blistering fast, which should be.
Like my iPhone starts up in asecond.
I can't wait a second.
Something pauses, even briefly.
We're like what's happening.
So the challenge is again theseare misconceptions that are out
there, so the challenge istrying to not just address them,
but then overcome them.
And that's where again productmanagement comes in.
(18:56):
My entire profession is to say,hey, we understand that your
iPhone can boot up in a secondand we understand that your
friend can do X, y, z at thisother carrier.
Here's how we're different.
Let's work within our sandboxand let's create amazing things
within there.
So it's really us trying toovercome those by doing the
quick explanation, because youcan't just say no, we can't do
it.
Then you just feel like eitheryou're being talked down to, but
(19:18):
at the same time, our role isin product management and even
within technology is to explainwhat we can do.
And then we partner with againgreat claims groups who are
coming up with amazing ideas tofigure out what we can do.
I'll actually turn thisquestion back to you two and
saying you've seen a lot more ofthese mismanaged expectations
over time and I know, greg, yousaid earlier, there's a
(19:41):
mismanaged expectation of sizing.
What do you think on technology?
Technology teams can do betterto help in that conversation
Like what?
As we, as technologists, whatdo we need to learn when
actually engaging in that backand forth?
Speaker 1 (19:52):
It's a great question .
I think on our end sometimes wedon't understand how hard
certain things are, not maybehow difficult, but how many
hours would be needed toaccomplish something, and so
that wish list we've talkedabout can get really long,
because we can all think ofthings that would make things
easier.
Well, if I just clicked here,then I wouldn't have to do these
(20:13):
five other things.
So I think, sometimesunderstanding also what other
things are being worked on thatare competing for that same time
so we could spend five weeksadding a new button to the
screen, or we could do thisother thing that would allow us
to automate small claims and sounderstanding the impact, I
(20:36):
think on both sides, sometimesmaybe the technology side might
not understand the lift thatwould come from something, and
then that's our job to make surewe're speaking up to explain
that, and then, on the other end, I think, understanding how
long it takes so that we'reprioritizing the right things.
I don't know what your thoughtsare, matt.
Speaker 2 (20:55):
Yeah, it's something similar.
I think spending enough timewith the people doing the job is
a lot of times the difficultyin making sure, like when I say
I want this, like that fivebutton example, or like I have
to click on these eight thingsto have this happen.
Instead of these eight things,like would it make sense to
click on it once, and then itjumps to eight.
(21:15):
What is steps two, three, four,five, six, seven doing?
Are they doing anything like?
Just you know?
Sometimes maybe it's the youknow three, four, five, six,
seven, doing?
Are they doing anything likeand just you know?
Sometimes maybe it's the youknow the data architect or maybe
it's the person that's actually, you know, the business analyst
.
Do they understand?
And then sometimes it claimspeople really don't understand.
Well, the reasons you have toclick on steps three, seven are
(21:38):
this like that triggers this,you know piping triggers this,
you know piping.
Like you know that, like samwas saying, like so I think
there's definitely roles on bothsides.
Like the reason yeah, thephrase that everybody hates is
like well, the reason we do itthat way, sometimes there is a
reason why it's done that way.
Right, yeah, yeah, because anactuarial will come over and hey
(22:00):
, you just cut out this process.
Speaker 3 (22:04):
We love those phone calls.
We love those phone calls andI'll actually add something out
there I think it's great to sayis we, as technologists, I think
need to get better at actuallygoing and sitting down next to
the person saying walk methrough, instead of just saying
I want X, y, z at the eighthstep, like skipping steps two
through seven, let me see whyyou don't need those.
And then again, that's wherethe dialogue can occur right,
hey, step number three.
(22:25):
And I need to give a shout outto my amazing team who thinks
through all these things.
So again, we have a team calledClaim to Fame where we really
focus on claims.
Shout out to that group wherethey'll say, hey, you forgot
about regulatory reporting hereor you forgot about NCCI.
We'll have these differentplugins of different steps.
We may need it, but thatdoesn't mean we can't.
We may need steps now, one andtwo, but we may not need three,
four, five and six.
(22:45):
So I think it's a good back andforth to be able to explain,
but too many times we just hearI want to go jump from one to
eight.
We write it off astechnologists says, can't do it,
and then it dies and you losethat potential savings over time
.
So rolling up the sleeves andgetting in the trenches is
important.
It just takes time, right.
Speaker 2 (23:03):
So how would you so?
Greg and I are both in claimsmanagement and we manage the
claims adjusters and claimsanalysts and examiners, etc.
So we have prioritizations thatare important to us.
You guys obviously haveprioritizations that are
important to IT and totechnology.
(23:23):
What's the best way to managethose prioritizations and the
limited resources?
Because, like the example.
Greg gave earlier.
How do we manage something whenthere's only a limited resource
, it resources and technologyresources?
Unfortunately, you know, wementioned, you know, google
earlier.
Like we don't have Google money, we're not going to throw like
(23:45):
all this at everything.
You know how do we manage thoseprioritizations?
Speaker 3 (23:49):
These are the questions that make us happy
that we're asking thesequestions, right?
Well, there's a simple answer.
Actually, matt and Greg, I willtell you.
The way we prioritize is liquorand guessing Joking okay, I
wish it was that easy.
Or just a dartboard.
It's like close your eyes andthat's no, I completely joke,
but I think a better way of forlack of a letter buzz phrase.
It's basically like it's thevoice of the tactical combined
with the insight of thestrategic right.
(24:11):
So with that is, we need toanytime you ignore sort of the
people who are living this dayin the day-to-day.
On the flip side, if all you'refocusing on is the details and
the weeds, you will miss thebigger strategic opportunities
out there.
So the first thing we need todo in order to prioritize the
chaos that is a huge backlog andseems like 15 new things have
(24:33):
popped up in the last 30 minutessince we've been talking is
combining those two, get thosetwo inputs and then after that,
you need to basically thennarrow it down quickly, because
there's always way too much thatyou want to do, just like my
wife has 15,000 home projectsshe wants me to do.
We can't talk about 15,000projects, we need to talk about
five, right?
So first is, get the inputs, doyour homework.
(24:54):
Again, with targeted homework,you can't boil the ocean, narrow
it down and then pick adirection and go.
I think it's sort of the ideathat a lot of places have
analysis, paralysis.
And again, there's a showcalled Prison Branko, maybe
years ago, where the guyWentworth Miller who played one
of the characters there she saidthere's only four rules you
need to remember Make a plan,execute the plan, expect the
(25:16):
plan to go off the rails, throwaway the plan.
So even with that, things willgo wrong at times.
It's not.
Any plan is not perfect, butyou need to make it and you need
to start executing it Right.
So, take those inputs, lower itdown and then go.
But in reality, if you're notthinking about the strategic,
then you're missing out.
And, greg, I'll actually askyou as basically as a claims
(25:37):
leader, what's important when weon the technology side should
be thinking about the strategic.
Should we be coming to you withthe ideas?
Should you be coming to us, oris there some better forum for
us to make sure we're notmissing that voice of the
strategic?
Speaker 1 (25:50):
Yeah, I feel like that conversation has to go both
ways.
But I do think you're right.
It's really important.
You don't want to have yourtechnology department so focused
on just better, and ourcompetition is going to be
looking for those ways to do itbetter.
We're all trying.
(26:20):
So if all our attention is justtrying to make sure that the
car runs, that's great, butwe're going to miss out on some
really important things.
So I think talking back andforth and creating that dialogue
of what's possible versus whatcan be dreamed up is, to me, I
think, some of the first stepsto really coming up with that.
That's what I would say anyway,matt, thoughts on your end.
(26:40):
Yeah, I like how.
Speaker 2 (26:42):
Sam put it.
I'm a big James Clear fan,where you just have to start
something about you know,ultimately, like you know, it's
great to have the plan, butsometimes you have to start and
finish something.
You can't keep worrying aboutthat being perfect, because then
you'll never get anythingstarted and you'll never finish
(27:04):
anything.
If it's always got to be 100%accurate, there's going to be
little glitches, like you said.
There's going to be little bugs.
Having people understand thatwe're going to do that, and then
Greg hit on it too, like wecan't create something.
That is also so bad, thoughthat it's creating all this
maintenance, right.
(27:25):
It people and technology peoplewant to work on new things,
right.
Unfortunately, like claims, alot of times it's like we're a
legacy.
There's a lot of legacy stuff.
There's years and years.
There's there's legalcomponents in this component and
you know reporting upwardcomponents, so there's all that
that needs to be factored in andsometimes that's you know, you
know the minutiae.
(27:46):
That's not glamorous and youknow nobody wants to work on
that stuff.
You know it's all the fun new.
You know.
Things that everyone wants towork on right.
Yes, very much so.
Speaker 1 (27:56):
I've heard it said that objects in motion tend to
stay in motion.
So I think there is somethingto getting started and just
getting moving.
Sam, I know on our end ourclaims people we're pretty good
at understanding a lot ofmedical terminology.
Now we're also pretty good atlearning a lot of legal because
we spend a lot of time in bothof those worlds.
We're also pretty good atlearning a lot of legal because
we spend a lot of time in bothof those worlds.
But when we start to get in thetechnical side, I think maybe
(28:19):
I'm the only one, but I do thinksometimes we struggle a little
bit.
So what are some of thechallenges or what do you do
from your perspective to breakdown some of these concepts and
help the folks and claimsunderstand what it is you're
trying to accomplish?
Speaker 3 (28:36):
Yeah, I think that's a great question, greg, because,
just as a technologist cominginto insurance, it's sometimes
you're like it's actually goingboth ways, because insurance is
not the easiest and moststraightforward at times, right,
and I had alphabet soup for myfirst, like when I first joined
I was like, oh my gosh, let'splay this game.
Which acronym is this?
Is this the insurance company'sacronym?
Berkeley acronym?
Is this the state acronym?
There's two acronyms for thesame thing.
(28:57):
How do I survive it?
But I think that idea is like,hopefully, things like this
podcast where you're explainingthings like in metaphors or
analogies.
This is like XYZ, or even mybad analogy of my wife giving me
the honey-do list.
How do we break this down intoa shared understanding that we
all have a chuckle as well, buttry to talk a similar language
and that's what productmanagement does, usually trying
(29:18):
to take stakeholders and thendevelopers.
So first of all, talk somethingout and if it makes no sense,
try it again and then sayexplain it like you were
explaining it to a five-year-oldand everyone seems to kind of
like roll their eyes, like no,okay, I can try to explain that,
but in reality it makes you tosimplify it and then get to a
way that you understand it.
Another thing that we do in ourteam a lot is basically diagram,
(29:40):
diagram, diagram big boxes,like you're going to a
whiteboard, and again we use adigital tool called Miro.
There's other versions outthere, but how do we just like
there's a box here and it goesto this box and let me get this
process to this box right?
So that's how sometimes weexplain the intricacies of APIs
and batch jobs and all buzzwordsoup.
(30:01):
When you hear again I'm surewhen people hear our
technologists talk, they're liketheir eyes are rolling.
What are they talking about?
Speaker 1 (30:07):
right.
Speaker 3 (30:08):
And so this is why at least one of our goals is to
explain what we're doing and whyis I say let's make it stupid,
come up with bullet points, notlong paragraphs, and explain to
it in a common language.
And then, if there's follow-upquestions and Neville is like,
well, I actually care about,there's that one person in the
room who's like I want to askthe detailed question, to be the
(30:28):
teacher's pet, right At thatpoint, then you say let's
sidebar this and let's go to thehuge deep dive.
The same idea as, I think, onboth ends.
Also, when adjusters like yousaid earlier, matt is like they
may just say I just wanted to dothis, but need to break it down
that way, I think both sidesneed to say explain it simply
and then have the follow-upquestions after that.
Speaker 1 (30:48):
Matt Levin, it's really easy to forget all of the
things we just assume we know.
And I could say well, we deniedthe TTD because of the IME, but
we need to keep in mind CMS andsomebody who's outside is like
what in the world did he justsay?
And somebody in claimsunderstands every word I just
said, and I do think it goesboth ways.
Speaker 3 (31:09):
I love that, greg, and I think especially because,
again, a lot of developerssometimes are not even having
that interaction.
So, again, if you ever get adeveloper on a call, you'll
rattle that off and they'll justbe completely like.
They're like, I understand noneof it, so I'm not even engaged,
right.
So I think the goal is issaying, just like you know, I've
seen great on this podcastpeople will explain what you
know.
What are these terms?
Let me explain the context.
(31:30):
Even if it's baseline, or mostinsurance adjusters or people
will know.
How do we here's, do we allunderstand what we're talking
about?
Yes, yes, yes or no?
Okay, let's explain 30 seconds.
There we go.
Now let's move on.
And I actually finallyunderstood like it took me a
while, but I actually understoodexactly what you said.
So that's where us, astechnologists, we're getting
there.
Speaker 2 (31:47):
We're getting there.
There's hope, right, creditcards.
So periodically the credit cardsystem would go down.
We still had the whole the old,you know, swiper probably too
(32:15):
young to even have seen any ofthose but greg and I aren't, you
know but no one knew how to useit.
And then there's, occasionallycredit cards don't even have
those raised ridges anymore, sothen you have to write it down
and people are like what do youmean anyway?
What happens when, whentechnology goes down and it's
all hands on deck for you?
Speaker 3 (32:28):
guys, you just, you just panic.
That's what you do.
You just panic, you know, justlike I don't know, no, it's like
then everyone it claims homeexactly, just like it's a
shutter down for the day.
I guess no one else is doing anymore work, right, like those
are.
Those are the good old days oflike snow days or something.
It looks like it's gonna snow.
I guess we just shut it downright, but uh, unfortunately
we're not in middle schoolanymore, right, and I'll have to
(32:48):
ask at the end, matt, how yousurvive those fun.
Probably the, the cha-chingsand people like you know,
customers I'm sure love that andlove taking time right.
What I will say is, likeanything, like any types, you've
got an emergency on claims,whether it's technology, people
forget that the techniques thatyou use, whether in you know
either on the claim side.
When there's a fire, whetherit's at home, when there's a
fire, whether anything, is youjust one, shut out the noise and
(33:09):
focus right.
The problem is with technologyis sometimes it's very hard to
diagnose the problem.
But you'll hear, hey, thesystem's slow or something like
that or something's not workingright.
And then of course everyonetrying to be helpful is like, oh
, and this other thing isn'tworking and this thing.
And then we have to like, okay,there's a lot of noise, how do
we focus it?
You start decoupling thesethings, breaking it down, and
(33:29):
you hear, oh yeah, that thing'sbeen broken for the last two
years.
Thank you for that help.
It's not helping right now inthe panic or the fire.
But I think communication isalso key because when something
breaks, just like you had there,the difference is if you put up
a sign saying, hey, our creditcard machine is broken, it's
going to take us a little longerto process payments, people are
like, oh, okay, they know, theyunderstand, at least I'm going
(33:51):
into it.
You see those sometimes like abig fast food addict.
When I go in, someone puts asign I'm saying we're
understaffed today, please bepatient.
I'm like, oh, okay, this is whythe line's taking so long.
So communication is also a way,during those panic moments, to
try to communicate it there,right, and finally send
consistent messaging and then atthe end give a wrap-up or
(34:13):
summary.
We had technology called abuzzword retrospective, where we
get in a room and again, mostpeople have retrospectives but
we get in a room and say why dideverything go so badly right?
And it's never a funconversation.
You don't want to be havingthese, but you really need to
have the tactical discipline andsay here's everything that went
wrong, summarize it and thensend it out to your stakeholders
for accountability.
(34:33):
I've had to send Greg a coupleof those emails last year and
they're not fun to send, but I'mhoping that it gives the
visibility saying hey, we'retaking ownership for what did
break and here's what we'redoing to solve it in the future.
Right, greg, when you get thoseemails from me, you're not
happy to see them, but I'mhoping you're at least liking
that we're taking ownership thatwe broke something, right?
Speaker 1 (34:55):
Absolutely, and I think some of the things you
know.
We don't have these types ofproblems very often, but when
they have happened, what Ireally appreciate one is just
the regular communication,knowing.
You know, sometimes it's everyhalf hour hour.
This is where we're at, this iswhat we're working on, this is
what we're doing to fix it.
Updates are going to follow,and then at the end that updates
are going to follow, and thenat the end, talking about, this
(35:15):
is what we found out, this iswhat we learned, this is how it
happened and this is what we'regoing to do so it doesn't happen
again.
I think that's a big part of it, and luckily we have a company
culture where we can acceptownership and there's not a lot
of finger pointing.
That allows us to get better.
So kudos to Berkeley fordeveloping that, but that's a
big part of the success.
Speaker 3 (35:33):
This is what I love working about this organization
and why I think we are theleader in our respective
insurance areas is the fact thatwe've got a great product and
again, like you said, it doesn'thappen often, but no company is
perfect.
It's how the company respondsand we take ownership and then
we say how do we not make surethis doesn't happen again?
A lot of places that I'veworked, even past, companies try
to sweep it under the rug,saying companies try to sweep it
(35:53):
under the rug saying hey, therewas no problem, or let's just
cross our fingers and toes andhope it doesn't happen again.
We actually take the methodicaltime to really not just figure
out what went wrong but enhanceso it doesn't happen in the
future, and that's why we'realways pushing forward, which is
appreciated.
Speaker 2 (36:04):
So that's awesome.
I mean, how many problemsbubble down to communication.
Right, when it's like hey, likeyou, you gave the example.
Like you gave the example, likeput up the sign telling people
you know we're moving slowertoday because we're understaffed
, we know this is down.
Hey, we know emails respondingslowly or we're not able to get
outgoing things or ingoingthings or whatever.
(36:25):
So just that communication.
I think you know it's criticaland hopefully this never happens
at our industry.
But that's why I had to give therestaurant analogy, right, yeah
, when those problems go down wehave the after action review,
or whatever you want to callthem, and we're able to improve
the process right.
(36:45):
Very much so.
Speaker 1 (36:46):
So, sam, this is the exciting part, I think, is where
do you see technology going toimprove claims outcomes in the
future?
So we've talked about some ofthe things we're doing now, but
where do you think this goes inthe next five years?
Speaker 3 (36:59):
Yeah, and I think that's a really great question,
greg.
And that's what again exciting.
We don't always just want to befixing the bad things.
We also want to be improvingand really enhancing and making
more fun for everyone involvedthe technologists, the adjusters
, policyholders.
There's a lot that technologycan do to really enhance and so,
for instance, the research firmMcKinsey recently had a report,
and again they call it what wascalled human in the loop.
(37:20):
We've talked a little bit aboutthis earlier, where technology
can take some of the mundaneheavy lifting but then keep the
human apprised of the steps and,if they need to stop or double
click or click in or, as yousaid earlier, find the documents
, read the document, show methey can, but the human is at
least tangential to that process, right?
So they don't have to doeverything, but they're in the
loop.
Another one is again thebuzzword that was really big
(37:42):
years ago, but you still hear ita lot is cloud computing, and
most people don't even know whatthat means or like how does
that impact me?
But the better way to thinkabout it is the Daft Punk song
from years ago harder, better,faster, stronger.
So cloud computing can makesystems faster, more reliable,
have better uptime, just overallbetter experience.
And so that's another shift tosaying hey, let's get away from
(38:03):
the thing that's been in thecloset, the server in the closet
for the last 50 years Don'tunplug.
You see those signs like don'tever unplug this.
More like we don't have to worryabout someone tripping and just
like, oh gosh, the system'sdown right.
Another one is the idea ofbuying versus building.
Some larger organizations likeours historically have been like
we got to build everythingourselves, we got to do
(38:23):
everything ourselves and manageeverything ourselves.
But really the great thingabout Berkeley, while we're
leading sort of this insuranceindustry, is we're thinking
about what are we great at andlet's execute that and let's
augment with other organizationsthat do their best at, and then
we make an ecosystem the bestright, we at Berkeley are
plugging in the piping, but wedon't have to build everything
from scratch, Sort of like,again, you're making your recipe
(38:46):
.
You don't have to do it the waygrandma did, where you make
every single ingredient fromthere.
I can get the best ingredientfrom here, the best ingredient
here, best ingredient from here,and still make that delicious
dessert.
Man, I'm making myself hungryat this point.
And then the last word is wetalked about machine learning
and AI, Again sort of with theidea of human in the loop.
(39:06):
But where can we leverage thosetechnology areas to really
solve problems that we didn'teven think about?
Right, Like one of the excitingthings that we in the
technology side saying go to thebasic question, the job to be
done of what are we trying to dohere and where can technology
assist so we're able to askbigger and bigger questions, it
seems like every year, but thenagain we then have to go back to
the drawing board.
(39:27):
Can it actually do it?
Is it regulations.
But it's exciting to happen tobe an insurance tech, so it gets
me up in the morning.
Speaker 1 (39:35):
I agree, but it's exciting to happen to be an
insurance tech, so, like I, youknow, gets me up in the morning,
I agree.
And when I think back on, whenI started my career, you know,
maybe four or five years in someof the data analytics was
starting to, you know, reallystart to get into insurance and
everything was about filling outa field, right.
There's a new box every fiveminutes.
Here's a new box, here's a newbox right, because we have to
(39:58):
capture the data and so, as afrontline adjuster, it was a
nightmare because it was likewell, now I've got to make sure
I find the 900 boxes and all thedata goes in the right boxes,
right.
But I look at where we are nowand some of the advances in
analytics and being able to justcognitively not only read notes
but start to make inferencesabout what those things mean,
that will allow us to do things,that, instead of having to fill
900 boxes out, we can actuallyget to that same place without
(40:21):
having to do that, which isexciting to me.
To think about where that goesnext, I don't even know what
that looks like 10 years fromnow, but it's very interesting
to see.
Very much so, sam, overall,I've really enjoyed having you
on here.
I can't end the episode withoutjoking.
Our audience will never hear it, but in the middle of this
technology episode, my computerhad a blue screen of death and
(40:44):
we had to stop the entirepodcast because everything
locked up and shut down.
So even in those moments thisstuff can happen, but the show
must go on.
The show must go on.
That's right.
Luckily, we have a great editorthat will make this sound
amazing at the end.
But I wanted to end this seasonby asking each of our guests to
tell us something that they'regrateful for.
(41:05):
I'm really a big proponent ofputting good vibes in the
universe.
I feel like there's plenty ofpeople doing the opposite, so my
small contribution to the worldis to make sure I put something
good in every time I couldthink of it.
So for you today, sam, what'ssomething you're grateful for?
Speaker 3 (41:21):
I love this question, by the way, and again listening
to other amazing adjustedpodcasts.
Shout out please go back to thelibrary, where there's a ton of
great episodes on Spotify oryour streaming platform of
choice.
Great content there.
But I have one professional andone personal.
So the professional thing I'mgrateful for is an amazing
claims product and analyst anddev team.
So again, I get to come on thispodcast and tell everyone about
(41:42):
it.
But we've got a lot of greatpeople who have been working for
decades, who really get all thecredit for supporting Berkeley
Industrial Comp.
And again, our alternativemarkets are BAMTEC.
It's too much of a mouthfuleven for me to say, and on a
personal level, I'll call it thefive Fs, right.
So my faith, my family,football film and finally, fast
food.
Okay, so those five really getme up in the morning and make me
(42:06):
excited.
So I'm grateful for all ofthose things and again, grateful
for this podcast and the bestco-host group product manager
could ask for.
So, thank you all for a greatconversation.
Speaker 1 (42:17):
Thanks, Sam, Really appreciate having you as a guest
today, and with that we willwrap up this episode and hope
you'll follow us in futureepisodes releasing every two
weeks on Monday.
You can also catch our blog onthe off weeks.
So again, remind everybody todo right, think differently and
don't forget to care.
And that's it, guys.
You.
Hello everybody and welcome to Adjusted.
I'm your host, greg Hamlin,coming at you from Sweet Home,
Alabama, and Berkeley IndustrialComp.
And I'm excited to share withyou this rebroadcast.
One of the hot topics in theindustry has been technology and
how technology is transformingthe insurance industry,
(00:34):
especially on the claim side,and in this episode we had the
opportunity to talk to Sam Neer,and Sam has a lot of experience
when it comes to the claim sideand technology, as he manages
quite a bit of this as a projectmanager for us, and so I really
enjoyed this dialogue as wetalked about how we go through
that process, whattransformation looks like, what
(00:58):
opportunities could still beahead of us and what companies
are doing to leverage that onthe claim side.
So I hope you enjoy thisepisode.
As always, I encourage you tolike, share and follow the
podcast, as well as to leave afive-star review so we have an
opportunity to help others findthis podcast and enjoy the
(01:23):
content we've created.
And, as always, I remind you todo right, think differently and
don't forget to care.
Enjoy.
Welcome to Adjusted.
I'm your host, greg Hamlin,coming at you from beautiful
Birmingham, alabama and BerkeleyIndustrial Comp.
And with me is my co-host forthe day, matt Yelling.
Speaker 2 (01:42):
Hello everyone.
This is Matthew Yelling, comingfrom St Louis, Missouri, along
the banks of the mightyMississippi.
Speaker 1 (01:49):
So glad to have Matt with us.
Today.
We actually have a specialguest with us.
That is with Berkeley.
Actually, it goes by theacronym BAM Tech, but Berkeley
Alternative Markets Technology.
Our special guest for the dayis Sam Neer, who is our group
product manager at BerkeleyAlternative Markets Tech.
So, sam, could you say hello toeverybody.
Speaker 3 (02:11):
Hey everybody, Very excited to be here.
Greg and Matt, and again I'mbased out of Raleigh, North
Carolina, where it's a littlebit warmer not as warm as
Birmingham, Alabama right now,but still there.
And I will say, no one wants tosay Berkeley Alternative
Markets Technology over and over, so we might just call it BAM
Tech just for all of our sanity,if that's okay.
Speaker 1 (02:30):
I think that's wise.
I think that's wise.
So the topic today this issomething we've never really
tackled in the nearly threeyears that I've been doing
adjusted is claims andtechnology and the role that it
plays in helping us do our jobs.
So I wanted to bring Sam intoday to talk to us a little bit
about that, since he is part ofthe Berkeley company that
(02:51):
manages our technology to helpus be successful.
Before we dive deep into thetopic, Sam, I wanted to ask you
a little bit about yourself, soI'm sure that you knew you were
going to be in workers'compensation, right?
Speaker 3 (03:05):
Greg, you knew it it's like.
And coming up as a small boy Iwould say to my dad with a glint
in my eye one day, dad, I cando work in workers' compensation
and a quippy line I'll have iseither the way I ended up in the
industry is either I found myway into making great technology
products with greatorganizations and really
affecting change, or I followthe goal of becoming the most
(03:25):
stable, uninteresting version ofmyself.
One of the two, and again maybeit's a little bit of column.
Speaker 1 (03:30):
A, maybe it's a little bit of column B hey there
is a lot of stability inworkers' compensation insurance,
so that's the one positiveright, it's something to be
accounted for.
Speaker 3 (03:41):
So, with the 30-second background again, I've
been working in digitaltechnology for the last 12 years
in product management, Startedoff in PR and analytics, then
moved to telehealth right at theheight of the pandemic Would
not recommend trying to jumpinto telehealth at that time and
I've only been in the insuranceindustry for a couple years now
.
But at the same time I'vereally just had a baptism by
fire but learned a lot and Ithink insurance is a great area
(04:03):
in SureTech that's having a lotof new frontier, a lot of old
legacy systems upgrading.
So it's a great time to bealive.
And again, my specialization isclaims.
I'm beginning to dabble alittle more on the policy and
underwriting side, but thetrends are out there that are
impacting all of us and reallyexcited to talk those through.
Speaker 1 (04:21):
That's awesome.
Well, I know the headlinesright now.
When you hear about technology,you hear about the mass layoffs
of companies like Twitter andFacebook and everybody else
Google.
So there really is something towhat you as you were joking.
There is some stability thatcomes with insurance and workers
compensation.
It might not be as glamorous,that's definitely not the
(04:42):
Silicon Valley dream, right butthere's definitely opportunities
for stability.
Speaker 3 (04:46):
And glamour is in the eyes of a beholder, right, and
I will say that the impact andthe scale of which we're being
able to operate here at Berkeleyis enormous, right, and so
that's what I do.
Enjoy that, and you getting tobuild technology that powers
really a lot of what happens inthe workplace is really awesome,
so I'm thrilled.
Okay, I tell my wife I'm thecoolest guy ever, and she
doesn't believe me yet, butshe'll get there soon.
Speaker 2 (05:08):
So when we talk about technology and claims claims
have been around for a whilewhat do we mean?
What's the role of technologyin claims?
Speaker 3 (05:19):
I think that's a great question.
Again, I'm going to start offwith a buzzword, because
everyone needs a cool buzzwordto begin, but I'm going to say
the technology in claims.
The word of the day isaugmentation, right?
So with this is you know, theclaimant experience is a very
human one, right?
You're in the moment, you're inthe accident.
How are you basically gettingthrough that process?
And that's why, at least howI'm viewing this is how can
(05:40):
technology assist in thatprocess to augment the adjuster
and the humans who are very muchinteracting with that claimant
at that moment and thatexperience, and also with the
policyholders and also withagents.
Again, there's a whole spectrumright here.
But with that is, some peoplejust try to say can we automate
it?
Can you just talk to a computer?
Can you never interact?
And yes, there's going to bethe small paper cuts or the
(06:00):
report.
Only there's going to be somesubset that can be automated.
But the way that I've beenviewing it, at least my time
here at Berkeley, is how can weuse the best of technology to
augment the best of humans toreally attack these problems?
So that's where I believe thatmost places will say just throw
a computer or machine learningmodel at it.
But in reality, how can weleverage technology to leverage
(06:23):
humans is the way that at leastI've been trying to approach it.
Speaker 2 (06:26):
Can you give an example a recent example maybe
of how you're using technologyto augment something in the
claims industry?
Speaker 3 (06:35):
Way too many, matt.
And again, this is what keepsme on my toes, right.
So with that, for instance,sometimes again we will use
machine learning or what somepeople call AI.
So ML is the buzzword, machinelearning which is sometimes
people call it AI, it's allthese different cool terms,
right?
Just saying hey, the Skynet isdoing something in the
background.
(06:55):
So from there we, at least someof our Berkeley time will say
hey, when a new claim comes in,based off the first report of
injury.
How do we get some of theseearly details to identify
whether it's going to be alarge-scale claim or just the
paper cut right.
And again, when it's thesmaller ones, we're saying hey,
an adjuster doesn't need to do17 of the normal steps.
Again, all the regulatory andmust-have requirements we're
(07:17):
always doing to meet the laws ofthe land, but some of the steps
of the process, all thefollow-ups.
We can reduce some of that onthat, when the machine learning
or AI or Skynet identifiessaying hey, you don't need to do
everything on this one, right,and we don't get it 100%, but
directly.
That saves some time.
So we just do the human steps,the augmentation where they need
to plug in and reduce some ofthat early.
(07:38):
So that's a good example wherewe're saying, hey, technology
will identify some things.
Example where we're saying, hey, technology will identify some
things that reduce some of thosesteps, but we still have the
human doing the must-have stepsin that process.
Speaker 1 (07:49):
I think it's really key one of the things you're
highlighting there and thatwe're not in a state where we're
looking for technology toreplace human beings, but if we
could redeploy some of thoseefforts so we have time to do
the things humans are reallygood at.
That's one of the things thatcan make us special if we allow
computers to help us andtechnology to help us improve
the processes.
(08:09):
Sam, what are some of themisconceptions you see when you
start to hear about howtechnology can improve claim
processes?
You're dealing with a lot ofclaims people who probably don't
all have high levels oftechnology background, and so
what are some of thosemisconceptions you sometimes
hear, where there's some, youknow, maybe they don't align
with what's actually, whatactually can be done.
Speaker 3 (08:30):
Well, yeah, I think that's a great question, greg,
and, but I will say it's.
There's two ends of thatspectrum, of those
misconceptions, right.
One is that technology, hey,anything that I don't like doing
right now, technology can solve.
It can solve world hunger justwith a single flitch.
Sam, here's an idea.
Just go make it happen, waveyour magic technology wand and
it's just like my parents, whenI go home for Christmas, think
(08:52):
that I, because I work intechnology, can solve all of
their router problems andprinter problems.
Some people think that, hey,technology can solve anything
you want there, just like theprinters, the TV's broken, can
you fix the Roku it you wantthere, just like the printers,
the TV's broken, can you fix theRoku it's like I can't do
everything, folks.
So that's one end of thespectrum and the other one is
(09:12):
that misconceptions is thattechnology can't do some of the
cool things that humanshistorically could do, and some
of those which, when youactually talk about it, makes
sense Like, hey, we wanttechnology to catch mistakes or
errors, right, oh yeah, ofcourse I would like that.
Or could it reduce mundane,repeatable tasks?
So you're doing like think,work and not just click, work,
right.
(09:32):
But the two ends of thosespectrums are always fun, and
especially insurance.
Some it tends to gravitatetowards oh, technology isn't
going to be able to do anything.
So that's the fun conversation,and I'll actually flip the
question back to you.
How have you all seen, likewhat are some of the
misconceptions you've found, orworking in this industry longer
than I have that you know thatcome up a lot as well.
Is there any different onesbesides from those two extremes?
Speaker 1 (09:56):
All right, Well, I was just going to say the one
I've seen the most in myexperience has been that often
technology can do things, butwhat's not being calculated is
the amount of time and effortand resources that would have to
be deployed to get there.
And so often I feel likeadjusters, managers, supervisors
(10:17):
have the want list, and thewant list might save them five
minutes, but it might take 10years to accomplish it today.
So then the prioritizationbecomes the issue.
It's not always that it can'thappen.
It probably could, but is itreally worth having that happen,
for what it's going to cost?
Speaker 2 (10:36):
Yeah, my example would be the first notice of
loss and automation of that andtriggering, maybe, keywords.
You gave the example ofhighlighting a more serious
claim, so you trigger it onwords like amputation, traumatic
brain injury and those go to amore experienced, high-level
(10:57):
claims person, versus you havesomething where it's like
contusion, laceration, and thosethings get assigned to a newer
person or somebody maybe withless experience.
So I mean, I've seen someexamples of that through
automation and augmentation Iguess you know the word we're
using.
(11:17):
So that would be an exampleI've seen of where it's been
applied in the industry versushaving it go basically across
the spectrum.
A new claim comes in and alwaysgoes to Sandra because Sandra
reviews every new claim.
It's like, well, if we canidentify those things and you
might identify 80 percent of thestuff and then Sandra just gets
(11:41):
20% and then that time thatSandra's freed up is now able to
go and do some other tasks,versus all she's doing is
reviewing new losses coming in.
So that's a recent example I'veseen.
Speaker 3 (11:53):
I love that example, man, because I will say it's
like it's a combination ofrouting rules saying hey, and
we're picking out words.
But this is where technology,like we say, can assist is
sometimes when you have to writethe exact if-then rules and
then you forgot a derivationright.
Some people call things verysimilarly.
There's also this idea intechnology called stemming,
where you know you do plurals orING or things like that, right,
and that's where you can say,hey, technology with natural
(12:15):
language processing, called NLP,come up with all variations of
this type of word or evenrecommend words.
And that's what it's gettingcrazy right now.
It's saying here I'll justrecommend these 20 other words.
Should these route to this typeof adjuster who's going to
handle this type of catastrophicclaim, for instance, or this
cat claim?
So that's where it can use thecombination of an if-then rule
set.
That's kind of basic but it'sgood in technology.
(12:37):
But then can technology figureout a few more examples around
that to help escalate that.
I love that example and again,it's one we're even looking to
utilize here to some degree.
And how can we expand andenhance that as well?
Speaker 2 (12:49):
And it's small things , and then you can continue to
build on them too.
Right, it's nice to have thehome run, but a lot of times
it's the little singles.
And how do you gain littleefficiencies, the 1%?
You know better improvement,and then you know that becomes
exponential improvement overtime, right?
So Sandra now has you know, 80%of her time freed up to do more
(13:09):
meaningful other tasks.
You know, we're not trying toeliminate her role.
We're trying to elevate herrole into like okay, let's, you
know, you're better utilized inthis capacity versus you know
reading.
Yeah, I mean, that's anotherexample.
I mean and the difficulty Iwork with a lot of attorneys and
nurses.
You know the difficulty I get alittle pushback from them is
(13:31):
I'm in the show me state, I'm inSt Louis, I'm in Missouri, they
want to see everything.
So that's some of thedifficulties I experience.
On the kind of the reverse end,like no, let me review the
document, give me the document.
Speaker 3 (13:43):
And one of the fun things about the changes in
technology is sometimes underthe hood.
It is not easy to decouple withmachine learning and AI and
insert buzzwords I mean, I jokeSkynet from the Terminator back
in the day, right, but to thatend is, it can't always show you
that way.
So that's where, sometimes,again, the idea of again
buzzword, back againaugmentation, where you could
say the computer or ML or AIrecommended this.
(14:06):
Do you want to take action onit?
So it's not basically makingthe choice all the way, but it's
allowing those to say hey, oh,I want to double look at the
document, the show me side, Iwant to.
Okay, let me just double checkto make sure the computer got it
right.
And if it does that 100 out of100 times or 99 out of 100,
maybe you start trusting it.
So that's the idea.
Is you don't, like you said, weall want to.
Using a bad sports analogy, weall like seeing the long pass
(14:28):
bomb to Randy Moss at the end,you know, with a touchdown.
I don't know why I'm going tothe Vikings in the 90s, but in
reality the teams that justchunk it Chunk it forward like
five yards at a time.
Five yards at a time.
The small incremental gains caneventually really get you there
, and it's not as glamorous asthe star play, but that's really
what's going to move the needlein the long run.
Speaker 1 (14:48):
Okay, so one of the questions I had, sam, is what is
technology good at in the claimspace?
Speaker 3 (14:54):
Yeah, and that's where I think the unglamorous
side sometimes of technology isthe piping right.
Technology is a goodthoroughfare of information,
connecting different systems,facilitating seamless and easy
communication interactions.
A couple of ideas that come tomind that, especially in the
claims space, especially as weas adjusters need to be more
easily contactable, or basicallyreaching out to claimants.
(15:15):
Things like digital paymentsDon't go say my old school
checked my grandma like mygrandma did.
Can I get something in?
You know?
Venmo?
Right, we've heard that.
Or texting, right, like whichkid under the age of 25 wants to
have a phone call?
They dread it.
That's like is that apunishment, right?
Am I being grounded because Ihave to answer the phone?
So, everything from the pipingto also thinking about where you
(15:37):
can identify patterns, right,technology could say I'm seeing
a lot of this problem.
Let me just say this is aproblem.
What should we do with it?
So we can't always say the whatto do with it.
That's the million-dollarquestion.
For instance, one of ourpartners at a similar company,
midwestern Employers Casualties,have this idea of identifying
problematic claim notes.
Hey, we've seen a bunch ofthese claim notes.
(15:57):
This one is flagged forattention.
Or another operating unit atBerkeley is also saying we've
seen claims like this.
You might want to go checkthose out.
So these are the ways thattechnology can find those
patterns, to then surface it,and then at that point the
adjuster or the human can takeit there.
And so, again, the reticence issometimes that technology is not
(16:19):
as good as humans in everythingit's like.
Well, that's majority clue.
True, we have a brain, we havethe ability to think, but it can
also look at 10 million claimnotes and come up with some
groupings of problems and say,hey, I've seen similar ones that
a human would never be able todo.
And then data validation andanything that we said earlier.
It's not think time, it's justchecking a box, because
(16:41):
computers, they do make errors.
Trust me, I've got a backlogfull of bugs to try to tackle,
but at the same time it's muchmore consistent than humans at
that part.
But trying to ask your computer, or even the newfangled chat
GPT, I tried.
You know that's the buzzwordnowadays.
I tried to ask it to write ahaiku about claims adjustment
and it failed miserably.
So anything that wantsperspective, but it did actually
(17:03):
have a good song to claims.
This is one aside about claims.
I asked it to do a song aboutclaims adjustments to the tune
of Baby Got Back and it did.
Says I like small claims and Icannot lie and it actually went
there.
So it started.
It's getting closer, but thehuge nuance is not.
There is where it needs to be.
Speaker 2 (17:20):
Great points huge nuance you know is not.
There is where it needs to be.
Great points, yeah.
So when you're working with,obviously you know individual
adjusters, you're working withinjured employees, you're
working with differentcomponents, what are some of the
challenges in managing thoseexpectations of the claim staff
when it comes to technology andwhen it comes to, you know,
enhancements and things like youknow Greg alluded to earlier,
(17:41):
like, and you alluded to whenyou're like fix all my stuff.
Like, what are some of the?
What are those expectations?
Like, how do you, how do younormalize?
Speaker 3 (17:51):
those.
That's great, man.
It's like it's the idea of Iwant it all and I want it now
Right.
Like that's the problem thatwe're dealing with right here is
like there's so much that wewant to fix.
Greg, you made a great pointearlier saying, hey, we don't
always necessarily know the size.
This will save me five minutes,but it's going to take Sam six
months to build Right.
So some of the misconceptionsthat I want to when thinking
(18:13):
about this question it's areally good question is that one
?
You hear about it elsewhere.
Well, we can do it right away.
Hey, I've heard about my friendwho works at a similar
insurance company.
They have this cool button thatbrews some coffee and does X, y
, z for my claim process.
Why can't I do that?
So the misconception is peopledon't forget all the piping, all
the underlying architecturethat goes up to make that.
Another is the response times.
(18:35):
We think that everything andevery system is supposed to be
blistering fast, which should be.
Like my iPhone starts up in asecond.
I can't wait a second.
Something pauses, even briefly.
We're like what's happening.
So the challenge is again theseare misconceptions that are out
there, so the challenge istrying to not just address them,
but then overcome them.
And that's where again productmanagement comes in.
(18:56):
My entire profession is to say,hey, we understand that your
iPhone can boot up in a secondand we understand that your
friend can do X, y, z at thisother carrier.
Here's how we're different.
Let's work within our sandboxand let's create amazing things
within there.
So it's really us trying toovercome those by doing the
quick explanation, because youcan't just say no, we can't do
it.
Then you just feel like eitheryou're being talked down to, but
(19:18):
at the same time, our role isin product management and even
within technology is to explainwhat we can do.
And then we partner with againgreat claims groups who are
coming up with amazing ideas tofigure out what we can do.
I'll actually turn thisquestion back to you two and
saying you've seen a lot more ofthese mismanaged expectations
over time and I know, greg, yousaid earlier, there's a
(19:41):
mismanaged expectation of sizing.
What do you think on technology?
Technology teams can do betterto help in that conversation
Like what?
As we, as technologists, whatdo we need to learn when
actually engaging in that backand forth?
Speaker 1 (19:52):
It's a great question .
I think on our end sometimes wedon't understand how hard
certain things are, not maybehow difficult, but how many
hours would be needed toaccomplish something, and so
that wish list we've talkedabout can get really long,
because we can all think ofthings that would make things
easier.
Well, if I just clicked here,then I wouldn't have to do these
(20:13):
five other things.
So I think, sometimesunderstanding also what other
things are being worked on thatare competing for that same time
so we could spend five weeksadding a new button to the
screen, or we could do thisother thing that would allow us
to automate small claims and sounderstanding the impact, I
(20:36):
think on both sides, sometimesmaybe the technology side might
not understand the lift thatwould come from something, and
then that's our job to make surewe're speaking up to explain
that, and then, on the other end, I think, understanding how
long it takes so that we'reprioritizing the right things.
I don't know what your thoughtsare, matt.
Speaker 2 (20:55):
Yeah, it's something similar.
I think spending enough timewith the people doing the job is
a lot of times the difficultyin making sure, like when I say
I want this, like that fivebutton example, or like I have
to click on these eight thingsto have this happen.
Instead of these eight things,like would it make sense to
click on it once, and then itjumps to eight.
(21:15):
What is steps two, three, four,five, six, seven doing?
Are they doing anything like?
Just you know?
Sometimes maybe it's the youknow three, four, five, six,
seven, doing?
Are they doing anything likeand just you know?
Sometimes maybe it's the youknow the data architect or maybe
it's the person that's actually, you know, the business analyst
.
Do they understand?
And then sometimes it claimspeople really don't understand.
Well, the reasons you have toclick on steps three, seven are
(21:38):
this like that triggers this,you know piping triggers this,
you know piping.
Like you know that, like samwas saying, like so I think
there's definitely roles on bothsides.
Like the reason yeah, thephrase that everybody hates is
like well, the reason we do itthat way, sometimes there is a
reason why it's done that way.
Right, yeah, yeah, because anactuarial will come over and hey
(22:00):
, you just cut out this process.
Speaker 3 (22:04):
We love those phone calls.
We love those phone calls andI'll actually add something out
there I think it's great to sayis we, as technologists, I think
need to get better at actuallygoing and sitting down next to
the person saying walk methrough, instead of just saying
I want X, y, z at the eighthstep, like skipping steps two
through seven, let me see whyyou don't need those.
And then again, that's wherethe dialogue can occur right,
hey, step number three.
(22:25):
And I need to give a shout outto my amazing team who thinks
through all these things.
So again, we have a team calledClaim to Fame where we really
focus on claims.
Shout out to that group wherethey'll say, hey, you forgot
about regulatory reporting hereor you forgot about NCCI.
We'll have these differentplugins of different steps.
We may need it, but thatdoesn't mean we can't.
We may need steps now, one andtwo, but we may not need three,
four, five and six.
(22:45):
So I think it's a good back andforth to be able to explain,
but too many times we just hearI want to go jump from one to
eight.
We write it off astechnologists says, can't do it,
and then it dies and you losethat potential savings over time
.
So rolling up the sleeves andgetting in the trenches is
important.
It just takes time, right.
Speaker 2 (23:03):
So how would you so?
Greg and I are both in claimsmanagement and we manage the
claims adjusters and claimsanalysts and examiners, etc.
So we have prioritizations thatare important to us.
You guys obviously haveprioritizations that are
important to IT and totechnology.
(23:23):
What's the best way to managethose prioritizations and the
limited resources?
Because, like the example.
Greg gave earlier.
How do we manage something whenthere's only a limited resource
, it resources and technologyresources?
Unfortunately, you know, wementioned, you know, google
earlier.
Like we don't have Google money, we're not going to throw like
(23:45):
all this at everything.
You know how do we manage thoseprioritizations?
Speaker 3 (23:49):
These are the questions that make us happy
that we're asking thesequestions, right?
Well, there's a simple answer.
Actually, matt and Greg, I willtell you.
The way we prioritize is liquorand guessing Joking okay, I
wish it was that easy.
Or just a dartboard.
It's like close your eyes andthat's no, I completely joke,
but I think a better way of forlack of a letter buzz phrase.
It's basically like it's thevoice of the tactical combined
with the insight of thestrategic right.
(24:11):
So with that is, we need toanytime you ignore sort of the
people who are living this dayin the day-to-day.
On the flip side, if all you'refocusing on is the details and
the weeds, you will miss thebigger strategic opportunities
out there.
So the first thing we need todo in order to prioritize the
chaos that is a huge backlog andseems like 15 new things have
(24:33):
popped up in the last 30 minutessince we've been talking is
combining those two, get thosetwo inputs and then after that,
you need to basically thennarrow it down quickly, because
there's always way too much thatyou want to do, just like my
wife has 15,000 home projectsshe wants me to do.
We can't talk about 15,000projects, we need to talk about
five, right?
So first is, get the inputs, doyour homework.
(24:54):
Again, with targeted homework,you can't boil the ocean, narrow
it down and then pick adirection and go.
I think it's sort of the ideathat a lot of places have
analysis, paralysis.
And again, there's a showcalled Prison Branko, maybe
years ago, where the guyWentworth Miller who played one
of the characters there she saidthere's only four rules you
need to remember Make a plan,execute the plan, expect the
(25:16):
plan to go off the rails, throwaway the plan.
So even with that, things willgo wrong at times.
It's not.
Any plan is not perfect, butyou need to make it and you need
to start executing it Right.
So, take those inputs, lower itdown and then go.
But in reality, if you're notthinking about the strategic,
then you're missing out.
And, greg, I'll actually askyou as basically as a claims
(25:37):
leader, what's important when weon the technology side should
be thinking about the strategic.
Should we be coming to you withthe ideas?
Should you be coming to us, oris there some better forum for
us to make sure we're notmissing that voice of the
strategic?
Speaker 1 (25:50):
Yeah, I feel like that conversation has to go both
ways.
But I do think you're right.
It's really important.
You don't want to have yourtechnology department so focused
on just better, and ourcompetition is going to be
looking for those ways to do itbetter.
We're all trying.
(26:20):
So if all our attention is justtrying to make sure that the
car runs, that's great, butwe're going to miss out on some
really important things.
So I think talking back andforth and creating that dialogue
of what's possible versus whatcan be dreamed up is, to me, I
think, some of the first stepsto really coming up with that.
That's what I would say anyway,matt, thoughts on your end.
(26:40):
Yeah, I like how.
Speaker 2 (26:42):
Sam put it.
I'm a big James Clear fan,where you just have to start
something about you know,ultimately, like you know, it's
great to have the plan, butsometimes you have to start and
finish something.
You can't keep worrying aboutthat being perfect, because then
you'll never get anythingstarted and you'll never finish
(27:04):
anything.
If it's always got to be 100%accurate, there's going to be
little glitches, like you said.
There's going to be little bugs.
Having people understand thatwe're going to do that, and then
Greg hit on it too, like wecan't create something.
That is also so bad, thoughthat it's creating all this
maintenance, right.
(27:25):
It people and technology peoplewant to work on new things,
right.
Unfortunately, like claims, alot of times it's like we're a
legacy.
There's a lot of legacy stuff.
There's years and years.
There's there's legalcomponents in this component and
you know reporting upwardcomponents, so there's all that
that needs to be factored in andsometimes that's you know, you
know the minutiae.
(27:46):
That's not glamorous and youknow nobody wants to work on
that stuff.
You know it's all the fun new.
You know.
Things that everyone wants towork on right.
Yes, very much so.
Speaker 1 (27:56):
I've heard it said that objects in motion tend to
stay in motion.
So I think there is somethingto getting started and just
getting moving.
Sam, I know on our end ourclaims people we're pretty good
at understanding a lot ofmedical terminology.
Now we're also pretty good atlearning a lot of legal because
we spend a lot of time in bothof those worlds.
We're also pretty good atlearning a lot of legal because
we spend a lot of time in bothof those worlds.
But when we start to get in thetechnical side, I think maybe
(28:19):
I'm the only one, but I do thinksometimes we struggle a little
bit.
So what are some of thechallenges or what do you do
from your perspective to breakdown some of these concepts and
help the folks and claimsunderstand what it is you're
trying to accomplish?
Speaker 3 (28:36):
Yeah, I think that's a great question, greg, because,
just as a technologist cominginto insurance, it's sometimes
you're like it's actually goingboth ways, because insurance is
not the easiest and moststraightforward at times, right,
and I had alphabet soup for myfirst, like when I first joined
I was like, oh my gosh, let'splay this game.
Which acronym is this?
Is this the insurance company'sacronym?
Berkeley acronym?
Is this the state acronym?
There's two acronyms for thesame thing.
(28:57):
How do I survive it?
But I think that idea is like,hopefully, things like this
podcast where you're explainingthings like in metaphors or
analogies.
This is like XYZ, or even mybad analogy of my wife giving me
the honey-do list.
How do we break this down intoa shared understanding that we
all have a chuckle as well, buttry to talk a similar language
and that's what productmanagement does, usually trying
(29:18):
to take stakeholders and thendevelopers.
So first of all, talk somethingout and if it makes no sense,
try it again and then sayexplain it like you were
explaining it to a five-year-oldand everyone seems to kind of
like roll their eyes, like no,okay, I can try to explain that,
but in reality it makes you tosimplify it and then get to a
way that you understand it.
Another thing that we do in ourteam a lot is basically diagram,
(29:40):
diagram, diagram big boxes,like you're going to a
whiteboard, and again we use adigital tool called Miro.
There's other versions outthere, but how do we just like
there's a box here and it goesto this box and let me get this
process to this box right?
So that's how sometimes weexplain the intricacies of APIs
and batch jobs and all buzzwordsoup.
(30:01):
When you hear again I'm surewhen people hear our
technologists talk, they're liketheir eyes are rolling.
What are they talking about?
Speaker 1 (30:07):
right.
Speaker 3 (30:08):
And so this is why at least one of our goals is to
explain what we're doing and whyis I say let's make it stupid,
come up with bullet points, notlong paragraphs, and explain to
it in a common language.
And then, if there's follow-upquestions and Neville is like,
well, I actually care about,there's that one person in the
room who's like I want to askthe detailed question, to be the
(30:28):
teacher's pet, right At thatpoint, then you say let's
sidebar this and let's go to thehuge deep dive.
The same idea as, I think, onboth ends.
Also, when adjusters like yousaid earlier, matt is like they
may just say I just wanted to dothis, but need to break it down
that way, I think both sidesneed to say explain it simply
and then have the follow-upquestions after that.
Speaker 1 (30:48):
Matt Levin, it's really easy to forget all of the
things we just assume we know.
And I could say well, we deniedthe TTD because of the IME, but
we need to keep in mind CMS andsomebody who's outside is like
what in the world did he justsay?
And somebody in claimsunderstands every word I just
said, and I do think it goesboth ways.
Speaker 3 (31:09):
I love that, greg, and I think especially because,
again, a lot of developerssometimes are not even having
that interaction.
So, again, if you ever get adeveloper on a call, you'll
rattle that off and they'll justbe completely like.
They're like, I understand noneof it, so I'm not even engaged,
right.
So I think the goal is issaying, just like you know, I've
seen great on this podcastpeople will explain what you
know.
What are these terms?
Let me explain the context.
(31:30):
Even if it's baseline, or mostinsurance adjusters or people
will know.
How do we here's, do we allunderstand what we're talking
about?
Yes, yes, yes or no?
Okay, let's explain 30 seconds.
There we go.
Now let's move on.
And I actually finallyunderstood like it took me a
while, but I actually understoodexactly what you said.
So that's where us, astechnologists, we're getting
there.
Speaker 2 (31:47):
We're getting there.
There's hope, right, creditcards.
So periodically the credit cardsystem would go down.
We still had the whole the old,you know, swiper probably too
(32:15):
young to even have seen any ofthose but greg and I aren't, you
know but no one knew how to useit.
And then there's, occasionallycredit cards don't even have
those raised ridges anymore, sothen you have to write it down
and people are like what do youmean anyway?
What happens when, whentechnology goes down and it's
all hands on deck for you?
Speaker 3 (32:28):
guys, you just, you just panic.
That's what you do.
You just panic, you know, justlike I don't know, no, it's like
then everyone it claims homeexactly, just like it's a
shutter down for the day.
I guess no one else is doing anymore work, right, like those
are.
Those are the good old days oflike snow days or something.
It looks like it's gonna snow.
I guess we just shut it downright, but uh, unfortunately
we're not in middle schoolanymore, right, and I'll have to
(32:48):
ask at the end, matt, how yousurvive those fun.
Probably the, the cha-chingsand people like you know,
customers I'm sure love that andlove taking time right.
What I will say is, likeanything, like any types, you've
got an emergency on claims,whether it's technology, people
forget that the techniques thatyou use, whether in you know
either on the claim side.
When there's a fire, whetherit's at home, when there's a
fire, whether anything, is youjust one, shut out the noise and
(33:09):
focus right.
The problem is with technologyis sometimes it's very hard to
diagnose the problem.
But you'll hear, hey, thesystem's slow or something like
that or something's not workingright.
And then of course everyonetrying to be helpful is like, oh
, and this other thing isn'tworking and this thing.
And then we have to like, okay,there's a lot of noise, how do
we focus it?
You start decoupling thesethings, breaking it down, and
(33:29):
you hear, oh yeah, that thing'sbeen broken for the last two
years.
Thank you for that help.
It's not helping right now inthe panic or the fire.
But I think communication isalso key because when something
breaks, just like you had there,the difference is if you put up
a sign saying, hey, our creditcard machine is broken, it's
going to take us a little longerto process payments, people are
like, oh, okay, they know, theyunderstand, at least I'm going
(33:51):
into it.
You see those sometimes like abig fast food addict.
When I go in, someone puts asign I'm saying we're
understaffed today, please bepatient.
I'm like, oh, okay, this is whythe line's taking so long.
So communication is also a way,during those panic moments, to
try to communicate it there,right, and finally send
consistent messaging and then atthe end give a wrap-up or
(34:13):
summary.
We had technology called abuzzword retrospective, where we
get in a room and again, mostpeople have retrospectives but
we get in a room and say why dideverything go so badly right?
And it's never a funconversation.
You don't want to be havingthese, but you really need to
have the tactical discipline andsay here's everything that went
wrong, summarize it and thensend it out to your stakeholders
for accountability.
(34:33):
I've had to send Greg a coupleof those emails last year and
they're not fun to send, but I'mhoping that it gives the
visibility saying hey, we'retaking ownership for what did
break and here's what we'redoing to solve it in the future.
Right, greg, when you get thoseemails from me, you're not
happy to see them, but I'mhoping you're at least liking
that we're taking ownership thatwe broke something, right?
Speaker 1 (34:55):
Absolutely, and I think some of the things you
know.
We don't have these types ofproblems very often, but when
they have happened, what Ireally appreciate one is just
the regular communication,knowing.
You know, sometimes it's everyhalf hour hour.
This is where we're at, this iswhat we're working on, this is
what we're doing to fix it.
Updates are going to follow,and then at the end that updates
are going to follow, and thenat the end, talking about, this
(35:15):
is what we found out, this iswhat we learned, this is how it
happened and this is what we'regoing to do so it doesn't happen
again.
I think that's a big part of it, and luckily we have a company
culture where we can acceptownership and there's not a lot
of finger pointing.
That allows us to get better.
So kudos to Berkeley fordeveloping that, but that's a
big part of the success.
Speaker 3 (35:33):
This is what I love working about this organization
and why I think we are theleader in our respective
insurance areas is the fact thatwe've got a great product and
again, like you said, it doesn'thappen often, but no company is
perfect.
It's how the company respondsand we take ownership and then
we say how do we not make surethis doesn't happen again?
A lot of places that I'veworked, even past, companies try
to sweep it under the rug,saying companies try to sweep it
(35:53):
under the rug saying hey, therewas no problem, or let's just
cross our fingers and toes andhope it doesn't happen again.
We actually take the methodicaltime to really not just figure
out what went wrong but enhanceso it doesn't happen in the
future, and that's why we'realways pushing forward, which is
appreciated.
Speaker 2 (36:04):
So that's awesome.
I mean, how many problemsbubble down to communication.
Right, when it's like hey, likeyou, you gave the example.
Like you gave the example, likeput up the sign telling people
you know we're moving slowertoday because we're understaffed
, we know this is down.
Hey, we know emails respondingslowly or we're not able to get
outgoing things or ingoingthings or whatever.
(36:25):
So just that communication.
I think you know it's criticaland hopefully this never happens
at our industry.
But that's why I had to give therestaurant analogy, right, yeah
, when those problems go down wehave the after action review,
or whatever you want to callthem, and we're able to improve
the process right.
(36:45):
Very much so.
Speaker 1 (36:46):
So, sam, this is the exciting part, I think, is where
do you see technology going toimprove claims outcomes in the
future?
So we've talked about some ofthe things we're doing now, but
where do you think this goes inthe next five years?
Speaker 3 (36:59):
Yeah, and I think that's a really great question,
greg.
And that's what again exciting.
We don't always just want to befixing the bad things.
We also want to be improvingand really enhancing and making
more fun for everyone involvedthe technologists, the adjusters
, policyholders.
There's a lot that technologycan do to really enhance and so,
for instance, the research firmMcKinsey recently had a report,
and again they call it what wascalled human in the loop.
(37:20):
We've talked a little bit aboutthis earlier, where technology
can take some of the mundaneheavy lifting but then keep the
human apprised of the steps and,if they need to stop or double
click or click in or, as yousaid earlier, find the documents
, read the document, show methey can, but the human is at
least tangential to that process, right?
So they don't have to doeverything, but they're in the
loop.
Another one is again thebuzzword that was really big
(37:42):
years ago, but you still hear ita lot is cloud computing, and
most people don't even know whatthat means or like how does
that impact me?
But the better way to thinkabout it is the Daft Punk song
from years ago harder, better,faster, stronger.
So cloud computing can makesystems faster, more reliable,
have better uptime, just overallbetter experience.
And so that's another shift tosaying hey, let's get away from
(38:03):
the thing that's been in thecloset, the server in the closet
for the last 50 years Don'tunplug.
You see those signs like don'tever unplug this.
More like we don't have to worryabout someone tripping and just
like, oh gosh, the system'sdown right.
Another one is the idea ofbuying versus building.
Some larger organizations likeours historically have been like
we got to build everythingourselves, we got to do
(38:23):
everything ourselves and manageeverything ourselves.
But really the great thingabout Berkeley, while we're
leading sort of this insuranceindustry, is we're thinking
about what are we great at andlet's execute that and let's
augment with other organizationsthat do their best at, and then
we make an ecosystem the bestright, we at Berkeley are
plugging in the piping, but wedon't have to build everything
from scratch, Sort of like,again, you're making your recipe
(38:46):
.
You don't have to do it the waygrandma did, where you make
every single ingredient fromthere.
I can get the best ingredientfrom here, the best ingredient
here, best ingredient from here,and still make that delicious
dessert.
Man, I'm making myself hungryat this point.
And then the last word is wetalked about machine learning
and AI, Again sort of with theidea of human in the loop.
(39:06):
But where can we leverage thosetechnology areas to really
solve problems that we didn'teven think about?
Right, Like one of the excitingthings that we in the
technology side saying go to thebasic question, the job to be
done of what are we trying to dohere and where can technology
assist so we're able to askbigger and bigger questions, it
seems like every year, but thenagain we then have to go back to
the drawing board.
(39:27):
Can it actually do it?
Is it regulations.
But it's exciting to happen tobe an insurance tech, so it gets
me up in the morning.
Speaker 1 (39:35):
I agree, but it's exciting to happen to be an
insurance tech, so, like I, youknow, gets me up in the morning,
I agree.
And when I think back on, whenI started my career, you know,
maybe four or five years in someof the data analytics was
starting to, you know, reallystart to get into insurance and
everything was about filling outa field, right.
There's a new box every fiveminutes.
Here's a new box, here's a newbox right, because we have to
(39:58):
capture the data and so, as afrontline adjuster, it was a
nightmare because it was likewell, now I've got to make sure
I find the 900 boxes and all thedata goes in the right boxes,
right.
But I look at where we are nowand some of the advances in
analytics and being able to justcognitively not only read notes
but start to make inferencesabout what those things mean,
that will allow us to do things,that, instead of having to fill
900 boxes out, we can actuallyget to that same place without
(40:21):
having to do that, which isexciting to me.
To think about where that goesnext, I don't even know what
that looks like 10 years fromnow, but it's very interesting
to see.
Very much so, sam, overall,I've really enjoyed having you
on here.
I can't end the episode withoutjoking.
Our audience will never hear it, but in the middle of this
technology episode, my computerhad a blue screen of death and
(40:44):
we had to stop the entirepodcast because everything
locked up and shut down.
So even in those moments thisstuff can happen, but the show
must go on.
The show must go on.
That's right.
Luckily, we have a great editorthat will make this sound
amazing at the end.
But I wanted to end this seasonby asking each of our guests to
tell us something that they'regrateful for.
(41:05):
I'm really a big proponent ofputting good vibes in the
universe.
I feel like there's plenty ofpeople doing the opposite, so my
small contribution to the worldis to make sure I put something
good in every time I couldthink of it.
So for you today, sam, what'ssomething you're grateful for?
Speaker 3 (41:21):
I love this question, by the way, and again listening
to other amazing adjustedpodcasts.
Shout out please go back to thelibrary, where there's a ton of
great episodes on Spotify oryour streaming platform of
choice.
Great content there.
But I have one professional andone personal.
So the professional thing I'mgrateful for is an amazing
claims product and analyst anddev team.
So again, I get to come on thispodcast and tell everyone about
(41:42):
it.
But we've got a lot of greatpeople who have been working for
decades, who really get all thecredit for supporting Berkeley
Industrial Comp.
And again, our alternativemarkets are BAMTEC.
It's too much of a mouthfuleven for me to say, and on a
personal level, I'll call it thefive Fs, right.
So my faith, my family,football film and finally, fast
food.
Okay, so those five really getme up in the morning and make me
(42:06):
excited.
So I'm grateful for all ofthose things and again, grateful
for this podcast and the bestco-host group product manager
could ask for.
So, thank you all for a greatconversation.
Speaker 1 (42:17):
Thanks, Sam, Really appreciate having you as a guest
today, and with that we willwrap up this episode and hope
you'll follow us in futureepisodes releasing every two
weeks on Monday.
You can also catch our blog onthe off weeks.
So again, remind everybody todo right, think differently and
don't forget to care.
And that's it, guys.
You.
Popular Podcasts
Bookmarked by Reese's Book Club
Welcome to Bookmarked by Reese’s Book Club — the podcast where great stories, bold women, and irresistible conversations collide! Hosted by award-winning journalist Danielle Robay, each week new episodes balance thoughtful literary insight with the fervor of buzzy book trends, pop culture and more. Bookmarked brings together celebrities, tastemakers, influencers and authors from Reese's Book Club and beyond to share stories that transcend the page. Pull up a chair. You’re not just listening — you’re part of the conversation.
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com