December 13, 2023 • 43 mins
In this Episode 10 of Insurtech Unscripted, Ali Safavi talks about innovations in the market. Did you know that the insurance industry saw its first technology wave back in the 1980s? Today's guest, Amir Kabir, general partner at AV8 and a well-known investor in the InsureTech space, is here to take us on a truly enlightening journey through the evolution of InsureTech. We'll start from the birth of big insurance companies in the early 1900s, moving into the tech infusion of the 80s and 90s, and finally arriving in the present day with a focus on the rise of venture capital funding and the birth of "InsureTech".
We'll also venture into the world of the second wave of InsureTech - highlighting the critical issue that startups in the field are trying to solve: a lack of customer-centricity in the insurance industry. We will discuss intriguing opportunities in niche insurance categories like collectibles insurance and the power of B2B2C distribution channels. Get ready to navigate through the challenge of identifying large and back able markets for digital innovation in the industry.
Finally, we'll ponder on the potential for monumental success in the InsureTech, discussing the prospects of a $50 billion exit and how it stands when compared to the market value of established insurance companies. We'll also touch on the role of tech in enhancing underwriting and profitability in niche insurance markets, and the potential of startups to become risk-bearing entities. Join us as we unravel these fascinating concepts and journey through the transformative world of InsureTech.
COVU is founded and advised by a team of insurance, finance, and tech industry veterans with a passion for transparent and unbiased advice. COVU's mission is to help both everyday people and professional advisors manage risk and insurance smarter. We help our customers find and fill their coverage gaps and achieve true peace of mind.
Sign up: https://covu.com
Let’s stay connected!
⚡ Instagram: / covu_inc
👍 Facebook: ...
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Ali Safavi (00:19):
Hello everyone.
My name is Ali Safavi.
I'm the CEO and co-founder ofCOVU and your host for
InsureTech Unscripted.
Today I'm honored to have oneof my very good friends.
I'm here with Amir Kabir, ageneral partner at AV8.
Obviously, I've known Amir for,I think, almost the beginning
of the InsureTech wave, at leastthe one that really picked up,
and we've been friends since andhe's obviously one of the most
(00:42):
knowledgeable investors that Iknow in the insurance space and
he has a lot of thoughts andthesis on where he thinks the
space is going.
So we're excited to have you onboard on the podcast, amir.
Maybe you could tell us alittle bit more about yourself
before we dive deeper.
Amir Kabir (00:57):
Yeah, sure, man, thanks for having me excited to
be part of this and talking toyou.
Yeah, we have, I don't know.
We've known each other formaybe 10 years now, I'm not sure
, but maybe a little bit less.
But yeah, as you said, sincekind of the whole hype cycle of
insurance, of InsureTech,started.
But yeah, a little bit aboutmyself.
I've been investing in theInsureTech space for a little
(01:20):
bit over 10 years and I kind ofgot into the FinTech InsureTech
world.
Coincidentally, I don't have aninsurance background.
I come from the tech side.
I was at operating roles forenterprise startups back in
Europe and when I came to the USI worked for a venture fund
(01:41):
actually with 66 ventures on theEast Coast for a while and got
into very early stage FinTechand InsureTech deals and did my
first InsureTech deal, I thinkin 2013 or so, which was simple
insurance, and that's how I gotinto venture.
From there I made my way to theWest Coast and helped build up
(02:02):
Munich Reventures, which I thinkis very familiar to a lot of
people in the insurance worldwho are kind of like the first
kind of corporate venture VCthat dives into insurance pretty
heavily and invested in allsorts of InsureTech version,
ones that people know like nextinsurance, Insure HDVI and so on
(02:23):
and so forth, and I joinedAviAventures a little bit over
two years ago as the fourthpartner to kind of lead the
FinTech practice and haveinvested so far, I think, in
seven or eight companies in theInsureTech world, which is
basically built upon the thesisthat I have written about the
next day of InsureTech and whatI've seen over the last 10 years
(02:45):
.
Ali Safavi (02:46):
Very cool.
So I usually like to start myquestioning with one of the most
controversial ideas, but justgiven the fact that you already
have put your thoughts on paperso clearly, it makes my job a
lot easier.
So let's start with.
I mean, I think the way thatyou looked at InsureTech is
interesting.
You've looked at it in wavesand you tried to compare it to
(03:06):
before.
So maybe you could starttalking a little bit about the
history of InsureTech, changesin insurance as an industry and
as relevant today and as we kindof try to think about the
future, how the history kind ofcomes into play.
Amir Kabir (03:19):
Yeah, that's a good point.
So I started writing aboutinsurance I think over a little
bit over two years ago and whatI tried to do was, in the part
one, was to kind of trace backthe history of insurance and the
evolution of InsureTech and seewhat happened actually the last
100 years, specifically in theUS when it comes down to
(03:40):
insurance and personal insuranceand so on and so forth.
So when you look at the firstwrite up that I did and the kind
of graph that I created, youcan see the big behemoth of
today.
The incumbents like the StateFarm and Geico and Progressive
(04:01):
kind of started in the veryearly 1900s, right 1920, 30
years around, and then the firstkind of technological wave in
insurance kind of came aboutwith the fact that it was within
like the 1980s, right I think1983 applied systems kind of
(04:21):
started 1985, insurity, selectquote 1997, insurion and so on
and so forth.
So these were kind of in 2000like guide buyers, so that the
80s, 90s kind of defined, I feel, the first kind of wave of
technology in insurance andgoing forward.
(04:44):
When you look at the graph andwhat happened is that within the
2000s a lot of like the firstversion of InsureTech, I would
say, kind of emerged which werebasically focused on price
comparison websites and somesort of distribution models
within insurance, like veryrudimentary distribution models,
(05:07):
and then in 2010 and onwards tolike 2020, that's where I feel
like the kind of term InsureTechwork was kind of minted let's
say, right, they're going intothe crypto world minted.
The InsureTech work has kind ofminted because people were
basically comparing likeinsurance as part of the
(05:32):
financial services ecosystem toFinTech and kind of counter word
here was InsureTech and I thinkthat emergence of that word,
insuretech, coupled with likeventure capital money flowing
into the ecosystem, started thefirst wave of InsureTech.
And within 2012, 13, 14, 15,obviously, the wave of
(05:56):
InsureTech took up and all ofthe prominent InsureTech
companies that people know about, as I mentioned before next
insurance lemonade, root, insure, oscar, health on the health
insurance side, ladder, life andso on and so forth.
(06:16):
They started around that timeand we had some kind of
interesting exit scenarios in2020, 21 where people the first
kind of the wave of insure taxwere able to create some sort of
a liquidity for early investorsand however quickly we saw that
(06:44):
like the public marketsdismissed somehow what insure
tax had built and the underlyingbusiness economics, they were
not really supporting the way togo public.
And so right now I feel likewe're entering insure tax, wave
(07:05):
two, and that's like based onthe article that I wrote
subsequently, where it's aboutlike what have you learned over
the last, let's say, maybe 10years when it comes down to
really insure tax, and how canwe move forward and what are the
learnings from there that wecan apply to the next day of
insure tax, if that makes sense.
Ali Safavi (07:28):
So let's jump into it.
So, basically, you're sayingthat this is the learnings from
insure tax 1.0, then, based onthat, insure tax 2.0 is created.
So, on a high level, what arethe big learnings that you feel
like a lot of investors orentrepreneurs have learned, or
haven't learned since or havemissed that you think worth
highlighting?
And then how's that insure tax?
Amir Kabir (07:48):
2.0?
.
I think the biggest kind oflearnings or miss was like hey,
technology will transforminsurance and everything will be
better, faster, cheaper andwhatever you wanna call it.
So obviously tech is part oflike insure tech.
The same way tech is part ofFinTech right, but in both
(08:12):
instances I feel, specificallybecause you are operating in a
very regulatory environment, theinsurance and the financial
services and FinTech have tocome first and I think this is
the biggest learning here thattechnology will not help you win
with an insurance.
No matter how great yourtechnology is, that will not
(08:34):
help you win and make big moneyfor you, and I think that was
the biggest learning here.
However, I think insure takewave one and again going back to
like why technology will nothelp you win, it's that because
insurance is not really a topline business.
(08:54):
So comparing insure tech tolike a SaaS company or like an
enterprise software company oreven like a consumer company in
the commerce world is completelyill equipped here, because it's
not a top line industry,meaning the more premium you get
in it doesn't mean you'rebetter or doing the better thing
.
It's more actually a bottomline industry where the
(09:17):
fundamental is unit economicsafter a line and the last ratios
play a very important role.
So I think this is the biggestlearning from my perspective
that I think people haverealized.
And then, on top of that, Ithink what people have seen as
well and goes back to thelearnings, is that insure take
(09:38):
wave one really focused on, likethe large, saturated markets,
highly commoditized productslike auto insurance, home
insurance, rents insurance.
When you look at those products, most of them, or some of them,
are mandatory.
Others are not.
But most of the people withinthat just focus in the United
(10:02):
States here have already autoinsurance.
Right, they have home insuranceinsurance.
So it's not that it's anunderserved market.
So what are you competing hereabout?
You're competing mostly aboutprice and maybe how do you
service customers.
So that was kind of the firstwave of insure take.
When you look at them, theywere able to quickly acquire
(10:25):
customers and show like a topline that people got excited
about.
However, the underlying, thebottom line, was not really
great here and over time it wasshowing and, honestly, if you
look at the public markets nowit still shows.
Right, those insure takes thatwent public and the numbers are
(10:45):
publicly available.
You can look at them.
It's not that they're making,they're the profitable
businesses.
So I think that was the biggestkind of thing here to focus on
like large, saturated marketswhere you basically are just
competing about price, right,because at the end of the day, I
(11:08):
have auto insurance fromprogressive and I pay, I think,
$140 a month.
If I get it for $50 a month,yeah, I take it right, but it
doesn't necessarily mean thatyou're covering your losses
there and it doesn't necessarilymean that you're getting the
right risk profile from acustomer perspective.
Right, and from insuranceperspective.
(11:29):
Those, I think also thelearnings are, those who switch
quickly or like look forinsurance, are actually the ones
you don't want, right.
And so the other thing is alsolike there was limited
differentiation, I feel, interms of like what, how, in
terms of underwriting, meaningwhat incumbents would do and
(11:50):
what insurtex would do was notmuch difference.
And it goes back to again it'sa very regulated industry.
So meaning like if you do autoinsurance and I've seen a bunch
of places too they would pitchme like, hey, we don't take FICO
into account, we take likeother stuff into account to
underwrite you.
Well, I'm like, well, that'sgreat, but you can't really do
(12:11):
that depending on which stateyou're in and the regulatory
bodies you have to comply withand you have to like, apply for,
like new products and so on andso forth.
So it's a very tough endeavorif you wanna really come up with
a new product.
So from the underwritingperspective there was also not
much differentiation.
Some would do more, some woulddo less, so it was not really
(12:33):
that much.
Going back to, like, theproducts that is insurtex
focused on were very like quoteunquote, easier products to
underwrite.
So and obviously thedistribution.
I think the learning there isthat customer acquisition is
(12:53):
very expensive and customerlifetime value, depending on the
product, is kind of limited.
So that means you constantlyhave to get new customers and I
said before, if you have toconstantly get new customers you
most likely will not get thebest risk out there.
So I stop here because I thinkthese are like the fundamental
(13:17):
things from the learningsperspective and what insurtex
Wave 1 kind of showed forventure investors and the
ecosystem.
However, I think it's veryimportant to outline that
insurtex Wave 1, the companiesreally navigated like incumbents
(13:37):
and service providers,consumers, regulators and the
broader financial market andshowing that, hey,
digitalization of insurance isinevitable and that's, I think,
a very important statement tomake, because all the work that
went into insurtex Wave 1 wasvery important, although maybe
(13:58):
the outcome of some of them, ormaybe most of them, was not
really great.
However, I think there was alot of learnings that can be
applied going forward.
Ali Safavi (14:05):
So let me ask you this like on paper and theory
and again I just want to makesure I kind of get the
hypothesis is that paperregulation changes and makes
Like is less involved with howunwriting is done for a product
like auto insurance, are yousaying, then we'll have truly
innovative startups in autoinsurance?
Amir Kabir (14:24):
Like Regulation is the main blocker for, you know,
like a true startup when in aspace like auto insurance, I
think so honestly because, again, you know the way you can go to
market is that you take anexisting product right, which a
lot of insure tech version oncedid, and you know so-called an
(14:44):
off the shelf product, and youdigitize that and, you know,
distribute it through variouschannels.
And then others you know took anoff the shelf product and came
up with new kind of innovativeunderwriting ideas around that
and had to first apply to getthat approved from the
regulatory body said to be ableto sell it to the consumer right
(15:07):
and others.
You know the other way to goabout it is also to use, you
know, ens paper and which is notso, which is not so
scrutinizing on a regulatoryside.
But I think that's that's.
I mean it's important thatregulatory barriers are there
right, otherwise everybody wouldgo around and, you know, do
(15:29):
whatever they want in terms of,like, writing insurance.
So I feel like there's no wayto get around it.
You have to work with those,you know, regulatory entities.
But the other learning here toois that you know it takes a
long time to get to market.
You know some insure tech tooksix to nine or 12 months to get
(15:50):
a product approved or a newproduct approved or a modified
product approved, and then theycould go to market right.
So the time to market is verylong and you need to have a long
breath if you want to win here.
Ali Safavi (16:03):
Very interesting.
Okay, and and obviously let'stalk a little bit about insure
tech 2.0 before we drill down insome of the areas that I was
curious to ask more about.
But now I'm talking about allthe learnings from insure tech
2.0.
Sorry, 1.0.
Let's introduce insure tech 2.0.
Amir Kabir (16:18):
So again, this is my kind of you know two cents here
, and the stuff that I wasthinking about over the last
years that I've seen is that, interms of like venture, I always
try to, you know, ask likewhat's the problem here, right,
you know that's?
That's, I think, a fundamentalquestion.
When you think about ventureand venture investing and
(16:40):
investing in startups is likewhat is the fundamental problem
that the startup is trying tosolve?
And insurance, you know, goingback to like version one, when
you think about it, you knowwhat was the fundamental problem
?
And from my perspective, thefundamental problem was that a
lot of people, or, like you know, the broader community or the
(17:02):
consumer, was thinking that youknow, there is not much customer
centricity in insurance, right,and people were like, hey, you
know, insurance companies reallydon't care about their
consumers.
That's why the NPS scores is solow and so on and so forth.
I mean, when you think aboutlike that insurance companies
(17:23):
kind of categorize you and me aspolicy holders rather than
consumers is kind of interesting.
It indicates that, right.
So that was the fundamental.
You know, I feel problem thatpeople saw and built around that
and try to build, like you know, nice apps and you know cost
consumer facing products aroundinsurance and you can buy
(17:45):
quickly.
You know we pay you quickly aclaim out and so on and so forth
, and all of this was great.
But again it happened in, likeyou know, large, saturated
markets where it was really hardto acquire the good customers
and you were competing againstthe behemothists that spent like
a billion dollars on customeracquisition every year.
So as a startup, even if youraised a lot of money, it was
(18:06):
really hard and is really hardto compete against those
incumbents.
So, going now to insure techwave two you know I feel
insurance is a very, very bigmarket, obviously, when you look
at it, and besides, like youknow, home auto renters, there's
tons of other insuranceproducts and niche insurance
(18:27):
categories where you knowinsurance operates and you know
commercial insurance obviouslyis a big one where I feel
there's a lot of opportunity.
But other categories that I kindof placed bets on is, you know,
for example, collectiblesinsurance.
Right, it's a very growingmarket.
The collectibles market is veryhuge.
(18:48):
But if you want to right now goand get your collectibles let's
assume you have a collection ofcollectibles of like a million
dollars if you call your homeowners insurance and be like hey
, can I get this?
You know insured under myhomeowner's policy, they're most
likely say no and if they sayyes, they give you a rider on
top of that which doesn't reallycover everything you have.
(19:10):
So there's no like reallyvisibility and some actually
want to have a real timeappraisal of your collection.
So it's a very cumbersomeprocess to get that and it goes
back to like what's the problemhere is that, hey, if you're
like an avid collector, it'sreally hard for you to get
insurance because nobody'sreally offering, you know, the
easy to way purchase ofinsurance.
(19:33):
So there's actually a problemand maybe we can build like a
new digital insurance productaround that and try to get it to
the consumer in variouschannels.
So I think insured tech wave twofocuses on these niche
insurance.
You know categories that havebeen you know that still to date
(19:53):
require a lot of manual.
You know processes oroperations.
So meaning you know you have tofill out like dozens of PDFs,
you have to send emails to youragents.
Your agent comes back and asks5,000 other questions to you.
Then you maybe have to gosomewhere to get something
appraised before you even get aninsurance quote.
(20:14):
So all of these you know, manualprocesses, I think, in various
insurance categories open up theopportunity for insured tech
wave two, where technology and,you know, digital underwriting
really makes sense because you,as of today, there's not much
there right.
And then the other thing heretoo, insured tech wave two,
(20:38):
going back to the learnings, isthat you know, if you really
focus on direct to consumer,it's going to be a very tough
road.
So most of the insured techs Ithink all of the insured techs
that I have back today arebasically mostly focused on
B2B2C.
So going through partners andpartner channels to distribute
your product is way morelucrative.
(20:58):
And I think that was also againa learning of insured tech wave
one, because all of the insuredtech wave ones kind of started
that to consumer and thenquickly realized, oh man, you
know it's hard, so maybe weshould, you know, talk to some
partners or agents or agenciesand see how they can distribute
our products.
Ali Safavi (21:17):
Very interesting.
So let me ask you a fewquestions more specifically now
in terms of how you're seeingthe future.
Number one is when we talkabout niche product sets, the
idea of it is great, right, butat the same time, we want to
make sure we're going out forideas that are big enough
markets.
Sure, the first thing that I'velearned and I've had is that a
lot of markets are recent,because, to me, insurance
(21:38):
industry has been one of thebest industries when it comes to
sales.
Industry has been built aroundsales and products like, let's
say, disability insurance.
The reason that is not as bigas life insurance is because
it's just harder to sell andjust having a startup isn't
justified like Mark and again.
For me maybe, like I don't evensee disabilities in each market
(22:00):
, but I'm saying for marketopportunities much smaller.
How many markets do you thinkare out there for insurance that
are large enough, markets thatare busy, backable and record
through digital innovation,versus just a company with an
underwriting model that is notreally like a software company?
Amir Kabir (22:17):
Yeah, absolutely, it's a very great question.
So when I kind of started ininsuretech world or investing
world, I always was like, hey,if a startup can have a billion
dollar outcome and you as aventure investor have 10% and
again I'm talking about like aseed stage investing and funds
(22:39):
that are, you know, smaller than$200 million, right, not like
the big behemoth that haveraised $5 billion.
You know, if you have a billiondollar outcome and someone owns
10% man, you kind of eitherreturn to your fund or you kind
of return half of your fund,which is amazing.
So if you have a couple ofthose, that's great.
But obviously, you know, whenyou look at these niche markets,
(23:04):
most of them are maybe $10billion, some of them are $20
billion, but the good thing hereis compared to like the $80
billion market and homeinsurance, the way I look at it
is like if you operate in this$20 billion market where nobody
really has technology, wherenobody is really trying to
innovate on the product side,where you know in comments are
(23:27):
still, you know, doing theirthing and think they're working,
and actually in some of thosemarkets you have, some of those
markets are very underserved aswell, meaning customers want
insurance but they don't.
They can get it or you knowit's hard to get for them.
If you can get like 10% of thatmarket or 20% of that market,
(23:47):
which would be $2 to $4 billion,and the outcome of your startup
could be around that, it's waymore lucrative than going for
like the $80 billion $90 billionmarket, which is highly
saturated, like incumbents arebasically dominating it.
There is no nobody'sunderserved there.
Like maybe a very tinypercentage of consumers cannot
(24:11):
get insurance and your digitalproduct will not really make
that big of a difference becausegoing to a state firm and
getting your home insurance willtake maybe two hours or maybe a
day and then getting your homeinsurance to an app might take
three minutes.
You know it doesn't make reallythat much of a difference
because the price will not makea difference.
(24:32):
So you invest in companies thatare going for these markets and
I have 10%, maybe 15% ownershipand that company exits for like
a billion, a billion, twobillion.
You know I have a great outcomeand that's totally fine for me,
right?
I think?
Going back to your question andthis is always the this has been
(24:53):
kind of something that wentthrough my mind.
I was thinking about a lot andmaybe people will challenge me
on that.
I feel like there's no way tohave $50 billion exit like a
Databricks or whatever you wantto call it, snowflake, whatever
all of these SaaS enterprisesoftware companies that exit for
(25:13):
a huge amount of money.
I don't feel we will see any$50 billion in short tech
outcome.
I mean I would be surprisedbecause think about it.
Look at, like, going back tothe incumbents, right.
I mean look at progressive,like state firm, whatever their
book of business.
I think they're writingwhatever $40 billion in premium,
$50 billion, very diversifiedbook of business, right?
(25:34):
They have home auto renters andall sorts of like other stuff.
What's their market value?
What's their market cap?
Ali Safavi (25:41):
I would say it's one to one to two depends on how
good it is like a book.
Amir Kabir (25:46):
Correct, Correct, yeah, yeah.
So one thing, one other, Ithink kind of, and it goes back
to like the version one and likethe niche kind of businesses,
right.
I think the best kind ofcomparison here is it's Kinsel.
You heard about Kinsel CapitalGroup, right?
I mean, they're kind of focusedvery much on like the niche
(26:08):
commercial market andspecialized insurance and their
market cap right now is ataround $7.6 billion.
Ali Safavi (26:17):
So I think you know that is a commercial risk right
Like that is large commercialniche risk Correct.
Amir Kabir (26:25):
But still niche and specialized right.
Ali Safavi (26:26):
So the reason I'm saying that is because to me
that's not a tech outcome asmuch as it is, because maybe
it's like the tech out here isthe thing that goes back to like
.
Amir Kabir (26:37):
That goes back to like insurance has to play a
fundamental role and tech willenable it.
Right, so you can't be like, oh,I'm a tech, I'm insured tech
and I want to be that wasanother thing Like, people value
these companies that you know,tech companies, comparables and
SaaS companies.
I don't think you can do that.
So that's why you know, if youspecifically operate in the, in
(27:00):
the so-called challenger worldof insured tech meaning those
who go after risk, after productand distribution the first
thing that matters there is theinsurance fundamentals.
The second thing is that thebook of business that you're
valued on and how profitable itis.
And I think third is that ifyou have a lot of good
technology that enables you tohave this good book of business,
(27:22):
because you're better inUnderwriting, then you can be
like, hey, this is the addedvalue I have.
But just saying, hey, because Ihave digital underwriting, I
have a great app and you know Ijust you group this product.
That's why I'm a tech company,insurance and insured tech.
I feel like that's.
That's hard to justify.
Ali Safavi (27:40):
There's definitely a line or there's no line.
Maybe I think that's that's.
That's another point whichbetween where does a company,
where's a company called astartup and what is a company
just called a company?
That is like small business.
Amir Kabir (27:54):
Sure.
Ali Safavi (27:56):
Because if we're talking about very large
commercial niche, these areusually capital plays right,
like, like companies that, likethe one they mentioned, they're
typically companies that arebacked by like a huge balance
shooter, like a big PE, that arejust taking very large risks
and they usually, you know, findthemselves in their
underwriting skills to be ableto, like, take on those risks
(28:18):
when other people can go to aLloyd's Crescent or whatever.
That is the idea here, from whatI understand, is that these
companies that have alwaysexisted, if they use more
technology in their underwriting, we think it might give them an
edge to better underwritingthis one off risks.
But these one off risks in alot of cases are not repeatable
(28:40):
businesses.
These are like one off risks,like a Lloyd model, right?
So if we're talking aboutrepeatable risks that you could
create a better AI engine, thatI'm very repeatable, it keeps
doing it then I wonder how manymarkets are like that out there
to say, like we're going tocreate, like a better insurance
for X and it's a market thatcustomers are looking for.
(29:05):
It's not a market they have topush it down to the throat and
it's a large enough market.
I just don't know how many arethere, unless we just go for,
like these, niche applicationsthat are like I'm just gonna set
up a capital group and usetechnology to better write every
individual risk that comes tomy door.
Amir Kabir (29:21):
I mean that goes back to like also when you look
at the article about the nextwave.
I think what you're sayingtouches base on what I think is
like the opportunity in embeddedinsurance.
Right, this is like a kind ofhas been a hot topic for quite
some years and I feel theembedded technology can enable
(29:41):
you know a bigger outcome here.
If you have the right tool sets, because you know, if you have
kind of an embedded play andyou're able to, you know, get a
ton of like input from a dataperspective and underwrite
better, you can maybe, you know,win also in those kind of flow
(30:05):
businesses, right, the wholemodel, renters, and kind of have
a bigger outcome there.
But I feel like, honestly, likeyou know, I know that venture
investors and we ourselves too,obviously we always want the
bigger, you know markets.
But I give you, I give anexample RPA, right, robotic
(30:25):
process automation.
Six, seven years ago when Ilooked into that, the market
size was five billion, I mean.
Now look at UI path, right, Imean, and so I think in every
like startup that had or has orhad a big outcome.
Even Uber, right, look at Uber.
(30:46):
I mean they started as a blackcar business in San Francisco
where the market size is 100million, and if you would have
said, okay, this is it right, Imean there would be a hard, hard
, hard to come to justify.
But eventually they kind ofwent into different kind of you
know categories around you knowtransportation, and I think it's
the same case here with thoseniche insurance businesses,
(31:08):
right.
As an example, I invested inCovertree.
They're focusing on homeinsurance for manufactured homes
, right, and it's a completelike.
It's a different kind ofunderwriting there, different
risk.
The market size, if you look atthat from the premium
perspective, just formanufactured homes, is around 10
billion a year, right, which Ithink is still great, but which
(31:31):
is great here there's a lot oflike adjacent other insurance
categories that fall into placethat you can, you know, play in
as well.
If you're Covertree, right,because if you look at the
demographic that lives in, youknow, in manufactured homes,
most of them have an RV, most ofthem have a motorbike, some of
(31:51):
them have a boat, right.
So if you look at those kind ofthree sectors in total with,
like, manufactured homes, youbasically are now at around, I
think, 50 billion market size,right.
So that's, I think, where youcan play and where, if you win
one category and you can applyyour technology, you're
underwriting whatever you haveinto others and your
(32:11):
distribution.
You can make a bigger outcome.
Ali Safavi (32:14):
So this is an example a 10 billion dollar
market.
If a company set up as an MGAand doesn't carry a risk, that
means they're probably aretaking their revenue be around
20% off it or less.
Amir Kabir (32:25):
It depends 20 to 30, depends on how you you know
what they negotiate.
Ali Safavi (32:28):
Party is like, then the market size is only two or
three billion from a revenueperspective, unless they carry
the full risk.
So we're talking about two orthree billion dollars in entire
market if they captured in totalmarket.
And but then you have to reallybet or prove that you could
sell adjacent markets to thosecustomers, which goes back to
(32:49):
the initial problem that it'sseen, which is a lot of people
already have other insurance andthen getting them to move over
the things that you already haveis going to be hard.
That's why I keep wondering,like how many big opportunities
are out there on paper?
I mean, it's indefinite,because you could always create
a better company, but, as yousaid, it's very hard to move
people from very incumbentcarriers.
Amir Kabir (33:08):
Yeah.
And so, going going back tothis, I think yes, if you just
look at it from the MGAperspective, I think it's will
be really, really hard to IPOand MGA Right.
So the intermediate outcomewould be here that an incumbent
that you know is alreadyservicing in that space comes
and buys the startup because itsees the technology and can plug
(33:30):
it into their ecosystem.
Right.
And yeah, yeah, exactly.
The outcome will probably notbe in the billions but maybe in
the upper 100 million or more.
But I think, from myperspective and the companies
that I've bet on, my goal is,with these companies, if it
works out on the MGA side andwe've seen it in the version one
(33:50):
, two, you know most of thembecome risk bearing entities
right and become eventuallycarriers of some sort and can
take risks themselves, and sothat's where the or you think
that should be part of theinsurance workup.
And only and only.
I feel, if you have figured itout and I think that was also, I
(34:11):
feel, a learning from versionone a lot of these insured texts
might have pivoted quickly tobecome a carrier or even started
out as a carrier.
I mean, look at Root, right.
I mean they started as acarrier.
And I mean, if you look at themarket size and market care, I
mean the outcome is not great.
Sure, they went public forwhatever six billion, but if you
look at it right now that themarket value is less than 100, I
(34:35):
think.
So I feel like you can onlybecome a carrier if you
gradually move into that, right?
Because, again, why do youbecome a carrier?
Is that, hey, I have the data,I have the underwriting skills
and I've operated into thatmarket, so I know what could
(34:56):
risk look like, and that's why Iwanna take risk myself rather
than like, hey, I wanna start asa carrier.
I think I know whatunderwriting works looks like, I
have this great data set andlet me do it right.
So I have another company Iinvested in which is actually a
full-stack carrier.
Right, they're a full-stackcarrier operating in the Latin
(35:18):
American market and selling lifeinsurance.
The reason there, why I wasexcited and I think that's the
best way to go is because theteam has done this twice before.
They had the data, they had thelearnings from the way they did
it before, and so for them itmakes the most sense to do that.
(35:39):
And again, latin America is waydifferent, too, in terms of the
market.
So they're the first kind ofdigital life insurance product
that exists more or less in theLatin American market, and they
distribute it through agents andagencies and brokerages rather
than, like, going back toconsumer.
Ali Safavi (35:58):
I think a part of the thesis that I'm very much
agreeing with you on is that youcan't I think a big learning, I
agree which is as an investor Ilearned the hard way and then
as an operator also.
Again, you can't just go to amarket that exists and compete
with auto insurance and say I'mjust digital, so I'm better,
because just being digital isnot value on its own, apart from
(36:20):
answering why and the wise aretypically not the right wise
when people pitch them.
So I'm there with you.
I think if we're truly toinnovate, we'll probably have to
innovate on distribution, oryou have to innovate on
unwriting, just being moreprofitable, or one thing that is
becoming a bigger thing, whichI saw in your notes as well,
which is around risk management.
(36:41):
How do we create a completesolution that manages risk, as
opposed to just sellinginsurance?
Cyber is a good example, wherewe have cybersecurity and risk
and all that.
Tell me everything about therisk management component as
part of the future MGA routes orfuture of insurance as a whole.
Amir Kabir (37:00):
Yeah, again, I mean that also depends on what market
you're operating in, right?
I think cyber is a good example, risk management is a big
component of that and I mean Ithink IoT was kind of like the
forerunner here, where you woulduse sensor data to risk
mitigate and then also doinsurance and one market
(37:22):
actually going back to insuredthat.
Wave two, which I alsomentioned in my article, is
medical malpractice, right,medical malpractice, you know,
in an old market, highlyunprofitable.
If you look at all the carriersthere, there's only one or two
carriers that are really doinggood risk, or maybe like a
handful, and the rest is justlike losing entities and again,
(37:46):
nobody there, or like very fewplayers have some sort of a
technological advantage and themarket they also.
I think medical malpractice isaround $15 billion, which
includes physicians, physicianassistants, nurses and, like you
know, big hospitals.
It's around $15 billion.
(38:06):
But going back to riskmanagement, I think you can
apply a lot of technology thereto you know, analyze how risk or
assess risk within, like youknow, physicians' offices or
hospitals and therefore be ableto provide better insurance or
(38:26):
mitigate.
You know claims, right.
So it really depends, I think,on the market when you think
about risk management associatedwith that.
I mean on a personal side, asyou probably know as well.
You know having these donglesin the cars and like data around
the cars, kind of like the riskmanagement component as well.
(38:49):
But interesting enough I don'tknow if you saw that today I was
reading about that Tesla isobviously, like you know, on the
forefront here with sellinginsurance to their customers and
basically saying, hey, giventhat we have all the data we
have like real time visibilityand the computer that you're
(39:09):
driving, we can give you betterinsurance.
But here apparently, you know,there's still a lot of learnings
to do, because a couple ofcustomers are suing actually
Tesla because the Tesla kind oflike wrongfully assess the data
(39:30):
around their driving behaviorbasically inflated car insurance
premiums based on misleadingcollision warnings and not
actual driving behavior, right,so that's kind of interesting.
Ali Safavi (39:43):
And you were like look at that and thought about
that, yeah.
So last thing that I'm alsoreally curious to hear your
thoughts on is what about allthis LLM AI stuff?
What are you excited about?
Do you feel like they're gonnamake a big change and what are
you thinking gonna happen?
And just wanna impact them.
Amir Kabir (40:00):
Yeah, I think, like I mean, ai is first of all,
nothing new, right, I mean ithas been around for decades.
I think what kind of now changeda lot of things is that it's
chat, gpt, and I think that'swhere, like everybody's like, oh
my God, like there's so muchopportunity here because like we
(40:20):
finally can talk to a computerand we get like real time
answers.
I think, from the insuranceperspective and again, as I
outlined also in my article,there's so many other problems
that we have to solve first,that now saying that AI and ML
will solve this all for us isthis kind of far fetched, right?
(40:43):
I mean, where this technologycan play a role, I think is
definitely on the claims side,right, you know, on a claims
part, I feel AI and ML and datacan play a huge role in
streamlining the claims processand making it more smooth, which
, again, if you do that, youwill have better and more
(41:05):
satisfied customers and so youwill have better retention when
it comes down to customers.
But in terms of, like, thesales process, you know you can
probably embed some AI and MLthere too.
But again, learning, I think,from version one was as well
that, depending on which marketyou operate in and the higher
(41:27):
the premium volume is, thehigher the likelihood that
someone wants to talk to someoneright?
So fully digitalizing aninsurance process and the sale
and servicing and the aftermath,I think is maybe doable and
very easily insurable risk right, but not like in the risk
(41:52):
categories that are associatedwith like bigger premiums.
Ali Safavi (41:57):
And last question for you if you were to start a
company today in insurance, whatcompany would you start or what
idea would you want to workwith?
Amir Kabir (42:05):
Again, I'm very much interested in you know products
and distribution and how youcan better create an insurance
product for a specific marketright, because most of these
products have been old or likehave been around for a long time
and I've looked in depth intoyou know medical medical
(42:27):
practice and I feel like that'sa very, very interesting space
and there's a lot of opportunityand there's actually one
company that is doing exactlywhat I was thinking of doing and
hopefully that's successful andhopefully maybe I can be part
of that from a ventureperspective.
Ali Safavi (42:44):
Cool.
Well, Amir, I took a lot ofyour time.
I appreciate you being on thepodcast and Thanks, man.
That was good I appreciate it.
Thanks for listening and enjoythe rest of your day.
Amir Kabir (42:57):
Thanks so much.
Thanks for having me Talk toyou soon.
Thank you.
Hello everyone.
My name is Ali Safavi.
I'm the CEO and co-founder ofCOVU and your host for
InsureTech Unscripted.
Today I'm honored to have oneof my very good friends.
I'm here with Amir Kabir, ageneral partner at AV8.
Obviously, I've known Amir for,I think, almost the beginning
of the InsureTech wave, at leastthe one that really picked up,
and we've been friends since andhe's obviously one of the most
(00:42):
knowledgeable investors that Iknow in the insurance space and
he has a lot of thoughts andthesis on where he thinks the
space is going.
So we're excited to have you onboard on the podcast, amir.
Maybe you could tell us alittle bit more about yourself
before we dive deeper.
Amir Kabir (00:57):
Yeah, sure, man, thanks for having me excited to
be part of this and talking toyou.
Yeah, we have, I don't know.
We've known each other formaybe 10 years now, I'm not sure
, but maybe a little bit less.
But yeah, as you said, sincekind of the whole hype cycle of
insurance, of InsureTech,started.
But yeah, a little bit aboutmyself.
I've been investing in theInsureTech space for a little
(01:20):
bit over 10 years and I kind ofgot into the FinTech InsureTech
world.
Coincidentally, I don't have aninsurance background.
I come from the tech side.
I was at operating roles forenterprise startups back in
Europe and when I came to the USI worked for a venture fund
(01:41):
actually with 66 ventures on theEast Coast for a while and got
into very early stage FinTechand InsureTech deals and did my
first InsureTech deal, I thinkin 2013 or so, which was simple
insurance, and that's how I gotinto venture.
From there I made my way to theWest Coast and helped build up
(02:02):
Munich Reventures, which I thinkis very familiar to a lot of
people in the insurance worldwho are kind of like the first
kind of corporate venture VCthat dives into insurance pretty
heavily and invested in allsorts of InsureTech version,
ones that people know like nextinsurance, Insure HDVI and so on
(02:23):
and so forth, and I joinedAviAventures a little bit over
two years ago as the fourthpartner to kind of lead the
FinTech practice and haveinvested so far, I think, in
seven or eight companies in theInsureTech world, which is
basically built upon the thesisthat I have written about the
next day of InsureTech and whatI've seen over the last 10 years
(02:45):
.
Ali Safavi (02:46):
Very cool.
So I usually like to start myquestioning with one of the most
controversial ideas, but justgiven the fact that you already
have put your thoughts on paperso clearly, it makes my job a
lot easier.
So let's start with.
I mean, I think the way thatyou looked at InsureTech is
interesting.
You've looked at it in wavesand you tried to compare it to
(03:06):
before.
So maybe you could starttalking a little bit about the
history of InsureTech, changesin insurance as an industry and
as relevant today and as we kindof try to think about the
future, how the history kind ofcomes into play.
Amir Kabir (03:19):
Yeah, that's a good point.
So I started writing aboutinsurance I think over a little
bit over two years ago and whatI tried to do was, in the part
one, was to kind of trace backthe history of insurance and the
evolution of InsureTech and seewhat happened actually the last
100 years, specifically in theUS when it comes down to
(03:40):
insurance and personal insuranceand so on and so forth.
So when you look at the firstwrite up that I did and the kind
of graph that I created, youcan see the big behemoth of
today.
The incumbents like the StateFarm and Geico and Progressive
(04:01):
kind of started in the veryearly 1900s, right 1920, 30
years around, and then the firstkind of technological wave in
insurance kind of came aboutwith the fact that it was within
like the 1980s, right I think1983 applied systems kind of
(04:21):
started 1985, insurity, selectquote 1997, insurion and so on
and so forth.
So these were kind of in 2000like guide buyers, so that the
80s, 90s kind of defined, I feel, the first kind of wave of
technology in insurance andgoing forward.
(04:44):
When you look at the graph andwhat happened is that within the
2000s a lot of like the firstversion of InsureTech, I would
say, kind of emerged which werebasically focused on price
comparison websites and somesort of distribution models
within insurance, like veryrudimentary distribution models,
(05:07):
and then in 2010 and onwards tolike 2020, that's where I feel
like the kind of term InsureTechwork was kind of minted let's
say, right, they're going intothe crypto world minted.
The InsureTech work has kind ofminted because people were
basically comparing likeinsurance as part of the
(05:32):
financial services ecosystem toFinTech and kind of counter word
here was InsureTech and I thinkthat emergence of that word,
insuretech, coupled with likeventure capital money flowing
into the ecosystem, started thefirst wave of InsureTech.
And within 2012, 13, 14, 15,obviously, the wave of
(05:56):
InsureTech took up and all ofthe prominent InsureTech
companies that people know about, as I mentioned before next
insurance lemonade, root, insure, oscar, health on the health
insurance side, ladder, life andso on and so forth.
(06:16):
They started around that timeand we had some kind of
interesting exit scenarios in2020, 21 where people the first
kind of the wave of insure taxwere able to create some sort of
a liquidity for early investorsand however quickly we saw that
(06:44):
like the public marketsdismissed somehow what insure
tax had built and the underlyingbusiness economics, they were
not really supporting the way togo public.
And so right now I feel likewe're entering insure tax, wave
(07:05):
two, and that's like based onthe article that I wrote
subsequently, where it's aboutlike what have you learned over
the last, let's say, maybe 10years when it comes down to
really insure tax, and how canwe move forward and what are the
learnings from there that wecan apply to the next day of
insure tax, if that makes sense.
Ali Safavi (07:28):
So let's jump into it.
So, basically, you're sayingthat this is the learnings from
insure tax 1.0, then, based onthat, insure tax 2.0 is created.
So, on a high level, what arethe big learnings that you feel
like a lot of investors orentrepreneurs have learned, or
haven't learned since or havemissed that you think worth
highlighting?
And then how's that insure tax?
Amir Kabir (07:48):
2.0?
.
I think the biggest kind oflearnings or miss was like hey,
technology will transforminsurance and everything will be
better, faster, cheaper andwhatever you wanna call it.
So obviously tech is part oflike insure tech.
The same way tech is part ofFinTech right, but in both
(08:12):
instances I feel, specificallybecause you are operating in a
very regulatory environment, theinsurance and the financial
services and FinTech have tocome first and I think this is
the biggest learning here thattechnology will not help you win
with an insurance.
No matter how great yourtechnology is, that will not
(08:34):
help you win and make big moneyfor you, and I think that was
the biggest learning here.
However, I think insure takewave one and again going back to
like why technology will nothelp you win, it's that because
insurance is not really a topline business.
(08:54):
So comparing insure tech tolike a SaaS company or like an
enterprise software company oreven like a consumer company in
the commerce world is completelyill equipped here, because it's
not a top line industry,meaning the more premium you get
in it doesn't mean you'rebetter or doing the better thing
.
It's more actually a bottomline industry where the
(09:17):
fundamental is unit economicsafter a line and the last ratios
play a very important role.
So I think this is the biggestlearning from my perspective
that I think people haverealized.
And then, on top of that, Ithink what people have seen as
well and goes back to thelearnings, is that insure take
(09:38):
wave one really focused on, likethe large, saturated markets,
highly commoditized productslike auto insurance, home
insurance, rents insurance.
When you look at those products, most of them, or some of them,
are mandatory.
Others are not.
But most of the people withinthat just focus in the United
(10:02):
States here have already autoinsurance.
Right, they have home insuranceinsurance.
So it's not that it's anunderserved market.
So what are you competing hereabout?
You're competing mostly aboutprice and maybe how do you
service customers.
So that was kind of the firstwave of insure take.
When you look at them, theywere able to quickly acquire
(10:25):
customers and show like a topline that people got excited
about.
However, the underlying, thebottom line, was not really
great here and over time it wasshowing and, honestly, if you
look at the public markets nowit still shows.
Right, those insure takes thatwent public and the numbers are
(10:45):
publicly available.
You can look at them.
It's not that they're making,they're the profitable
businesses.
So I think that was the biggestkind of thing here to focus on
like large, saturated marketswhere you basically are just
competing about price, right,because at the end of the day, I
(11:08):
have auto insurance fromprogressive and I pay, I think,
$140 a month.
If I get it for $50 a month,yeah, I take it right, but it
doesn't necessarily mean thatyou're covering your losses
there and it doesn't necessarilymean that you're getting the
right risk profile from acustomer perspective.
Right, and from insuranceperspective.
(11:29):
Those, I think also thelearnings are, those who switch
quickly or like look forinsurance, are actually the ones
you don't want, right.
And so the other thing is alsolike there was limited
differentiation, I feel, interms of like what, how, in
terms of underwriting, meaningwhat incumbents would do and
(11:50):
what insurtex would do was notmuch difference.
And it goes back to again it'sa very regulated industry.
So meaning like if you do autoinsurance and I've seen a bunch
of places too they would pitchme like, hey, we don't take FICO
into account, we take likeother stuff into account to
underwrite you.
Well, I'm like, well, that'sgreat, but you can't really do
(12:11):
that depending on which stateyou're in and the regulatory
bodies you have to comply withand you have to like, apply for,
like new products and so on andso forth.
So it's a very tough endeavorif you wanna really come up with
a new product.
So from the underwritingperspective there was also not
much differentiation.
Some would do more, some woulddo less, so it was not really
(12:33):
that much.
Going back to, like, theproducts that is insurtex
focused on were very like quoteunquote, easier products to
underwrite.
So and obviously thedistribution.
I think the learning there isthat customer acquisition is
(12:53):
very expensive and customerlifetime value, depending on the
product, is kind of limited.
So that means you constantlyhave to get new customers and I
said before, if you have toconstantly get new customers you
most likely will not get thebest risk out there.
So I stop here because I thinkthese are like the fundamental
(13:17):
things from the learningsperspective and what insurtex
Wave 1 kind of showed forventure investors and the
ecosystem.
However, I think it's veryimportant to outline that
insurtex Wave 1, the companiesreally navigated like incumbents
(13:37):
and service providers,consumers, regulators and the
broader financial market andshowing that, hey,
digitalization of insurance isinevitable and that's, I think,
a very important statement tomake, because all the work that
went into insurtex Wave 1 wasvery important, although maybe
(13:58):
the outcome of some of them, ormaybe most of them, was not
really great.
However, I think there was alot of learnings that can be
applied going forward.
Ali Safavi (14:05):
So let me ask you this like on paper and theory
and again I just want to makesure I kind of get the
hypothesis is that paperregulation changes and makes
Like is less involved with howunwriting is done for a product
like auto insurance, are yousaying, then we'll have truly
innovative startups in autoinsurance?
Amir Kabir (14:24):
Like Regulation is the main blocker for, you know,
like a true startup when in aspace like auto insurance, I
think so honestly because, again, you know the way you can go to
market is that you take anexisting product right, which a
lot of insure tech version oncedid, and you know so-called an
(14:44):
off the shelf product, and youdigitize that and, you know,
distribute it through variouschannels.
And then others you know took anoff the shelf product and came
up with new kind of innovativeunderwriting ideas around that
and had to first apply to getthat approved from the
regulatory body said to be ableto sell it to the consumer right
(15:07):
and others.
You know the other way to goabout it is also to use, you
know, ens paper and which is notso, which is not so
scrutinizing on a regulatoryside.
But I think that's that's.
I mean it's important thatregulatory barriers are there
right, otherwise everybody wouldgo around and, you know, do
(15:29):
whatever they want in terms of,like, writing insurance.
So I feel like there's no wayto get around it.
You have to work with those,you know, regulatory entities.
But the other learning here toois that you know it takes a
long time to get to market.
You know some insure tech tooksix to nine or 12 months to get
(15:50):
a product approved or a newproduct approved or a modified
product approved, and then theycould go to market right.
So the time to market is verylong and you need to have a long
breath if you want to win here.
Ali Safavi (16:03):
Very interesting.
Okay, and and obviously let'stalk a little bit about insure
tech 2.0 before we drill down insome of the areas that I was
curious to ask more about.
But now I'm talking about allthe learnings from insure tech
2.0.
Sorry, 1.0.
Let's introduce insure tech 2.0.
Amir Kabir (16:18):
So again, this is my kind of you know two cents here
, and the stuff that I wasthinking about over the last
years that I've seen is that, interms of like venture, I always
try to, you know, ask likewhat's the problem here, right,
you know that's?
That's, I think, a fundamentalquestion.
When you think about ventureand venture investing and
(16:40):
investing in startups is likewhat is the fundamental problem
that the startup is trying tosolve?
And insurance, you know, goingback to like version one, when
you think about it, you knowwhat was the fundamental problem
?
And from my perspective, thefundamental problem was that a
lot of people, or, like you know, the broader community or the
(17:02):
consumer, was thinking that youknow, there is not much customer
centricity in insurance, right,and people were like, hey, you
know, insurance companies reallydon't care about their
consumers.
That's why the NPS scores is solow and so on and so forth.
I mean, when you think aboutlike that insurance companies
(17:23):
kind of categorize you and me aspolicy holders rather than
consumers is kind of interesting.
It indicates that, right.
So that was the fundamental.
You know, I feel problem thatpeople saw and built around that
and try to build, like you know, nice apps and you know cost
consumer facing products aroundinsurance and you can buy
(17:45):
quickly.
You know we pay you quickly aclaim out and so on and so forth
, and all of this was great.
But again it happened in, likeyou know, large, saturated
markets where it was really hardto acquire the good customers
and you were competing againstthe behemothists that spent like
a billion dollars on customeracquisition every year.
So as a startup, even if youraised a lot of money, it was
(18:06):
really hard and is really hardto compete against those
incumbents.
So, going now to insure techwave two you know I feel
insurance is a very, very bigmarket, obviously, when you look
at it, and besides, like youknow, home auto renters, there's
tons of other insuranceproducts and niche insurance
(18:27):
categories where you knowinsurance operates and you know
commercial insurance obviouslyis a big one where I feel
there's a lot of opportunity.
But other categories that I kindof placed bets on is, you know,
for example, collectiblesinsurance.
Right, it's a very growingmarket.
The collectibles market is veryhuge.
(18:48):
But if you want to right now goand get your collectibles let's
assume you have a collection ofcollectibles of like a million
dollars if you call your homeowners insurance and be like hey
, can I get this?
You know insured under myhomeowner's policy, they're most
likely say no and if they sayyes, they give you a rider on
top of that which doesn't reallycover everything you have.
(19:10):
So there's no like reallyvisibility and some actually
want to have a real timeappraisal of your collection.
So it's a very cumbersomeprocess to get that and it goes
back to like what's the problemhere is that, hey, if you're
like an avid collector, it'sreally hard for you to get
insurance because nobody'sreally offering, you know, the
easy to way purchase ofinsurance.
(19:33):
So there's actually a problemand maybe we can build like a
new digital insurance productaround that and try to get it to
the consumer in variouschannels.
So I think insured tech wave twofocuses on these niche
insurance.
You know categories that havebeen you know that still to date
(19:53):
require a lot of manual.
You know processes oroperations.
So meaning you know you have tofill out like dozens of PDFs,
you have to send emails to youragents.
Your agent comes back and asks5,000 other questions to you.
Then you maybe have to gosomewhere to get something
appraised before you even get aninsurance quote.
(20:14):
So all of these you know, manualprocesses, I think, in various
insurance categories open up theopportunity for insured tech
wave two, where technology and,you know, digital underwriting
really makes sense because you,as of today, there's not much
there right.
And then the other thing heretoo, insured tech wave two,
(20:38):
going back to the learnings, isthat you know, if you really
focus on direct to consumer,it's going to be a very tough
road.
So most of the insured techs Ithink all of the insured techs
that I have back today arebasically mostly focused on
B2B2C.
So going through partners andpartner channels to distribute
your product is way morelucrative.
(20:58):
And I think that was also againa learning of insured tech wave
one, because all of the insuredtech wave ones kind of started
that to consumer and thenquickly realized, oh man, you
know it's hard, so maybe weshould, you know, talk to some
partners or agents or agenciesand see how they can distribute
our products.
Ali Safavi (21:17):
Very interesting.
So let me ask you a fewquestions more specifically now
in terms of how you're seeingthe future.
Number one is when we talkabout niche product sets, the
idea of it is great, right, butat the same time, we want to
make sure we're going out forideas that are big enough
markets.
Sure, the first thing that I'velearned and I've had is that a
lot of markets are recent,because, to me, insurance
(21:38):
industry has been one of thebest industries when it comes to
sales.
Industry has been built aroundsales and products like, let's
say, disability insurance.
The reason that is not as bigas life insurance is because
it's just harder to sell andjust having a startup isn't
justified like Mark and again.
For me maybe, like I don't evensee disabilities in each market
(22:00):
, but I'm saying for marketopportunities much smaller.
How many markets do you thinkare out there for insurance that
are large enough, markets thatare busy, backable and record
through digital innovation,versus just a company with an
underwriting model that is notreally like a software company?
Amir Kabir (22:17):
Yeah, absolutely, it's a very great question.
So when I kind of started ininsuretech world or investing
world, I always was like, hey,if a startup can have a billion
dollar outcome and you as aventure investor have 10% and
again I'm talking about like aseed stage investing and funds
(22:39):
that are, you know, smaller than$200 million, right, not like
the big behemoth that haveraised $5 billion.
You know, if you have a billiondollar outcome and someone owns
10% man, you kind of eitherreturn to your fund or you kind
of return half of your fund,which is amazing.
So if you have a couple ofthose, that's great.
But obviously, you know, whenyou look at these niche markets,
(23:04):
most of them are maybe $10billion, some of them are $20
billion, but the good thing hereis compared to like the $80
billion market and homeinsurance, the way I look at it
is like if you operate in this$20 billion market where nobody
really has technology, wherenobody is really trying to
innovate on the product side,where you know in comments are
(23:27):
still, you know, doing theirthing and think they're working,
and actually in some of thosemarkets you have, some of those
markets are very underserved aswell, meaning customers want
insurance but they don't.
They can get it or you knowit's hard to get for them.
If you can get like 10% of thatmarket or 20% of that market,
(23:47):
which would be $2 to $4 billion,and the outcome of your startup
could be around that, it's waymore lucrative than going for
like the $80 billion $90 billionmarket, which is highly
saturated, like incumbents arebasically dominating it.
There is no nobody'sunderserved there.
Like maybe a very tinypercentage of consumers cannot
(24:11):
get insurance and your digitalproduct will not really make
that big of a difference becausegoing to a state firm and
getting your home insurance willtake maybe two hours or maybe a
day and then getting your homeinsurance to an app might take
three minutes.
You know it doesn't make reallythat much of a difference
because the price will not makea difference.
(24:32):
So you invest in companies thatare going for these markets and
I have 10%, maybe 15% ownershipand that company exits for like
a billion, a billion, twobillion.
You know I have a great outcomeand that's totally fine for me,
right?
I think?
Going back to your question andthis is always the this has been
(24:53):
kind of something that wentthrough my mind.
I was thinking about a lot andmaybe people will challenge me
on that.
I feel like there's no way tohave $50 billion exit like a
Databricks or whatever you wantto call it, snowflake, whatever
all of these SaaS enterprisesoftware companies that exit for
(25:13):
a huge amount of money.
I don't feel we will see any$50 billion in short tech
outcome.
I mean I would be surprisedbecause think about it.
Look at, like, going back tothe incumbents, right.
I mean look at progressive,like state firm, whatever their
book of business.
I think they're writingwhatever $40 billion in premium,
$50 billion, very diversifiedbook of business, right?
(25:34):
They have home auto renters andall sorts of like other stuff.
What's their market value?
What's their market cap?
Ali Safavi (25:41):
I would say it's one to one to two depends on how
good it is like a book.
Amir Kabir (25:46):
Correct, Correct, yeah, yeah.
So one thing, one other, Ithink kind of, and it goes back
to like the version one and likethe niche kind of businesses,
right.
I think the best kind ofcomparison here is it's Kinsel.
You heard about Kinsel CapitalGroup, right?
I mean, they're kind of focusedvery much on like the niche
(26:08):
commercial market andspecialized insurance and their
market cap right now is ataround $7.6 billion.
Ali Safavi (26:17):
So I think you know that is a commercial risk right
Like that is large commercialniche risk Correct.
Amir Kabir (26:25):
But still niche and specialized right.
Ali Safavi (26:26):
So the reason I'm saying that is because to me
that's not a tech outcome asmuch as it is, because maybe
it's like the tech out here isthe thing that goes back to like
.
Amir Kabir (26:37):
That goes back to like insurance has to play a
fundamental role and tech willenable it.
Right, so you can't be like, oh,I'm a tech, I'm insured tech
and I want to be that wasanother thing Like, people value
these companies that you know,tech companies, comparables and
SaaS companies.
I don't think you can do that.
So that's why you know, if youspecifically operate in the, in
(27:00):
the so-called challenger worldof insured tech meaning those
who go after risk, after productand distribution the first
thing that matters there is theinsurance fundamentals.
The second thing is that thebook of business that you're
valued on and how profitable itis.
And I think third is that ifyou have a lot of good
technology that enables you tohave this good book of business,
(27:22):
because you're better inUnderwriting, then you can be
like, hey, this is the addedvalue I have.
But just saying, hey, because Ihave digital underwriting, I
have a great app and you know Ijust you group this product.
That's why I'm a tech company,insurance and insured tech.
I feel like that's.
That's hard to justify.
Ali Safavi (27:40):
There's definitely a line or there's no line.
Maybe I think that's that's.
That's another point whichbetween where does a company,
where's a company called astartup and what is a company
just called a company?
That is like small business.
Amir Kabir (27:54):
Sure.
Ali Safavi (27:56):
Because if we're talking about very large
commercial niche, these areusually capital plays right,
like, like companies that, likethe one they mentioned, they're
typically companies that arebacked by like a huge balance
shooter, like a big PE, that arejust taking very large risks
and they usually, you know, findthemselves in their
underwriting skills to be ableto, like, take on those risks
(28:18):
when other people can go to aLloyd's Crescent or whatever.
That is the idea here, from whatI understand, is that these
companies that have alwaysexisted, if they use more
technology in their underwriting, we think it might give them an
edge to better underwritingthis one off risks.
But these one off risks in alot of cases are not repeatable
(28:40):
businesses.
These are like one off risks,like a Lloyd model, right?
So if we're talking aboutrepeatable risks that you could
create a better AI engine, thatI'm very repeatable, it keeps
doing it then I wonder how manymarkets are like that out there
to say, like we're going tocreate, like a better insurance
for X and it's a market thatcustomers are looking for.
(29:05):
It's not a market they have topush it down to the throat and
it's a large enough market.
I just don't know how many arethere, unless we just go for,
like these, niche applicationsthat are like I'm just gonna set
up a capital group and usetechnology to better write every
individual risk that comes tomy door.
Amir Kabir (29:21):
I mean that goes back to like also when you look
at the article about the nextwave.
I think what you're sayingtouches base on what I think is
like the opportunity in embeddedinsurance.
Right, this is like a kind ofhas been a hot topic for quite
some years and I feel theembedded technology can enable
(29:41):
you know a bigger outcome here.
If you have the right tool sets, because you know, if you have
kind of an embedded play andyou're able to, you know, get a
ton of like input from a dataperspective and underwrite
better, you can maybe, you know,win also in those kind of flow
(30:05):
businesses, right, the wholemodel, renters, and kind of have
a bigger outcome there.
But I feel like, honestly, likeyou know, I know that venture
investors and we ourselves too,obviously we always want the
bigger, you know markets.
But I give you, I give anexample RPA, right, robotic
(30:25):
process automation.
Six, seven years ago when Ilooked into that, the market
size was five billion, I mean.
Now look at UI path, right, Imean, and so I think in every
like startup that had or has orhad a big outcome.
Even Uber, right, look at Uber.
(30:46):
I mean they started as a blackcar business in San Francisco
where the market size is 100million, and if you would have
said, okay, this is it right, Imean there would be a hard, hard
, hard to come to justify.
But eventually they kind ofwent into different kind of you
know categories around you knowtransportation, and I think it's
the same case here with thoseniche insurance businesses,
(31:08):
right.
As an example, I invested inCovertree.
They're focusing on homeinsurance for manufactured homes
, right, and it's a completelike.
It's a different kind ofunderwriting there, different
risk.
The market size, if you look atthat from the premium
perspective, just formanufactured homes, is around 10
billion a year, right, which Ithink is still great, but which
(31:31):
is great here there's a lot oflike adjacent other insurance
categories that fall into placethat you can, you know, play in
as well.
If you're Covertree, right,because if you look at the
demographic that lives in, youknow, in manufactured homes,
most of them have an RV, most ofthem have a motorbike, some of
(31:51):
them have a boat, right.
So if you look at those kind ofthree sectors in total with,
like, manufactured homes, youbasically are now at around, I
think, 50 billion market size,right.
So that's, I think, where youcan play and where, if you win
one category and you can applyyour technology, you're
underwriting whatever you haveinto others and your
(32:11):
distribution.
You can make a bigger outcome.
Ali Safavi (32:14):
So this is an example a 10 billion dollar
market.
If a company set up as an MGAand doesn't carry a risk, that
means they're probably aretaking their revenue be around
20% off it or less.
Amir Kabir (32:25):
It depends 20 to 30, depends on how you you know
what they negotiate.
Ali Safavi (32:28):
Party is like, then the market size is only two or
three billion from a revenueperspective, unless they carry
the full risk.
So we're talking about two orthree billion dollars in entire
market if they captured in totalmarket.
And but then you have to reallybet or prove that you could
sell adjacent markets to thosecustomers, which goes back to
(32:49):
the initial problem that it'sseen, which is a lot of people
already have other insurance andthen getting them to move over
the things that you already haveis going to be hard.
That's why I keep wondering,like how many big opportunities
are out there on paper?
I mean, it's indefinite,because you could always create
a better company, but, as yousaid, it's very hard to move
people from very incumbentcarriers.
Amir Kabir (33:08):
Yeah.
And so, going going back tothis, I think yes, if you just
look at it from the MGAperspective, I think it's will
be really, really hard to IPOand MGA Right.
So the intermediate outcomewould be here that an incumbent
that you know is alreadyservicing in that space comes
and buys the startup because itsees the technology and can plug
(33:30):
it into their ecosystem.
Right.
And yeah, yeah, exactly.
The outcome will probably notbe in the billions but maybe in
the upper 100 million or more.
But I think, from myperspective and the companies
that I've bet on, my goal is,with these companies, if it
works out on the MGA side andwe've seen it in the version one
(33:50):
, two, you know most of thembecome risk bearing entities
right and become eventuallycarriers of some sort and can
take risks themselves, and sothat's where the or you think
that should be part of theinsurance workup.
And only and only.
I feel, if you have figured itout and I think that was also, I
(34:11):
feel, a learning from versionone a lot of these insured texts
might have pivoted quickly tobecome a carrier or even started
out as a carrier.
I mean, look at Root, right.
I mean they started as acarrier.
And I mean, if you look at themarket size and market care, I
mean the outcome is not great.
Sure, they went public forwhatever six billion, but if you
look at it right now that themarket value is less than 100, I
(34:35):
think.
So I feel like you can onlybecome a carrier if you
gradually move into that, right?
Because, again, why do youbecome a carrier?
Is that, hey, I have the data,I have the underwriting skills
and I've operated into thatmarket, so I know what could
(34:56):
risk look like, and that's why Iwanna take risk myself rather
than like, hey, I wanna start asa carrier.
I think I know whatunderwriting works looks like, I
have this great data set andlet me do it right.
So I have another company Iinvested in which is actually a
full-stack carrier.
Right, they're a full-stackcarrier operating in the Latin
(35:18):
American market and selling lifeinsurance.
The reason there, why I wasexcited and I think that's the
best way to go is because theteam has done this twice before.
They had the data, they had thelearnings from the way they did
it before, and so for them itmakes the most sense to do that.
(35:39):
And again, latin America is waydifferent, too, in terms of the
market.
So they're the first kind ofdigital life insurance product
that exists more or less in theLatin American market, and they
distribute it through agents andagencies and brokerages rather
than, like, going back toconsumer.
Ali Safavi (35:58):
I think a part of the thesis that I'm very much
agreeing with you on is that youcan't I think a big learning, I
agree which is as an investor Ilearned the hard way and then
as an operator also.
Again, you can't just go to amarket that exists and compete
with auto insurance and say I'mjust digital, so I'm better,
because just being digital isnot value on its own, apart from
(36:20):
answering why and the wise aretypically not the right wise
when people pitch them.
So I'm there with you.
I think if we're truly toinnovate, we'll probably have to
innovate on distribution, oryou have to innovate on
unwriting, just being moreprofitable, or one thing that is
becoming a bigger thing, whichI saw in your notes as well,
which is around risk management.
(36:41):
How do we create a completesolution that manages risk, as
opposed to just sellinginsurance?
Cyber is a good example, wherewe have cybersecurity and risk
and all that.
Tell me everything about therisk management component as
part of the future MGA routes orfuture of insurance as a whole.
Amir Kabir (37:00):
Yeah, again, I mean that also depends on what market
you're operating in, right?
I think cyber is a good example, risk management is a big
component of that and I mean Ithink IoT was kind of like the
forerunner here, where you woulduse sensor data to risk
mitigate and then also doinsurance and one market
(37:22):
actually going back to insuredthat.
Wave two, which I alsomentioned in my article, is
medical malpractice, right,medical malpractice, you know,
in an old market, highlyunprofitable.
If you look at all the carriersthere, there's only one or two
carriers that are really doinggood risk, or maybe like a
handful, and the rest is justlike losing entities and again,
(37:46):
nobody there, or like very fewplayers have some sort of a
technological advantage and themarket they also.
I think medical malpractice isaround $15 billion, which
includes physicians, physicianassistants, nurses and, like you
know, big hospitals.
It's around $15 billion.
(38:06):
But going back to riskmanagement, I think you can
apply a lot of technology thereto you know, analyze how risk or
assess risk within, like youknow, physicians' offices or
hospitals and therefore be ableto provide better insurance or
(38:26):
mitigate.
You know claims, right.
So it really depends, I think,on the market when you think
about risk management associatedwith that.
I mean on a personal side, asyou probably know as well.
You know having these donglesin the cars and like data around
the cars, kind of like the riskmanagement component as well.
(38:49):
But interesting enough I don'tknow if you saw that today I was
reading about that Tesla isobviously, like you know, on the
forefront here with sellinginsurance to their customers and
basically saying, hey, giventhat we have all the data we
have like real time visibilityand the computer that you're
(39:09):
driving, we can give you betterinsurance.
But here apparently, you know,there's still a lot of learnings
to do, because a couple ofcustomers are suing actually
Tesla because the Tesla kind oflike wrongfully assess the data
(39:30):
around their driving behaviorbasically inflated car insurance
premiums based on misleadingcollision warnings and not
actual driving behavior, right,so that's kind of interesting.
Ali Safavi (39:43):
And you were like look at that and thought about
that, yeah.
So last thing that I'm alsoreally curious to hear your
thoughts on is what about allthis LLM AI stuff?
What are you excited about?
Do you feel like they're gonnamake a big change and what are
you thinking gonna happen?
And just wanna impact them.
Amir Kabir (40:00):
Yeah, I think, like I mean, ai is first of all,
nothing new, right, I mean ithas been around for decades.
I think what kind of now changeda lot of things is that it's
chat, gpt, and I think that'swhere, like everybody's like, oh
my God, like there's so muchopportunity here because like we
(40:20):
finally can talk to a computerand we get like real time
answers.
I think, from the insuranceperspective and again, as I
outlined also in my article,there's so many other problems
that we have to solve first,that now saying that AI and ML
will solve this all for us isthis kind of far fetched, right?
(40:43):
I mean, where this technologycan play a role, I think is
definitely on the claims side,right, you know, on a claims
part, I feel AI and ML and datacan play a huge role in
streamlining the claims processand making it more smooth, which
, again, if you do that, youwill have better and more
(41:05):
satisfied customers and so youwill have better retention when
it comes down to customers.
But in terms of, like, thesales process, you know you can
probably embed some AI and MLthere too.
But again, learning, I think,from version one was as well
that, depending on which marketyou operate in and the higher
(41:27):
the premium volume is, thehigher the likelihood that
someone wants to talk to someoneright?
So fully digitalizing aninsurance process and the sale
and servicing and the aftermath,I think is maybe doable and
very easily insurable risk right, but not like in the risk
(41:52):
categories that are associatedwith like bigger premiums.
Ali Safavi (41:57):
And last question for you if you were to start a
company today in insurance, whatcompany would you start or what
idea would you want to workwith?
Amir Kabir (42:05):
Again, I'm very much interested in you know products
and distribution and how youcan better create an insurance
product for a specific marketright, because most of these
products have been old or likehave been around for a long time
and I've looked in depth intoyou know medical medical
(42:27):
practice and I feel like that'sa very, very interesting space
and there's a lot of opportunityand there's actually one
company that is doing exactlywhat I was thinking of doing and
hopefully that's successful andhopefully maybe I can be part
of that from a ventureperspective.
Ali Safavi (42:44):
Cool.
Well, Amir, I took a lot ofyour time.
I appreciate you being on thepodcast and Thanks, man.
That was good I appreciate it.
Thanks for listening and enjoythe rest of your day.
Amir Kabir (42:57):
Thanks so much.
Thanks for having me Talk toyou soon.
Thank you.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
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
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.