Julie Kheyfets, CEO of Block Renovation, joins us to share her insights on revolutionizing the home renovation experience through technology and AI. We discuss the challenges homeowners face and how Block addresses these issues while exploring ambitious plans for scaling and innovation in the industry.
• Julie's career journey and transition to CEO of Block Renovation
• The challenges in the home renovation industry and Block's approach
• Shifting from a one-stop-shop to a flexible marketplace for homeowners
• The role of AI and technology in simplifying the planning process
• Addressing homeowner overwhelm and decision fatigue
• The importance of information and safety nets in renovations
• Strategies for scaling the business and fostering a growth mindset
• Quickfire insights on leadership and productivity tools
All Links: linktr.ee/startup_recruiting
LinkedIn: www.linkedin.com/in/riecekeck/
Twitter/X: x.com/tech_headhunter
Recruitment: www.mindhire.ai
Youtube: https://www.youtube.com/@seedtoexitpod
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
I think different chapters of a company's lifetime
require different leaders.
Coda did a phenomenal job ofconceptualizing the company,
going after this reallydifficult problem space of
renovations, setting a fantasticculture of humility, excellence
, hard work, collaboration Allthose things will carry forward
(00:23):
In this next chapter.
What matters a lot is buildingout our AI capabilities, so
using artificial intelligence toput world-class tools for
design, planning and hiring inthe hands of homeowners, and
then scaling the company as well.
Speaker 2 (00:41):
Hey, thanks for tuning in.
Today I'm joined by JulieKafitz, the CEO of Block
Renovation.
Block is a company on a missionto transform the home
renovation experience throughtechnology, data and seamless
project management.
Julie brings a wealth ofexperience, having held
leadership roles at Facebook,tractable and the Boston
Consulting Group before steppingin as Block's Chief Operating
(01:01):
Officer and is now leading thecompany as CEO.
So in this conversation, we'regoing to talk about a few things
, including what drew Julie totackling efficiencies in the
home renovation industry, howBlock is solving one of the most
fragmented and complex consumerexperiences, how she's
transitioned from COO to CEO andhow she's starting to make her
mark on the company.
Lessons from raising capitaland scaling a fast-growing
(01:23):
startup.
And the role of AI, data andautomation in shaping the future
of home renovations.
So, whether you're a founder, anoperator or just someone who's
interested in how technology isreshaping old-school industries,
this episode is packed withinsights that you're not going
to want to miss.
No-transcript that power thestartup ecosystem.
(10:46):
I'm your host, rhys Keck,founder of MindHire, a talent
acquisition firm specializing inhelping startups build
exceptional teams.
Each week, I sit down withfounders, investors and industry
leaders to explore the journeysbehind iconic companies and
game-changing ideas.
Whether you're building,investing or just curious about
what it takes to succeed in thestartup world, I want this
(11:08):
podcast to be your go-toresource for actionable insights
and inspiring conversations.
Now, if you enjoy the show,please don't forget to subscribe
, leave a review or share itwith your network.
Your support means the worldand really helps bring more
incredible conversations to life.
Julie, thanks for coming on theshow.
Speaker 1 (11:25):
Thanks for having me, rhys, excited to be here.
Speaker 2 (11:27):
Likewise.
So, before we get intoeverything, you have a super
interesting background.
You've spent some time inconsulting.
You've spent some time at somereally well-known tech companies
like Facebook.
You had a VP role at Tractable.
I'm just curious, now thatyou've ended up at Block, could
you mind just to start thingsoff walking me through your
background a little bit, how yougot to where you're at today
and what drew you to applyingtech to the home renovation
(11:50):
space?
Speaker 1 (11:51):
Yeah, great question.
So I started my career inmanagement consulting at BCG.
I did that for a number ofyears, used that as an
opportunity to see a lot ofdifferent industries and
functions.
I really enjoy learning, so Igot to learn very broadly.
But doing that, what I reallywas missing was actually
(12:11):
operating.
In consulting, you write thestrategy deck, you do a lot of
good thinking about where acompany should go or what it
should do, but you don't do thething.
And I wanted to do the thingand actually take that strategy,
implement it, learn from it,iterate, and so I went into tech
after consulting and I spentsome time at Facebook on the
(12:32):
business operations team, whichwas really interesting.
This was over 10 years ago,starting in 2013, when Facebook
was still in this crazy growthmode and I was embedded in the
revenue organization in ad sales.
So I saw how that was scalingreally quickly, which was really
interesting very culturallycool, but still felt really big.
And at that point Facebook was10,000 people, a fraction of
(12:55):
what it is today.
But even then there was a lotof kind of meetings and decks to
get things done, and so I knewI wanted to go towards something
early stage.
I then spent four years atTractable, the company you
mentioned.
What Tractable does is computervision for accident and
disaster recovery.
So it has an AI platform thattakes photos and videos of
damaged cars and damaged homesand figures out how to repair
(13:16):
them, and that platform is soldto insurance companies, to
restoration companies, repairers, things like that.
And I launched and built outthe North America business for
Tractable, which was reallyinteresting.
So we essentially tried tofigure out what was the product
market fit in North America.
The company was headquarteredin London, had sold a bunch in
(13:36):
Europe and Asia and we had tobring on the first clients,
build out the team, build outthe go-to-market for the United
States and Canada, but not thego-to-market for the United
States and Canada and that.
So I did that from 2018 to 2022.
So before the latest wave of AI, at that point AI was much
harder to build.
You know, before building anauto damage assessment AI, we
(13:57):
had to teach a model how to evenrecognize car parts Like what
is a hood, recognize that thereis a car hood in this photo.
That took months of work, youknow, let alone assessing the
damage to the hood or how torepair it.
So it was a different era of AI, but it was really cool to see
that brought to market and tobring it to market and put it in
the hands of millions of users.
(14:18):
When Block approached me, I justbecame so intrigued by the home
renovation space, in partbecause it is such a massive
market.
Americans spend $470 billionevery year renovating their
homes.
That's not includingmaintenance or small repairs.
That is major renovations, andit seems like almost everybody
(14:39):
still has a nightmare of anexperience that's both
homeowners and contractors andso what really intrigued me was,
you know, just seeing a spacethat was so large and still so
difficult.
There aren't many industriesleft like that.
It also is largely untouched bytechnology.
So there are lead generationplatforms like Angie's List and
(15:02):
Thumbtack, you know.
There are CRMs for contractors.
There isn't something reallythat creates a great experience
for homeowners and contractors,and that's what we're trying to
do with Block.
So what really excited me wasseeing this really difficult
space and trying to figure outhow to create a much better
experience for it withtechnology.
Speaker 2 (15:20):
Yeah, I mean I can 100% agree with you on what you
just described.
I mean, I personally have acouple of rentals and you know I
don't manage them personally,but you know, having to deal
with the contractors or themaintenance repair folks, it's
they're running off of phones,their things are being written
down on paper, it's almostcompletely untouched by
(15:40):
technology and it's it's frankly, it's a really frustrating
experience that I've had severaltimes as a homeowner.
So when I, when I came acrossacross just Block as a company,
I was like oh wow, this issolving a really interesting
problem.
So you came on there a coupleof years ago and talk to me
about what your experience hasbeen like at Block and what
you've done since then.
Speaker 1 (16:01):
Yeah, it's been really great.
We have a really great teamthat deeply understands the
space and is super motivated tocreate something much better.
When I came on, I came on aschief operating officer, so that
was two years ago, at the endof 2022.
And at that point our offering,our model, was still a bit of a
one-stop shop.
So you would come to us as ahomeowner.
(16:23):
We would match you with a greatcontractor from our network for
your build, and then we wouldalso support you with things
like your architectural plans,your permit with the city, your
approval if you're in a co-op ora condo.
We would source, warehouse anddeliver on a single pallet all
of your materials.
So you didn't have to wait forhundreds of deliveries coming
(16:46):
piecemeal.
So it was a reallyoperationally intensive process
and that process and thatoffering made sense for a small
segment of homeowners who didn'thave strong preferences.
They kind of said, okay, I'llpick that template and you guys
handle it for me.
I'm going to go work at mysuper busy job.
But if you had very specifictile or fixtures that you loved
(17:09):
like, let's say, you travel toSpain and you had found some
gorgeous tile, you had shippedover something else it was hard
for us to accommodate because wehad this templated, streamlined
process, and so what weactually did over the last
couple of years is evolve theoffering.
So we stopped requiringhomeowners to use our materials,
follow our architectural plans.
We leaned in much more to themarketplace side of the platform
(17:31):
, which meant helping homeownersto discover the right
contractors for them, getproposals, evaluate those
proposals, negotiate andtransact on the platform.
We also started to build out awhole suite of planning tools.
On the platform.
We also started to build out awhole suite of planning tools.
So all the thinking, the workthat happens before you hire a
contractor, just trying tofigure out what you're going to
(17:52):
do with your home, how muchit'll cost, what it'll look like
.
We support that too.
Speaker 2 (17:55):
Got it.
So you recently took on the CEOrole, just as recently as a few
weeks ago, so congratulations.
So the founder was Coda, so hedecided to step back.
He's still serving on the board.
Talk to me a little bit moreabout what it was like as you
started to have thoseconversations, how you thought
(18:15):
about it and just really how itall came to be, how you thought
about it and just really how itall came to be.
Speaker 1 (18:20):
Yeah, you know, leadership transitions can be
complex, especially at startups,which often don't have the same
reps as large companies do.
Transitioning CEO to CEO.
Kodai did a really fantasticjob of thinking about this
pretty far in advance and givingus the time to plan it and
execute it, and so we were justreally deliberate about how to
(18:43):
do this transition, how to havethe right conversations with our
investors and our board, withour team, how to set up the
leadership team around me tosupport this transition.
He is still really involved asexecutive chair on the board,
just less in the day-to-day, andthe fact that we've been
working side-by-side as CEO andCOO for the last two years that
helped a lot as well.
Speaker 2 (19:04):
I can imagine that, as you step into the CEO role,
you want to make an impact, youwant to put your proverbial
stamp on things, but then also,at the same time, you want to be
respectful of the legacy andwhat has been built so far.
How do you navigate, balancingthose two things?
Speaker 1 (19:20):
Yeah, I think different chapters of a
company's lifetime requiredifferent leaders.
Coda did a phenomenal job ofconceptualizing the company,
going after this reallydifficult problem space of
renovations, setting a fantasticculture of humility, excellence
(19:40):
, hard work, collaboration allthose things will carry forward
In this next chapter.
What matters a lot is buildingout our AI capabilities, so
using artificial intelligence toput world-class tools for
design, planning and hiring inthe hands of homeowners, and
then scaling the company as well, and those are things where I
(20:01):
bring the right track record,and so I think there's natural
complementarity in thefoundation that Coda has built
and the momentum that I caninject for this next chapter.
Speaker 2 (20:11):
When you talk about scaling, what does that look
like in terms of the vision ofwhat you want to do next?
Speaker 1 (20:16):
Yeah, we started to expand geographically a lot more
quickly, more aggressively thisyear.
So initially Block launchedjust in New York City,
especially with our much moreoperationally complex 1.0
original model.
It was hard to go and bringthat model elsewhere.
We'd need to open up warehouses, put up supply chains.
So in the first six, sevenyears we opened up only a couple
(20:41):
of more cities.
Just three months ago wealready opened up Chicago,
philadelphia, connecticut.
We will continue to expand cityby city across the country,
probably every six months or so.
So that's a big part of it.
The other part is just handlingmore of the renovation journey.
So historically we've focusedso much on helping homeowners to
(21:01):
hire contractors and to executetheir projects.
But that's the last 10% of thejourney.
The average American renovatestheir home about every three
years and 90% of that threeyears is just trying to figure
out what to do.
You know you wake up and youlook at your kitchen.
You're like this is kind ofoutdated, I don't know the right
counter space, I just wish itcould be right.
(21:24):
It just doesn't feel right.
But going from it doesn't feelright to actually knowing what
to do and how much it'll costand who to hire, that is what we
call planning, and that'susually a multi-year process.
So more tooling around that.
Putting more data athomeowners' fingertips.
So more tooling around that.
Putting more data athomeowners' fingertips, helping
them to figure out design scope,budget iteratively on their own
schedule, without having tohire an architect.
(21:46):
That'll cost them $20,000.
Speaker 2 (21:50):
That's where we're focused on the product side as
well.
Why does it take years?
Is it just a matter of nottaking action and knowing it's
just one of those things thatneed to be worked on, but you
just haven't bothered to do it.
Speaker 1 (22:02):
That's such a great question.
I think it's largely because itfeels overwhelming.
The thing we see over and overagain is that a lot of
homeowners just feel stuck.
They know they want to changetheir space, but when it comes
to figuring out what to do orhow much to spend, or even where
(22:22):
to start should I do my kitchenfirst or my bath first they
Google around, they ask around,but there's no reliable source
of information and it feels likesuch a consequential set of
decisions that people just getoverwhelmed and they stop.
And then they start again.
They start to look at Pinterestagain and Instagram again.
They're like oh, I really likethat.
But then, going from I reallylike that to actually investing
(22:45):
the capital, trying to figureout do I need to move my family
to a hotel, you know, can Iactually live through the
renovation?
It just seems so overwhelmingand there are hundreds of
decisions you have to make.
So what we're focused on is howdo we empower homeowners, put
the right tools and data intheir hands and the right
support in their hands so theyfeel empowered, and so they
(23:07):
don't feel alone.
Speaker 2 (23:07):
So what is building that process look like from an
AI perspective?
I'm just curious.
I'd imagine it's multimodal.
I'm sure there's a chatbotbuilt in.
I'm sure you can upload imagesand say hey, here's the space.
What is?
What are the AI capabilitiesthat you're actually building?
Speaker 1 (23:21):
Yeah, yeah, I love talking about this because
there's so much you can do withAI.
Today.
There's a lot of AI hype, whichI'm always aware of.
You know.
One thing we observed with therenovation planning specifically
is people really struggle toarticulate what they need and
what they want, but they alsohave very strong reactions to it
.
So if I show you a space, a lotof people, if they've been
(23:43):
thinking about renovating, willinstantly say, oh, that feels
right, or oh, no, that's not forme.
You know, think of it like asimple bathroom, five by eight
feet.
I can one person.
You show them white, with whiteaccents, and they're like, oh,
that's so sterile, that's notfor me, and they want something
moody and dark.
You know blues and greens andblacks.
(24:03):
And you show somebody else thatmoody and dark, and they're
like, oh, that's not right atall.
I want something clean, modern,industrial.
And what we figured out is thereneeds to be an accessible,
delightful, easy way to showhomeowners different things that
are possible, to allow them tonarrow down what they really
have in mind.
(24:24):
And in a way, that's whatarchitects and designers do.
They show you mock-ups andyou're like, oh, I like that one
, but maybe not the cabinets.
Can we switch that out?
And you iterate.
So what we're building is aniterative process that gives you
different stimuli or examplesto react to and allows you,
through your instincts, tonarrow down.
The key, though, is doing thatin a way that feels personalized
(24:45):
to you, so I can show yougeneric bathrooms or kitchens or
spaces, but you might say, well, that one is twice as big as my
bathroom.
That's clearly not right for me.
That won't fit here.
Or that one has a window.
I don't have a window, so itneeds to feel sufficiently
relevant and personalized to youand your space.
And the hard part is that everysingle home is different.
Unlike new construction, whereyou have a piece of land that's
(25:09):
almost a blank slate withrenovation, you have a home
that's been there probably fordecades, that probably had work
done on it, permitted,unpermitted.
You might open up a wall, findstuff that you didn't know was
there, and so we need to accountfor that in kind of giving
people this information andthese options and these choices.
Speaker 2 (25:28):
Yeah, I was going to ask that actually because, again
, having lived through a coupleof renovations myself, you get
in there and you find somethingthat you didn't think was there.
There's always surprises, right, or scope changes, so how does
the platform and team adjust andadapt to that?
Speaker 1 (25:46):
Yeah, yeah.
That's one of the great joys ofrenovation you know there are
always surprises because, again,every home is different.
So we do a couple of things.
One is upfront.
We try to share as muchinformation as possible.
So when homeowners getproposals from contractors, they
see things like it might takethis long, to this long, like
(26:07):
these are the kinds of things wedon't know yet, that we will
see once we actually open upwalls and see what's there.
So I think information upfrontis something that's often
missing in the industry,especially for first-time
homeowners.
So often first-time homeownerskind of figure out their vision
and maybe they use onlineplatforms like Pinterest or
(26:27):
Instagram and they just seethese beautiful aftershots and
they think, oh, it's going to bea smooth journey, I just need a
trustworthy contractor to buildthat.
But your home is a constructionsite for a number of weeks and
there will be surprises.
So sharing information upfrontis really important.
The other thing that we do isalways provide a safety net, so
(26:47):
homeowners and contractors worktogether to get through these
surprises.
Our contractors are equipped todo that, but homeowners always
know.
If something is really strange,we're there to help.
So if there is a change order,your contractor comes to you and
says we opened up a wall.
There's stuff we didn't expectto see there.
That's going to cost another$10,000.
Usually the homeowner andcontractor work it out.
(27:10):
But if the homeowner, for somereason, is I don't know maybe
skeptical or confused, they canalways come to us.
We review, change orders forfree.
We'll always intervene.
You know, on the off chanceyour contractor hasn't called
you back in 24 hours.
You can always reach out to us.
So we're a safety net.
So you know you always havesomebody who has your back
throughout the process.
Speaker 2 (27:32):
And the other thing too with renovations they're
obviously incredibly expensive.
I'm curious has there ever beenthe thought or discussion
around introducing some sort offinancing component to the
company?
Speaker 1 (27:42):
Yeah, really good question.
Very top of mind right now.
For sure, we used to havefinancing partnerships with
various companies when rateswere lower.
We put that on hold justbecause borrowing costs were so
high in the last couple of years.
And we'll actually berelaunching a financing
partnership shortly offeringHELOCs through a third party,
which should help homeowners whoaren't ready to wire five
(28:04):
figures tomorrow.
Speaker 2 (28:05):
Yeah, that makes sense.
I'd like to shift gears alittle bit and talk a little bit
more about you and yourthoughts and philosophies behind
building the company.
So when I was reading the pressrelease, coda mentioned that
you had a really deeply heldgrowth mindset.
So I'm just curious how did youdevelop that and how does that
affect your general leadershipstyle on a daily basis?
Speaker 1 (28:27):
Yeah, those are kind words from him, so I'm really
grateful for that.
Yeah, I think growth mindsetit's a term that's used a lot
and you can interpret it indifferent ways.
What it means to me is a realcommitment to learning, and a
commitment to learning to theextent that you're not afraid of
setbacks or failure.
(28:48):
And for me, I think I'm just sodriven by learning about users,
by learning about what we cando, by learning how we can be
better, that I am prettyenthusiastic about running at
hard, hard problems andexpecting that we're going to
(29:08):
fail many, many times.
That's one of the things thatappealed to me about the
renovation space.
It just seems so damn hard.
The fact that you have a marketthat's almost half a trillion
dollars that nobody has solvedis a massive testament to how
hard it is.
But I find that reallyappealing.
And I think what drives mepersonally, both in my
(29:31):
professional life and mypersonal life, is these problems
that might be impossible ormight just be really hard, and
you don't know until you try.
And when you try you're goingto fail many, many times.
But every time you're failingyou are also learning, because
you're getting signal thatfailure itself is signal and
that might be signal on whatusers need or want.
(29:51):
It might be signal on whatmarketing channels work or don't
.
I think, when Codas' growthmindset how I interpret that is
this perseverance and thiscommitment to learning.
Speaker 2 (30:03):
I love that.
And when we talk about signalfrom failure, because I think
there's two ways that you caninterpret that signal right.
There's this isn't working andwe need to change, or there's we
pick ourselves up and dustourselves off and try again.
So how do you decide whetherit's time to shift gears or just
keep powering through?
Speaker 1 (30:23):
Yeah, that is one of the perennial questions of
building a company, you know.
I think when we think aboutfailure, it's important to
stratify.
What that failure is Is whensomething isn't working.
Is it that we're losingconviction in the problem space?
So, for example, you know, arewe losing conviction that
(30:46):
homeowners need help renovatingtheir homes?
Is it that the current solutionisn't working?
Are we offering a particularproduct feature that's not
resonating, or are we tacklingthis problem space in a method
that's not working?
Or is it distribution?
So, are we maybe pursuing thewrong homeowners or homeowners
(31:09):
in the wrong channel, or are wecommunicating in the wrong way?
And I think, with startups, it'sreally important to have
conviction in and commitment tothe problem space first and
foremost.
And when people think aboutpivoting, pivoting to me means
you go, pursue an entirelydifferent problem space.
You go from renovations tosomething completely different.
(31:31):
If you're changing the featuresets, the actual product, the
solution or your distribution,that's not a pivot.
That is guaranteed.
You know your product and yourdistribution will change many
times.
And so when we think aboutsetbacks and something not
working, I think it's reallyimportant to diagnose what
exactly isn't clicking the waywe expected it to.
Speaker 2 (31:53):
I love that.
Can you tell me a little bitmore about the internal
operations of the company?
How have you built out yourteams?
What does that look like?
Where is everyone located, etcetera?
Speaker 1 (32:03):
Yeah, our team is split 50-50 between on-site in
New York our offices in Brooklynand Dumbo and remote, and
initially the team was entirelyin New York plus engineering a
couple of engineers in Poland.
During COVID we hired morefolks remotely.
Now we're leaning more towardhiring in New York again.
So almost all roles we hire forgoing forward will be in New
(32:25):
York.
Our existing remote employeeswe of course support and we
bring them on site a couple oftimes a year and we have a
really great hybrid culture aswell for most of our teams, who
are hybrid.
But going forward, we do wantto build more and more of a
center of gravity here in NewYork.
Speaker 2 (32:44):
How do you think about, when you know we were
talking about earlier thatgrowth mindset?
How do you screen for that whenit comes to potential new hires
, or just how do you think abouthiring generally?
Speaker 1 (32:52):
Yeah, when it comes to potential new hires or just
how do you think about hiringgenerally?
Yeah, with growth mindset, oneof the things that is really
inherent to our culture ishumility, and you know, we we
talk from time to time about alow ego culture, and I think
that's that's actually part ofthe foundation that Coda built
culturally, and what that meansto me is people who are really
(33:15):
committed to figuring out theright answers for users, for our
homeowners, our contractors,our designers, our architects
and people who care less aboutbeing right.
And that comes back to thisnotion that we're actually going
to be wrong a lot.
We're going to be wrong 90% ofthe time.
What's important is that we'retrying different things and
we're learning.
Every time we have a setback orwe feel like something isn't
(33:38):
going well, we celebrate that.
That's real signal.
The thing to be worried aboutis when you're not getting
signal, when you're taking toolong to build something.
Get it out in front of users.
When you're overthinking amarketing campaign something.
Get it out in front of userswhen you're overthinking a
marketing campaign.
So, when it comes to hiring, welook for people who are really
(34:00):
committed to figuring out theright thing for users and
working collaboratively to dothat together.
Speaker 2 (34:02):
I love that.
And then, as you think aboutthis next phase of scale,
opening it up in all these newcities, building and really
focusing on these AI productshow do you think about doing
that?
From either a scaling from apeople perspective or a tech
perspective, or both?
Do you tend to lean one way,building and really focusing on
these AI products?
How do you think about doingthat?
From either a scaling from apeople perspective or a tech
perspective, or both?
Do you tend to lean one way orthe other?
Speaker 1 (34:19):
Yeah, when it comes to people, we hire on an as
needed basis.
You know, we I think in thisday and age, most startups
especially don't create a hiringplan January 1st and go hire
against it.
Tooling is also changing somuch.
You know you could use AI for alot of things.
It supplements a lot of peopleand we give ourselves the
(34:42):
flexibility to hire throughout ayear, multiple years, as needed
to support our growth.
For example, seo was not a bigpriority for us until recently
and then we ran some experimentswith an outside consultant.
They were fruitful.
We brought on an SEO lead, webrought on writers.
We could do all of that reallyquickly because we had the
flexibility.
We didn't have to say, oh no,we need to wait until our hiring
(35:04):
plan in 2025 to go and kickthis off.
So for us, the most importantthing is to give ourselves the
flexibility to adapt as we learnis to give ourselves the
flexibility to adapt as we learnWith our technology.
The place where we maintain alot of flexibility is in the AI
models we use.
So today we're leaning onvarious foundation models.
That is text interpretation,that's image interpretation,
(35:25):
that's image generation, and thestate of the art changes every
single day.
I mean, I read AI papers everyday because new models come out
every day, and so it's reallyimportant, as you architect a
product that uses different AIcapabilities, is to allow
yourself the flexibility to plugdifferent models in at
different points, and that's abig thing that we're focused on
(35:46):
as well.
Speaker 2 (35:46):
Yeah, I was going to ask you that because obviously,
like you said, stuff is changingevery single day and the
abilities of the foundationmodels change.
So do you anticipate and do youbuild in certain assumptions
around maybe future capabilitiesas you're thinking about
scaling up your own AI products,or what is your overall
relationship with the foundationmodels and how you think of
(36:09):
them affecting the business?
Speaker 1 (36:10):
Yeah, I think in today's day and age, if you're
building any kind of AI product,you need to be super selective
about where you train acompletely proprietary model
from scratch versus where youtake an out-of-the-box model and
fine-tune weights versus whereyou take an out-of-the-box model
and prompt engineer it.
So we're super selective aboutthings like that.
(36:31):
I think it's also reallyimportant to be selective about
where you optimize versus waitfor the technology to catch up.
So one thing we don't solve fortoday is inference cost, so
what we pay to foundation modelsper token to deliver results in
the product.
That's because the cost ofinference is falling constantly
(36:53):
and we're assuming that, giventhe amount of competition in the
market, that's because the costof inference is falling
constantly and we're assumingthat, given the amount of
competition in the market, we'regoing to see that cost continue
to fall.
If at some point it plateaus,we'll have room to optimize, use
fewer tokens here and there bea little bit more efficient.
But we're super selective onokay, we have limited resources
and time and capabilities.
Where do we focus our manpowerto actually optimize and drive
step change results versus?
(37:13):
Where do we focus our manpowerto actually optimize and drive
step change results versus wheredo we expect the technology to
advance on its own?
Speaker 2 (37:20):
Love that.
So a couple of quick firequestions for you.
For one thing, so you're anultra marathon runner, which is,
which is to be honest, crazy.
I can't comprehend.
I I generally like I thinkabout myself, I mean, I'm in
generally pretty decent shape,but when I think about that I'm
(37:41):
like that is just so far beyondanything that I could ever do.
I'm just curious how did youget into that?
What was your?
What's your journey been on onthat side aside from work?
Speaker 1 (37:50):
Yeah, moderation has never been a strength of mine.
I'm either 200% into somethingor 200% out of it.
And you know I ran in highschool.
Our team was not very good.
I jogged a little bit incollege and then my first year
out of college I started workingas a management consultant and
I threw myself into work and Istopped exercising and I, like
(38:15):
would work on the weekends andkind of see friends but do a lot
of work as well.
And at my one year review thefeedback I got was you're
working too hard.
Everybody thinks you're goingto burn out.
Chill out like find a hobby,make it easy.
And you know consulting is aplace where you're expected to
work hard.
So that was surprising feedback, but it came through loud and
(38:36):
clear.
And so I started to run againand I kind of started to run
three miles here and there, fivemiles here and there, and then
on the weekend I would extend myrun longer and longer and
eventually ran a marathon,really enjoyed it.
And marathons, I think, arepretty binary.
You either love it or you hateit.
And you've checked off the listand you swear to never do it
again.
Uh, I, I loved it.
(38:57):
Um, and then I met some ultramarathoners here in New York and
I was like, oh, you can golonger than 26.
Okay, and they just kept goingon longer and longer and um, I
think, uh, the ultra marathoncommunity is a little bit
different.
Like marathoners are reallyintense.
You get a lot of like I don'tknow to stereotype investment
(39:18):
bakers who are working a hundredhours but also running.
Ultra marathoners usually are alittle bit more low key.
You know, you're just likeshuffling around in the woods
for a long time.
You're like chatting andlaughing as you run because
you're going at such a slow pace.
So I love the community and Ilove the adventure and I love
the exploration of what humansare capable of.
(39:39):
It's not only what your bodycan do.
It's so much about what yourmind does.
You just watch your own mind asyou're shuffling in the woods
for hours and hours and yourmind does crazy things.
It's just like you can watchyourself, over the course of a
day, go through really highhighs and really low lows, and
it's really fascinating.
Speaker 2 (40:00):
Next quick fire question what is one book that
you think every leader shouldread?
Speaker 1 (40:06):
That's a good one.
My favorite book in the worldis Thinking Fast and Slow by
Daniel Kahneman, and I used tothink psychology as a field was
just bullshit.
I'd like written it off.
There's too much pop psych andself-help.
But Kahneman, who passed in thelast couple of years, was just
such a tremendous thinker,extremely rigorous, very
(40:27):
epistemically humble, and in hisbook book he really deeply
explores the nature of humancognition and how, how we think,
and he talks about these twosystems system one, which is
impulsive and rapid, and systemtwo, which is kind of more
deliberate and slower, alongwith the tremendous amount of
cognitive bias that exists inthe different types of it.
(40:49):
And I think it's a really goodmap to understand how humans
think, whether it's your ownmind, your, your colleagues,
your team, partners, customers,investors.
I really, really recommend ityou sold me.
Speaker 2 (41:01):
I'm going to go pick it up on Amazon after this.
Speaker 1 (41:03):
A lot of studies, so it's okay.
Speaker 2 (41:07):
No, it's fine, I geek out over that type of stuff.
And then last one what is yourgo-to strategy or tool that you
use to stay just organized andproductive with everything you
have going on?
Speaker 1 (41:16):
Yeah, I think for me it's really helpful to always
have a clear line of sight towhat matters, and a lot of the
time the number one thing Iprioritize is speed of signal.
How can we learn as quickly aspossible, day-to-day,
hour-to-hour?
I feel like our most scarceresource is time and what I try
(41:37):
to maximize for in that time islearning, and what that means is
whether we're building products, whether we're launching
marketing campaigns, whetherwe're selling, trying new things
as quickly as possible and asdecisively as possible to learn.
That comes back to beingcomfortable with failing.
It also comes back to realizingthat most decisions are two-way
(41:58):
doors.
We're not doing brain surgery.
It's okay to take a risk evenif we're not sure, but there's
such an opportunity cost towaiting and kind of waffling and
being indecisive.
So I encourage my teams and tryto really empower my teams to
make decisions.
Get out there, try stuff.
Even if it doesn't work, welearn and that's really valuable
.
Speaker 2 (42:18):
I love that.
Well, julie, thank you so muchfor coming on the show.
I really appreciate all theknowledge you dropped here.
Thanks, rhys.
Thanks for having meconversation insightful and
valuable.
If you enjoyed the episode,please take a moment to
subscribe, leave a review andshare it with your network.
Your support means the worldhelps us continue to grow and
(42:39):
bring more incredible guestsonto the show.
Now for more content andupdates, follow me on LinkedIn
or Twitter, or you can check outMindHire, where we help
startups build exceptional teams.
Thanks again for listening andI'll see you in the next episode
of Seed to Exit.
I think different chapters of a company's lifetime
require different leaders.
Coda did a phenomenal job ofconceptualizing the company,
going after this reallydifficult problem space of
renovations, setting a fantasticculture of humility, excellence
, hard work, collaboration Allthose things will carry forward
(00:23):
In this next chapter.
What matters a lot is buildingout our AI capabilities, so
using artificial intelligence toput world-class tools for
design, planning and hiring inthe hands of homeowners, and
then scaling the company as well.
Speaker 2 (00:41):
Hey, thanks for tuning in.
Today I'm joined by JulieKafitz, the CEO of Block
Renovation.
Block is a company on a missionto transform the home
renovation experience throughtechnology, data and seamless
project management.
Julie brings a wealth ofexperience, having held
leadership roles at Facebook,tractable and the Boston
Consulting Group before steppingin as Block's Chief Operating
(01:01):
Officer and is now leading thecompany as CEO.
So in this conversation, we'regoing to talk about a few things
, including what drew Julie totackling efficiencies in the
home renovation industry, howBlock is solving one of the most
fragmented and complex consumerexperiences, how she's
transitioned from COO to CEO andhow she's starting to make her
mark on the company.
Lessons from raising capitaland scaling a fast-growing
(01:23):
startup.
And the role of AI, data andautomation in shaping the future
of home renovations.
So, whether you're a founder, anoperator or just someone who's
interested in how technology isreshaping old-school industries,
this episode is packed withinsights that you're not going
to want to miss.
No-transcript that power thestartup ecosystem.
(10:46):
I'm your host, rhys Keck,founder of MindHire, a talent
acquisition firm specializing inhelping startups build
exceptional teams.
Each week, I sit down withfounders, investors and industry
leaders to explore the journeysbehind iconic companies and
game-changing ideas.
Whether you're building,investing or just curious about
what it takes to succeed in thestartup world, I want this
(11:08):
podcast to be your go-toresource for actionable insights
and inspiring conversations.
Now, if you enjoy the show,please don't forget to subscribe
, leave a review or share itwith your network.
Your support means the worldand really helps bring more
incredible conversations to life.
Julie, thanks for coming on theshow.
Speaker 1 (11:25):
Thanks for having me, rhys, excited to be here.
Speaker 2 (11:27):
Likewise.
So, before we get intoeverything, you have a super
interesting background.
You've spent some time inconsulting.
You've spent some time at somereally well-known tech companies
like Facebook.
You had a VP role at Tractable.
I'm just curious, now thatyou've ended up at Block, could
you mind just to start thingsoff walking me through your
background a little bit, how yougot to where you're at today
and what drew you to applyingtech to the home renovation
(11:50):
space?
Speaker 1 (11:51):
Yeah, great question.
So I started my career inmanagement consulting at BCG.
I did that for a number ofyears, used that as an
opportunity to see a lot ofdifferent industries and
functions.
I really enjoy learning, so Igot to learn very broadly.
But doing that, what I reallywas missing was actually
(12:11):
operating.
In consulting, you write thestrategy deck, you do a lot of
good thinking about where acompany should go or what it
should do, but you don't do thething.
And I wanted to do the thingand actually take that strategy,
implement it, learn from it,iterate, and so I went into tech
after consulting and I spentsome time at Facebook on the
(12:32):
business operations team, whichwas really interesting.
This was over 10 years ago,starting in 2013, when Facebook
was still in this crazy growthmode and I was embedded in the
revenue organization in ad sales.
So I saw how that was scalingreally quickly, which was really
interesting very culturallycool, but still felt really big.
And at that point Facebook was10,000 people, a fraction of
(12:55):
what it is today.
But even then there was a lotof kind of meetings and decks to
get things done, and so I knewI wanted to go towards something
early stage.
I then spent four years atTractable, the company you
mentioned.
What Tractable does is computervision for accident and
disaster recovery.
So it has an AI platform thattakes photos and videos of
damaged cars and damaged homesand figures out how to repair
(13:16):
them, and that platform is soldto insurance companies, to
restoration companies, repairers, things like that.
And I launched and built outthe North America business for
Tractable, which was reallyinteresting.
So we essentially tried tofigure out what was the product
market fit in North America.
The company was headquarteredin London, had sold a bunch in
(13:36):
Europe and Asia and we had tobring on the first clients,
build out the team, build outthe go-to-market for the United
States and Canada, but not thego-to-market for the United
States and Canada and that.
So I did that from 2018 to 2022.
So before the latest wave of AI, at that point AI was much
harder to build.
You know, before building anauto damage assessment AI, we
(13:57):
had to teach a model how to evenrecognize car parts Like what
is a hood, recognize that thereis a car hood in this photo.
That took months of work, youknow, let alone assessing the
damage to the hood or how torepair it.
So it was a different era of AI, but it was really cool to see
that brought to market and tobring it to market and put it in
the hands of millions of users.
(14:18):
When Block approached me, I justbecame so intrigued by the home
renovation space, in partbecause it is such a massive
market.
Americans spend $470 billionevery year renovating their
homes.
That's not includingmaintenance or small repairs.
That is major renovations, andit seems like almost everybody
(14:39):
still has a nightmare of anexperience that's both
homeowners and contractors andso what really intrigued me was,
you know, just seeing a spacethat was so large and still so
difficult.
There aren't many industriesleft like that.
It also is largely untouched bytechnology.
So there are lead generationplatforms like Angie's List and
(15:02):
Thumbtack, you know.
There are CRMs for contractors.
There isn't something reallythat creates a great experience
for homeowners and contractors,and that's what we're trying to
do with Block.
So what really excited me wasseeing this really difficult
space and trying to figure outhow to create a much better
experience for it withtechnology.
Speaker 2 (15:20):
Yeah, I mean I can 100% agree with you on what you
just described.
I mean, I personally have acouple of rentals and you know I
don't manage them personally,but you know, having to deal
with the contractors or themaintenance repair folks, it's
they're running off of phones,their things are being written
down on paper, it's almostcompletely untouched by
(15:40):
technology and it's it's frankly, it's a really frustrating
experience that I've had severaltimes as a homeowner.
So when I, when I came acrossacross just Block as a company,
I was like oh wow, this issolving a really interesting
problem.
So you came on there a coupleof years ago and talk to me
about what your experience hasbeen like at Block and what
you've done since then.
Speaker 1 (16:01):
Yeah, it's been really great.
We have a really great teamthat deeply understands the
space and is super motivated tocreate something much better.
When I came on, I came on aschief operating officer, so that
was two years ago, at the endof 2022.
And at that point our offering,our model, was still a bit of a
one-stop shop.
So you would come to us as ahomeowner.
(16:23):
We would match you with a greatcontractor from our network for
your build, and then we wouldalso support you with things
like your architectural plans,your permit with the city, your
approval if you're in a co-op ora condo.
We would source, warehouse anddeliver on a single pallet all
of your materials.
So you didn't have to wait forhundreds of deliveries coming
(16:46):
piecemeal.
So it was a reallyoperationally intensive process
and that process and thatoffering made sense for a small
segment of homeowners who didn'thave strong preferences.
They kind of said, okay, I'llpick that template and you guys
handle it for me.
I'm going to go work at mysuper busy job.
But if you had very specifictile or fixtures that you loved
(17:09):
like, let's say, you travel toSpain and you had found some
gorgeous tile, you had shippedover something else it was hard
for us to accommodate because wehad this templated, streamlined
process, and so what weactually did over the last
couple of years is evolve theoffering.
So we stopped requiringhomeowners to use our materials,
follow our architectural plans.
We leaned in much more to themarketplace side of the platform
(17:31):
, which meant helping homeownersto discover the right
contractors for them, getproposals, evaluate those
proposals, negotiate andtransact on the platform.
We also started to build out awhole suite of planning tools.
On the platform.
We also started to build out awhole suite of planning tools.
So all the thinking, the workthat happens before you hire a
contractor, just trying tofigure out what you're going to
(17:52):
do with your home, how muchit'll cost, what it'll look like
.
We support that too.
Speaker 2 (17:55):
Got it.
So you recently took on the CEOrole, just as recently as a few
weeks ago, so congratulations.
So the founder was Coda, so hedecided to step back.
He's still serving on the board.
Talk to me a little bit moreabout what it was like as you
started to have thoseconversations, how you thought
(18:15):
about it and just really how itall came to be, how you thought
about it and just really how itall came to be.
Speaker 1 (18:20):
Yeah, you know, leadership transitions can be
complex, especially at startups,which often don't have the same
reps as large companies do.
Transitioning CEO to CEO.
Kodai did a really fantasticjob of thinking about this
pretty far in advance and givingus the time to plan it and
execute it, and so we were justreally deliberate about how to
(18:43):
do this transition, how to havethe right conversations with our
investors and our board, withour team, how to set up the
leadership team around me tosupport this transition.
He is still really involved asexecutive chair on the board,
just less in the day-to-day, andthe fact that we've been
working side-by-side as CEO andCOO for the last two years that
helped a lot as well.
Speaker 2 (19:04):
I can imagine that, as you step into the CEO role,
you want to make an impact, youwant to put your proverbial
stamp on things, but then also,at the same time, you want to be
respectful of the legacy andwhat has been built so far.
How do you navigate, balancingthose two things?
Speaker 1 (19:20):
Yeah, I think different chapters of a
company's lifetime requiredifferent leaders.
Coda did a phenomenal job ofconceptualizing the company,
going after this reallydifficult problem space of
renovations, setting a fantasticculture of humility, excellence
(19:40):
, hard work, collaboration allthose things will carry forward
In this next chapter.
What matters a lot is buildingout our AI capabilities, so
using artificial intelligence toput world-class tools for
design, planning and hiring inthe hands of homeowners, and
then scaling the company as well, and those are things where I
(20:01):
bring the right track record,and so I think there's natural
complementarity in thefoundation that Coda has built
and the momentum that I caninject for this next chapter.
Speaker 2 (20:11):
When you talk about scaling, what does that look
like in terms of the vision ofwhat you want to do next?
Speaker 1 (20:16):
Yeah, we started to expand geographically a lot more
quickly, more aggressively thisyear.
So initially Block launchedjust in New York City,
especially with our much moreoperationally complex 1.0
original model.
It was hard to go and bringthat model elsewhere.
We'd need to open up warehouses, put up supply chains.
So in the first six, sevenyears we opened up only a couple
(20:41):
of more cities.
Just three months ago wealready opened up Chicago,
philadelphia, connecticut.
We will continue to expand cityby city across the country,
probably every six months or so.
So that's a big part of it.
The other part is just handlingmore of the renovation journey.
So historically we've focusedso much on helping homeowners to
(21:01):
hire contractors and to executetheir projects.
But that's the last 10% of thejourney.
The average American renovatestheir home about every three
years and 90% of that threeyears is just trying to figure
out what to do.
You know you wake up and youlook at your kitchen.
You're like this is kind ofoutdated, I don't know the right
counter space, I just wish itcould be right.
(21:24):
It just doesn't feel right.
But going from it doesn't feelright to actually knowing what
to do and how much it'll costand who to hire, that is what we
call planning, and that'susually a multi-year process.
So more tooling around that.
Putting more data athomeowners' fingertips.
So more tooling around that.
Putting more data athomeowners' fingertips, helping
them to figure out design scope,budget iteratively on their own
schedule, without having tohire an architect.
(21:46):
That'll cost them $20,000.
Speaker 2 (21:50):
That's where we're focused on the product side as
well.
Why does it take years?
Is it just a matter of nottaking action and knowing it's
just one of those things thatneed to be worked on, but you
just haven't bothered to do it.
Speaker 1 (22:02):
That's such a great question.
I think it's largely because itfeels overwhelming.
The thing we see over and overagain is that a lot of
homeowners just feel stuck.
They know they want to changetheir space, but when it comes
to figuring out what to do orhow much to spend, or even where
(22:22):
to start should I do my kitchenfirst or my bath first they
Google around, they ask around,but there's no reliable source
of information and it feels likesuch a consequential set of
decisions that people just getoverwhelmed and they stop.
And then they start again.
They start to look at Pinterestagain and Instagram again.
They're like oh, I really likethat.
But then, going from I reallylike that to actually investing
(22:45):
the capital, trying to figureout do I need to move my family
to a hotel, you know, can Iactually live through the
renovation?
It just seems so overwhelmingand there are hundreds of
decisions you have to make.
So what we're focused on is howdo we empower homeowners, put
the right tools and data intheir hands and the right
support in their hands so theyfeel empowered, and so they
(23:07):
don't feel alone.
Speaker 2 (23:07):
So what is building that process look like from an
AI perspective?
I'm just curious.
I'd imagine it's multimodal.
I'm sure there's a chatbotbuilt in.
I'm sure you can upload imagesand say hey, here's the space.
What is?
What are the AI capabilitiesthat you're actually building?
Speaker 1 (23:21):
Yeah, yeah, I love talking about this because
there's so much you can do withAI.
Today.
There's a lot of AI hype, whichI'm always aware of.
You know.
One thing we observed with therenovation planning specifically
is people really struggle toarticulate what they need and
what they want, but they alsohave very strong reactions to it
.
So if I show you a space, a lotof people, if they've been
(23:43):
thinking about renovating, willinstantly say, oh, that feels
right, or oh, no, that's not forme.
You know, think of it like asimple bathroom, five by eight
feet.
I can one person.
You show them white, with whiteaccents, and they're like, oh,
that's so sterile, that's notfor me, and they want something
moody and dark.
You know blues and greens andblacks.
(24:03):
And you show somebody else thatmoody and dark, and they're
like, oh, that's not right atall.
I want something clean, modern,industrial.
And what we figured out is thereneeds to be an accessible,
delightful, easy way to showhomeowners different things that
are possible, to allow them tonarrow down what they really
have in mind.
(24:24):
And in a way, that's whatarchitects and designers do.
They show you mock-ups andyou're like, oh, I like that one
, but maybe not the cabinets.
Can we switch that out?
And you iterate.
So what we're building is aniterative process that gives you
different stimuli or examplesto react to and allows you,
through your instincts, tonarrow down.
The key, though, is doing thatin a way that feels personalized
(24:45):
to you, so I can show yougeneric bathrooms or kitchens or
spaces, but you might say, well, that one is twice as big as my
bathroom.
That's clearly not right for me.
That won't fit here.
Or that one has a window.
I don't have a window, so itneeds to feel sufficiently
relevant and personalized to youand your space.
And the hard part is that everysingle home is different.
Unlike new construction, whereyou have a piece of land that's
(25:09):
almost a blank slate withrenovation, you have a home
that's been there probably fordecades, that probably had work
done on it, permitted,unpermitted.
You might open up a wall, findstuff that you didn't know was
there, and so we need to accountfor that in kind of giving
people this information andthese options and these choices.
Speaker 2 (25:28):
Yeah, I was going to ask that actually because, again
, having lived through a coupleof renovations myself, you get
in there and you find somethingthat you didn't think was there.
There's always surprises, right, or scope changes, so how does
the platform and team adjust andadapt to that?
Speaker 1 (25:46):
Yeah, yeah.
That's one of the great joys ofrenovation you know there are
always surprises because, again,every home is different.
So we do a couple of things.
One is upfront.
We try to share as muchinformation as possible.
So when homeowners getproposals from contractors, they
see things like it might takethis long, to this long, like
(26:07):
these are the kinds of things wedon't know yet, that we will
see once we actually open upwalls and see what's there.
So I think information upfrontis something that's often
missing in the industry,especially for first-time
homeowners.
So often first-time homeownerskind of figure out their vision
and maybe they use onlineplatforms like Pinterest or
(26:27):
Instagram and they just seethese beautiful aftershots and
they think, oh, it's going to bea smooth journey, I just need a
trustworthy contractor to buildthat.
But your home is a constructionsite for a number of weeks and
there will be surprises.
So sharing information upfrontis really important.
The other thing that we do isalways provide a safety net, so
(26:47):
homeowners and contractors worktogether to get through these
surprises.
Our contractors are equipped todo that, but homeowners always
know.
If something is really strange,we're there to help.
So if there is a change order,your contractor comes to you and
says we opened up a wall.
There's stuff we didn't expectto see there.
That's going to cost another$10,000.
Usually the homeowner andcontractor work it out.
(27:10):
But if the homeowner, for somereason, is I don't know maybe
skeptical or confused, they canalways come to us.
We review, change orders forfree.
We'll always intervene.
You know, on the off chanceyour contractor hasn't called
you back in 24 hours.
You can always reach out to us.
So we're a safety net.
So you know you always havesomebody who has your back
throughout the process.
Speaker 2 (27:32):
And the other thing too with renovations they're
obviously incredibly expensive.
I'm curious has there ever beenthe thought or discussion
around introducing some sort offinancing component to the
company?
Speaker 1 (27:42):
Yeah, really good question.
Very top of mind right now.
For sure, we used to havefinancing partnerships with
various companies when rateswere lower.
We put that on hold justbecause borrowing costs were so
high in the last couple of years.
And we'll actually berelaunching a financing
partnership shortly offeringHELOCs through a third party,
which should help homeowners whoaren't ready to wire five
(28:04):
figures tomorrow.
Speaker 2 (28:05):
Yeah, that makes sense.
I'd like to shift gears alittle bit and talk a little bit
more about you and yourthoughts and philosophies behind
building the company.
So when I was reading the pressrelease, coda mentioned that
you had a really deeply heldgrowth mindset.
So I'm just curious how did youdevelop that and how does that
affect your general leadershipstyle on a daily basis?
Speaker 1 (28:27):
Yeah, those are kind words from him, so I'm really
grateful for that.
Yeah, I think growth mindsetit's a term that's used a lot
and you can interpret it indifferent ways.
What it means to me is a realcommitment to learning, and a
commitment to learning to theextent that you're not afraid of
setbacks or failure.
(28:48):
And for me, I think I'm just sodriven by learning about users,
by learning about what we cando, by learning how we can be
better, that I am prettyenthusiastic about running at
hard, hard problems andexpecting that we're going to
(29:08):
fail many, many times.
That's one of the things thatappealed to me about the
renovation space.
It just seems so damn hard.
The fact that you have a marketthat's almost half a trillion
dollars that nobody has solvedis a massive testament to how
hard it is.
But I find that reallyappealing.
And I think what drives mepersonally, both in my
(29:31):
professional life and mypersonal life, is these problems
that might be impossible ormight just be really hard, and
you don't know until you try.
And when you try you're goingto fail many, many times.
But every time you're failingyou are also learning, because
you're getting signal thatfailure itself is signal and
that might be signal on whatusers need or want.
(29:51):
It might be signal on whatmarketing channels work or don't
.
I think, when Codas' growthmindset how I interpret that is
this perseverance and thiscommitment to learning.
Speaker 2 (30:03):
I love that.
And when we talk about signalfrom failure, because I think
there's two ways that you caninterpret that signal right.
There's this isn't working andwe need to change, or there's we
pick ourselves up and dustourselves off and try again.
So how do you decide whetherit's time to shift gears or just
keep powering through?
Speaker 1 (30:23):
Yeah, that is one of the perennial questions of
building a company, you know.
I think when we think aboutfailure, it's important to
stratify.
What that failure is Is whensomething isn't working.
Is it that we're losingconviction in the problem space?
So, for example, you know, arewe losing conviction that
(30:46):
homeowners need help renovatingtheir homes?
Is it that the current solutionisn't working?
Are we offering a particularproduct feature that's not
resonating, or are we tacklingthis problem space in a method
that's not working?
Or is it distribution?
So, are we maybe pursuing thewrong homeowners or homeowners
(31:09):
in the wrong channel, or are wecommunicating in the wrong way?
And I think, with startups, it'sreally important to have
conviction in and commitment tothe problem space first and
foremost.
And when people think aboutpivoting, pivoting to me means
you go, pursue an entirelydifferent problem space.
You go from renovations tosomething completely different.
(31:31):
If you're changing the featuresets, the actual product, the
solution or your distribution,that's not a pivot.
That is guaranteed.
You know your product and yourdistribution will change many
times.
And so when we think aboutsetbacks and something not
working, I think it's reallyimportant to diagnose what
exactly isn't clicking the waywe expected it to.
Speaker 2 (31:53):
I love that.
Can you tell me a little bitmore about the internal
operations of the company?
How have you built out yourteams?
What does that look like?
Where is everyone located, etcetera?
Speaker 1 (32:03):
Yeah, our team is split 50-50 between on-site in
New York our offices in Brooklynand Dumbo and remote, and
initially the team was entirelyin New York plus engineering a
couple of engineers in Poland.
During COVID we hired morefolks remotely.
Now we're leaning more towardhiring in New York again.
So almost all roles we hire forgoing forward will be in New
(32:25):
York.
Our existing remote employeeswe of course support and we
bring them on site a couple oftimes a year and we have a
really great hybrid culture aswell for most of our teams, who
are hybrid.
But going forward, we do wantto build more and more of a
center of gravity here in NewYork.
Speaker 2 (32:44):
How do you think about, when you know we were
talking about earlier thatgrowth mindset?
How do you screen for that whenit comes to potential new hires
, or just how do you think abouthiring generally?
Speaker 1 (32:52):
Yeah, when it comes to potential new hires or just
how do you think about hiringgenerally?
Yeah, with growth mindset, oneof the things that is really
inherent to our culture ishumility, and you know, we we
talk from time to time about alow ego culture, and I think
that's that's actually part ofthe foundation that Coda built
culturally, and what that meansto me is people who are really
(33:15):
committed to figuring out theright answers for users, for our
homeowners, our contractors,our designers, our architects
and people who care less aboutbeing right.
And that comes back to thisnotion that we're actually going
to be wrong a lot.
We're going to be wrong 90% ofthe time.
What's important is that we'retrying different things and
we're learning.
Every time we have a setback orwe feel like something isn't
(33:38):
going well, we celebrate that.
That's real signal.
The thing to be worried aboutis when you're not getting
signal, when you're taking toolong to build something.
Get it out in front of users.
When you're overthinking amarketing campaign something.
Get it out in front of userswhen you're overthinking a
marketing campaign.
So, when it comes to hiring, welook for people who are really
(34:00):
committed to figuring out theright thing for users and
working collaboratively to dothat together.
Speaker 2 (34:02):
I love that.
And then, as you think aboutthis next phase of scale,
opening it up in all these newcities, building and really
focusing on these AI productshow do you think about doing
that?
From either a scaling from apeople perspective or a tech
perspective, or both?
Do you tend to lean one way,building and really focusing on
these AI products?
How do you think about doingthat?
From either a scaling from apeople perspective or a tech
perspective, or both?
Do you tend to lean one way orthe other?
Speaker 1 (34:19):
Yeah, when it comes to people, we hire on an as
needed basis.
You know, we I think in thisday and age, most startups
especially don't create a hiringplan January 1st and go hire
against it.
Tooling is also changing somuch.
You know you could use AI for alot of things.
It supplements a lot of peopleand we give ourselves the
(34:42):
flexibility to hire throughout ayear, multiple years, as needed
to support our growth.
For example, seo was not a bigpriority for us until recently
and then we ran some experimentswith an outside consultant.
They were fruitful.
We brought on an SEO lead, webrought on writers.
We could do all of that reallyquickly because we had the
flexibility.
We didn't have to say, oh no,we need to wait until our hiring
(35:04):
plan in 2025 to go and kickthis off.
So for us, the most importantthing is to give ourselves the
flexibility to adapt as we learnis to give ourselves the
flexibility to adapt as we learnWith our technology.
The place where we maintain alot of flexibility is in the AI
models we use.
So today we're leaning onvarious foundation models.
That is text interpretation,that's image interpretation,
(35:25):
that's image generation, and thestate of the art changes every
single day.
I mean, I read AI papers everyday because new models come out
every day, and so it's reallyimportant, as you architect a
product that uses different AIcapabilities, is to allow
yourself the flexibility to plugdifferent models in at
different points, and that's abig thing that we're focused on
(35:46):
as well.
Speaker 2 (35:46):
Yeah, I was going to ask you that because obviously,
like you said, stuff is changingevery single day and the
abilities of the foundationmodels change.
So do you anticipate and do youbuild in certain assumptions
around maybe future capabilitiesas you're thinking about
scaling up your own AI products,or what is your overall
relationship with the foundationmodels and how you think of
(36:09):
them affecting the business?
Speaker 1 (36:10):
Yeah, I think in today's day and age, if you're
building any kind of AI product,you need to be super selective
about where you train acompletely proprietary model
from scratch versus where youtake an out-of-the-box model and
fine-tune weights versus whereyou take an out-of-the-box model
and prompt engineer it.
So we're super selective aboutthings like that.
(36:31):
I think it's also reallyimportant to be selective about
where you optimize versus waitfor the technology to catch up.
So one thing we don't solve fortoday is inference cost, so
what we pay to foundation modelsper token to deliver results in
the product.
That's because the cost ofinference is falling constantly
(36:53):
and we're assuming that, giventhe amount of competition in the
market, that's because the costof inference is falling
constantly and we're assumingthat, given the amount of
competition in the market, we'regoing to see that cost continue
to fall.
If at some point it plateaus,we'll have room to optimize, use
fewer tokens here and there bea little bit more efficient.
But we're super selective onokay, we have limited resources
and time and capabilities.
Where do we focus our manpowerto actually optimize and drive
step change results versus?
(37:13):
Where do we focus our manpowerto actually optimize and drive
step change results versus wheredo we expect the technology to
advance on its own?
Speaker 2 (37:20):
Love that.
So a couple of quick firequestions for you.
For one thing, so you're anultra marathon runner, which is,
which is to be honest, crazy.
I can't comprehend.
I I generally like I thinkabout myself, I mean, I'm in
generally pretty decent shape,but when I think about that I'm
(37:41):
like that is just so far beyondanything that I could ever do.
I'm just curious how did youget into that?
What was your?
What's your journey been on onthat side aside from work?
Speaker 1 (37:50):
Yeah, moderation has never been a strength of mine.
I'm either 200% into somethingor 200% out of it.
And you know I ran in highschool.
Our team was not very good.
I jogged a little bit incollege and then my first year
out of college I started workingas a management consultant and
I threw myself into work and Istopped exercising and I, like
(38:15):
would work on the weekends andkind of see friends but do a lot
of work as well.
And at my one year review thefeedback I got was you're
working too hard.
Everybody thinks you're goingto burn out.
Chill out like find a hobby,make it easy.
And you know consulting is aplace where you're expected to
work hard.
So that was surprising feedback, but it came through loud and
(38:36):
clear.
And so I started to run againand I kind of started to run
three miles here and there, fivemiles here and there, and then
on the weekend I would extend myrun longer and longer and
eventually ran a marathon,really enjoyed it.
And marathons, I think, arepretty binary.
You either love it or you hateit.
And you've checked off the listand you swear to never do it
again.
Uh, I, I loved it.
(38:57):
Um, and then I met some ultramarathoners here in New York and
I was like, oh, you can golonger than 26.
Okay, and they just kept goingon longer and longer and um, I
think, uh, the ultra marathoncommunity is a little bit
different.
Like marathoners are reallyintense.
You get a lot of like I don'tknow to stereotype investment
(39:18):
bakers who are working a hundredhours but also running.
Ultra marathoners usually are alittle bit more low key.
You know, you're just likeshuffling around in the woods
for a long time.
You're like chatting andlaughing as you run because
you're going at such a slow pace.
So I love the community and Ilove the adventure and I love
the exploration of what humansare capable of.
(39:39):
It's not only what your bodycan do.
It's so much about what yourmind does.
You just watch your own mind asyou're shuffling in the woods
for hours and hours and yourmind does crazy things.
It's just like you can watchyourself, over the course of a
day, go through really highhighs and really low lows, and
it's really fascinating.
Speaker 2 (40:00):
Next quick fire question what is one book that
you think every leader shouldread?
Speaker 1 (40:06):
That's a good one.
My favorite book in the worldis Thinking Fast and Slow by
Daniel Kahneman, and I used tothink psychology as a field was
just bullshit.
I'd like written it off.
There's too much pop psych andself-help.
But Kahneman, who passed in thelast couple of years, was just
such a tremendous thinker,extremely rigorous, very
(40:27):
epistemically humble, and in hisbook book he really deeply
explores the nature of humancognition and how, how we think,
and he talks about these twosystems system one, which is
impulsive and rapid, and systemtwo, which is kind of more
deliberate and slower, alongwith the tremendous amount of
cognitive bias that exists inthe different types of it.
(40:49):
And I think it's a really goodmap to understand how humans
think, whether it's your ownmind, your, your colleagues,
your team, partners, customers,investors.
I really, really recommend ityou sold me.
Speaker 2 (41:01):
I'm going to go pick it up on Amazon after this.
Speaker 1 (41:03):
A lot of studies, so it's okay.
Speaker 2 (41:07):
No, it's fine, I geek out over that type of stuff.
And then last one what is yourgo-to strategy or tool that you
use to stay just organized andproductive with everything you
have going on?
Speaker 1 (41:16):
Yeah, I think for me it's really helpful to always
have a clear line of sight towhat matters, and a lot of the
time the number one thing Iprioritize is speed of signal.
How can we learn as quickly aspossible, day-to-day,
hour-to-hour?
I feel like our most scarceresource is time and what I try
(41:37):
to maximize for in that time islearning, and what that means is
whether we're building products, whether we're launching
marketing campaigns, whetherwe're selling, trying new things
as quickly as possible and asdecisively as possible to learn.
That comes back to beingcomfortable with failing.
It also comes back to realizingthat most decisions are two-way
(41:58):
doors.
We're not doing brain surgery.
It's okay to take a risk evenif we're not sure, but there's
such an opportunity cost towaiting and kind of waffling and
being indecisive.
So I encourage my teams and tryto really empower my teams to
make decisions.
Get out there, try stuff.
Even if it doesn't work, welearn and that's really valuable
.
Speaker 2 (42:18):
I love that.
Well, julie, thank you so muchfor coming on the show.
I really appreciate all theknowledge you dropped here.
Thanks, rhys.
Thanks for having meconversation insightful and
valuable.
If you enjoyed the episode,please take a moment to
subscribe, leave a review andshare it with your network.
Your support means the worldhelps us continue to grow and
(42:39):
bring more incredible guestsonto the show.
Now for more content andupdates, follow me on LinkedIn
or Twitter, or you can check outMindHire, where we help
startups build exceptional teams.
Thanks again for listening andI'll see you in the next episode
of Seed to Exit.
Popular Podcasts
True Crime Tonight
If you eat, sleep, and breathe true crime, TRUE CRIME TONIGHT is serving up your nightly fix. Five nights a week, KT STUDIOS & iHEART RADIO invite listeners to pull up a seat for an unfiltered look at the biggest cases making headlines, celebrity scandals, and the trials everyone is watching. With a mix of expert analysis, hot takes, and listener call-ins, TRUE CRIME TONIGHT goes beyond the headlines to uncover the twists, turns, and unanswered questions that keep us all obsessed—because, at TRUE CRIME TONIGHT, there’s a seat for everyone. Whether breaking down crime scene forensics, scrutinizing serial killers, or debating the most binge-worthy true crime docs, True Crime Tonight is the fresh, fast-paced, and slightly addictive home for true crime lovers.
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
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.