Monetizing Open Source with Peer Richelsen (Cal.com)

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Peer Richelsen (00:00):
You see this landing page of an open source
library and it says used byFacebook, Netflix, Amazon, right
, and then you scroll down andyou see donations by Steve and
Anna and you're like what?
Like how is Steve and Annadonating and Facebook and Amazon
and Meta and everyone's usingthis for free and no one donates

(00:23):
back To me.
That's the opposite of of thestrategy of open source.
It's almost like a socialwelfare system where the top one
percent of customers should payfor the bottom 99 and if you
can figure out a way to do that,now you have a free product.
People love it, they startusing it, they share it with
others, they share it withenterprise companies, enterprise

(00:44):
companies keep paying.
Free product gets better and aslong as you can get this cycle
going, you're always in a bettershape than trying to monetize
the bottom 99%, which is reallyhard.

Ariel Camus (01:01):
Welcome everyone.
This is Ariel Camus, and youare listening to the Only Thing
that Matters the podcast where Iinterview successful founders

(01:22):
to deconstruct their path toproduct market fit and uncover
the principles and frameworksbehind their success.
My guest today is PeerRichelsen, the co-founder and
CEO of Cal.
com.
Peer has raised over $30million from Cal.
com and has built an open-sourcescheduling platform for
individuals, teams and builders,offering a more flexible

(01:42):
solution than traditionalcalendar tools.
In this episode, we dive intothe key strategies for building
and monetizing a successful opensource business leveraging a
community of eager contributorsto achieve product market fit,
embracing the open startupmovement.
And the principles of buildingand managing a distributed

(02:04):
global team.
We also discussed his strongopinions on offering global
salaries.
I hope you enjoy thisconversation a lot With all of
Peer Richelsen .
Hello Peer, it's so good tohave you here.
Thank you so much for joiningus today.
We got to know each other, likeduring the COVID times,

(02:25):
supporting each other throughthose, you know, difficult
moments of adapting our work,our companies, our way of
approaching work, and a lot haschanged since then and I look
forward to catching up with youtoday on everything that has
happened with you as anentrepreneur and your company,
happened with you as anentrepreneur and your company.

(02:47):
Um, let's start with that, like, can you give me a little bit
of?
Uh, yeah, what is cal?
Um, if you, when you explain itto people, how do you explain
what it is?

yeah, yeah, first off.
Uh, you were hosting officehours doing covet and everyone
was like anxious as, and so you,you kind of like ease the pain,
I guess, of COVID and pandemic,so thank you for that.
Ariel was hosting foundermeetups online, which is fun.
What is Cal?
I would say what Cal isn't.

(03:17):
It's not a calendar, as may itsound like.
We're not a cron alternative orVim Cal or Amy, we don't do any
calendar UI.
I'm incredibly happy we're notdoing that because I think it's
there's so much microinteractions with a calendar and
people think, oh, it's just acalendar, but like it is a very,

(03:40):
very long and strong, like longroadmap to get to a point where
people are like, okay, this isnice.
So we only do the schedulingpart and we do the scheduling
for individuals, for teams andfor builders.
And we just recently launchedthe builders part where, if you

(04:01):
want to build anything with Cal.
com, you want to build amarketplace, you want to build a
SaaS tool that has a schedulingcomponent, a telehealth
platform where users are bookingusers.
It's very different to all theother scheduling tools that want
to connect my calendar with mycustomers.

(04:22):
So that's called Cal.
com platform and that's a moredeveloper focused tool where you
get like a set of apis andfront-end sdks and ui kits.
So really we're doing both theconsumer facing kind of
scheduling tool where you as anindividual sign up and it makes
it super easy.

(04:42):
The same way I joined your linkto to book this podcast, um, so
that's like one one part of ourbusiness.
And then recently last uh, lastmonday we launched a developer
focus one perfect and and itmakes a lot of sense.
I think the approach you took tolaunching first the b2c side,
the consumer side, um, the waythat I think about it, at least

(05:04):
in the current product that I'mbuilding, is it allows you to
just have to satisfy onestakeholder and then, once
that's incredibly good, then youcan move to the B2B side of
things, for example, and they'realso more homogeneous, right
Like, the type of customerreaching out that signs up for a
consumer-facing product is ageneric person who wants to do

(05:25):
meetings right, it's verydifferent than a developer who
has 5, 10, 20 different ideas ofwhere they want to build this
in Like, the platform producthas inbound from anywhere from,
as I said, like, healthcare tohiring marketplaces or ADHD

(05:46):
coaches or group violencesessions.
Right Like it's such adifferent market.
Each customer, each lead issuch a different beast than
reaching out to every founder todo their fundraising with Cal.
com.
Right, like, that's a littlebit easier to start with in the

(06:06):
beginning.

Makes total sense and I'll ask the question that I
know you're well prepared toanswer, and it's how is this
different from Calendly?
Right, and I know that isprobably part of the origin
story of the company.
I would love to hear how it isdifferent and how did cal come
to to be what it is today yeah,no, I mean hands down.

I was a calendar user for many years, which is
also why I'm still very, verymuch interested in this industry
.
I think it's very hard to justlook at the market and start to
do something that someone elsedoes If you don't know the
background, like the contextright, like you need to become
an industry expert.
And when we met, I was buildingleanhirecom I don't know if you

(06:53):
remember and that was amarketplace to connect remote,
actually founders for initiallyfounders only with remote
companies, only with remotecompanies.
And I always keep telling thestory how I used Calendly as a
scheduling infrastructure tomake the meeting happen.
Right, so very easy product youwould.

(07:14):
During the onboarding flowwhere you create your account,
there was a string input thatsounds like copy paste your
Calendly link and then, if wefind a company, we just forward
that link to that company in theemail.
So very easy stuff, so verysimple prototype.
But as we got more traffic andmore people, we realized that A,

(07:34):
we don't see if a bookinghappens because it's a black box
, right, you just send the linkand they don't report it back.
And second, people may not havea Calendly account.
They copy it and paste it witha typo.
They forgot to upgrade theiraccount, so their link is a 404
and needs an upgrade.
And others are just straight upnot willing to pay for it or

(07:59):
they don't want to use it, butthey still want to use your
product.
And so after a while I waslooking into the market and to
me, having been in the industryfor so many years, I know that
open source solves that issue.
Whenever you want somethingmore custom that you can
integrate into your productthat's more sustainable and

(08:19):
long-term and you can makechanges and make you really feel
like it's your product.
I always look for open sourceprojects and alternatives or
repositories and I went onGoogle and asked for, like
Calendly, open source, likethat's literally it, and I was
hoping to find a popular GitHubrepository and just start

(08:40):
cloning it and couldn't find it.
And the same way, I couldn'tfind it.
There was many posts on HappenNews and Reddit of other
entrepreneurs or developersasking the same question, right
and so if you're an entrepreneurand you see other entrepreneurs
struggling to find a solutionthat they really need, like

(09:01):
these threads have like hundredsof messages like no, I don't
know a solution, but let me knowif you find one.
Right, like if you find thattype of product and you have
that pain yourself, I thinkthat's like basically a light
beam shining on you telling youto like you should probably

(09:21):
start looking into thisdirection.
So I just started an opensource repository, really with
no intent to commercialize it.

I still was running lean, higher as a business, um,
and then eventually that githubrepository just gained more
traction than the initialbusiness how, how did, how did
you validate or how did youdecide that something that was
an open source repository withno monetization behind it, with,

(09:52):
I guess, uncertainty on well,to a certain extent, uncertainty
on the willingness to pay?
I guess there was somecertainty because people were
paying for it candidly andthat's cool.
But maybe the type of peoplelooking for an open-source
alternative had lowerwillingness to pay, not just
need for more versatility.
So what did you do to say, okay, let's check if this could be a

(10:16):
business?

Well, it's good if people reach out to you and
ask you for like hey, how can Ipay you?
That's always a good sign.
Did that happen?
Yeah, I mean we've had likeafter launch, by the way.
So we launched on Product HuntProduct went absolutely crazy.
Product of the day, big oneyeah, like 2,000 something

(10:40):
upvotes, probably one of themost upvoted products by now, I
agree, something upvotesprobably one of the most upvotes
upvoted products by now.

I agree that that's really um, I see products of
the day going with like 500,sometimes like like 91 000 votes
and, yeah, like 2 200 200 yeah,and so that is.

That is a sell shock right to me as a founder,
to my co-founder who you knowactually worked on on 90 of that
prototype at bailey, um, andbut also that makes a.
That has a reason why it's sopopular, right, because people
are craving and looking for it,um, and so, yeah, we from day
one we even had like largercorporations reaching out.

(11:18):
Um, none of those deals likeconverted in the early stage
because your product is justvery horrible and shit, but some
of them we converted a year ortwo years after.
But when you build somethingthat has just a sell-shop moment
and we talk about this in YCalso a lot of finding product
market fit is like you know itwhen you have it and you can't

(11:40):
really describe it otherwise.
And product market fittypically also comes with a
commercial market fit typically.

Yeah, when, when you did the launch, I think that
was like April 2020.

2021, I think 21 .
Yeah 21.
You were in the Thomas.
Now, right, that's it.

Yeah, yeah, your anniversary Actually, yes, very
close.

Which day?

of April, was it, do you?

remember.
Yeah, let me open my product.
I'm not sure we shouldcelebrate.
That's funny.

It happens to me all the time as an entrepreneur,
we I know there's so manyanniversaries Exactly, and we're
so focused on the future andthe next milestone and we forget
to celebrate the past.
But we also fail, I think, veryoften to celebrate the, the
little victories of the momentbecause we're always chasing
something else, and it's such abig um lesson for life, right

(12:34):
like the importance ofcelebrating every step, because
I don't know, at least to me,there is no guarantee,
especially as an entrepreneur,that there is a big destination
out there.

So you better enjoy the journey, otherwise
you're setting things up forfailure 100 absolutely um so
where do I see that 15th ofapril, wow, okay, so really on
the day two days ago again yeah,and we did.
We did launch on the 15th ofapril again.
So, wow, that's nice.
I mean we always try to launchon the 15th, but I didn't know

(13:04):
it was actually also the sameday, so oh, pretty cool oh yeah,
so you, you recently had the4.0 launch and okay, I, I get, I
hadn't paid attention to thespecific dates.
That's a cool one, um if you goback in time, every major
release is on the 15th, even theminor ones.
So we launch every month on the15th, even the minor ones.

So we launch every month on the 15th.
It reminds me to the way thatApple used to do releases.
I think having a date thatforces you, that rallies the
entire team to make somethingamazing happen, I think it's a
very powerful thing.

It's the best thing we ever came up with.
And, to be honest, we actuallyand I got a little bit of heat
by some engineers, but, like wealso do our versioning as
Calvary I don't know if you'veheard of it.
So, semba is, when you have abreaking change, you make a
major update.
So you like inform people thatyou know if you go from 3.0, 4.0
is a breaking change.
I never liked that because Idon't think the average person

(14:01):
cares, nor you know, sometimesyou ship something and you don't
know it's a major breakingchange.
And so there's many things whereI'd rather make sure we don't
have any breaking changes everand we've been very backwards,
compatible in the past and thenreally focus on have a yearly

(14:21):
cadence.
So if I look back, if I seesomeone out in the wild using
Calicom 3.0, I know, okay,that's a year ago, right, it's
much easier than 3.74.56659.
I'm like, is that a breakingchange?
I don't know.
And so, and from a marketingpoint of view, you just make it
so much easier to communicate,like where's the state of the

(14:42):
product?
And you know, after 12 samefirefox is like version number
27 or something like.
You just know if a product ismature and I, I, if we tell
people every year we launch amajor release and and it's at
5.0, they know the product'sfive years in the making, so
it's like roughly five years um.

So we?
We spend so much time oftenfocused on the features, on the
technical side and we forgetthat we have to sell right.
And sometimes very simplemessages even if in their
simplicity is where the key isright, and just not complicating
the versions, but also usingthem in your advantage to be
able to sell.
It makes total sense.
Let's talk about selling.

(15:20):
You mentioned that often withproduct market fit comes some
kind of a commercial success aswell.
At which point, before or afteryou launched officially Product
Hunt, the first version, youstarted seeing revenue.
How were those early days ofactually starting to charge for
the product?

Yeah, we actually made revenue before we
gave people access to theproduct.
It was very interesting.
It was a time where we hadlaunched the website, we had
launched a prototype, but it wasbarely functional really, and
we almost oversold our vision onthe landing page of more of a

(16:03):
promise market fit than aproduct market fit, where we
know we could build these things.
We just haven't gotten the timeyet to to ship these, and so we
had kind of like these, um, uh,angel donation almost, where
it's like, um, like a, apre-sale race, really right,
like a pre-sale of the product.
And so, I my surprise, we'vedone like a decent amount of

(16:25):
money, like a decent amount ofrevenue.
Before we had certain thingsready, which is great because
the revenue graph goes from, youknow, zero to something rather
quick.
And the second thing we did iswe started with a paid offering
only.
So when we released the beta,after like two, three weeks

(16:46):
after product, and we we had itpaid only, which means you, in
order to use the product forfree, you would need to self
host.
So it's always an option, evenup to this day.
It's an.
It's an option with a easydocker file.
But, um, we wanted to reallywork with people who care, and

(17:09):
that's the one who pay you right, they care and they give higher
quality feedback.
They stay around longer theythey are, they want to be early,
they want to be you know, userid number 37, right and and brag
about it.
And so we had this really smallniche, maybe like a couple

(17:31):
hundred people that really lovedthe product, and then we just
closely worked with them.
We invited them into our Slackchannel.
We had this big monolith Slackchannel, so we had a community
mixed with our core team.
We had no separation.

These were contributors to the codebase, or
some of them were justsuggesting.

Customers, contributors, fans.

They didn't need to pay for that.
They just needed to becommitted to making the product
better.

Yeah, yeah, and so one interesting metric was
just to see how the communitygrows.
Because if you're an opensource, you want to make sure
that you have a good developerand community ecosystem, and so
having everyone in Slackdecreased the amount of friction
to talk to people.

(18:22):
You have a general channelwhere everyone's asking
questions and then if someonehas an issue, I can make a new
private group with an engineerfrom our core team with that
customer right.
So it's really easy.
Immediately stopped, likeimmediately started breaking.
Once we had like 3000 people inthe channel, because now it's
like the wild west and it's likeit's really horrendous for

(18:45):
larger communities.
Um, I do like the idea fromlike zero to 200 people.
Um, it's, it just feels magicto like have everyone in one
channel and kind of like beworking on this together.
But, um, after a certain stage,we just move to discord.
Um, frankly speaking, I don'teven like discord nowadays.

(19:05):
I think it's a hot mess ofnotifications.
I, I don't think it's, and itdoesn't help with seo, it
doesn't help with searchability.
People ask, keep asking, thesame questions over and over
again without doing any research.
I, rather I might be moving offof discord in the next couple
of weeks or months.

Makes sense, but in this initial zero-to-one stage,
being able to talk to yourcustomers as YC, pounds over and
over on us.
It's so, so, so important, andyou can do that because it's a
small little volume of people.
But do you think this can alsowork for non-open source
companies?

(19:45):
How will you do it differently,if any difference, if you are
not an open source company?

Well, I mean, I think there are examples of
large communities that are notopen source.
I think Raycast comes to mymind.
They are like a bit of a hybridwhere all of their app store
extensions are open source.
Um, because they they they hadthe correct analysis that that,
um, github is the best place tobuild apps on, like integrations

(20:15):
.
Um, the core product, raycast,is not open source, so I would
not put them in a open sourcebucket, but at least they have a
very, very active community andecosystem.

Because I guess the commonality is not open source
but you have people buildingstuff for the product and I
think that allows them to havethis very closed, tight-knit
community where you can talk topeople very closely.

Yeah, and what they've found out is by
empowering people to make itsuper easy to build apps.
The same way, google andAndroid and iOS try to make it
easy for you to build apps.
I mean, that's everyone'sincentive really to make it as
easy as possible.
So I think they found a goodway to have their marketplace on

(21:00):
the web while still owning,being the gatekeeper of what
goes in and what not.
I mean, ideally I'd love tohave Raycast fully open source,
but I don't think that's goingto happen.
Other communities yeah, I meanany company that is community
driven in the first place, right, like companies like On Deck,

(21:20):
maybe companies like yours.
You probably want to have acommunity.

Yc is, at the end of the day, a corporation, a VC
firm, absolutely and for thoseof you that might not have gone
through YC and I would love tohear your experience, pierre.
But like having access toBookface, the private community
for YC founders it's crazy thevalue that I get from that, like
almost every day of my life.

(21:47):
It's insane, and I think it'sthe only email that I open
religiously is the daily digestof what's happening there.

What's going on?
I don't know if your experienceis similar.
Same.
Yeah, it is after all theseyears and all these new people
and people complain about ourbatch W19, that it was too large
.
It was the largest batch, Ithink, up until then, and they
were like, oh, what if thisreduces the quality.

But I think YC has always been it's just more
quality people, really, it'smore of good, and so, uh,
quality hasn't really decreasedever and that's really hard for
a community to grow it's reallyhard to do and I guess you got a
taste of the opposite as youscale into getting people that
probably show up to the discordserver just to get support and

(22:40):
not because they're invested inmaking the person better, and
spam and scams and insaneexpectations to an open source
repository, like they demand youto reply within five minutes
because they didn't install npmbefore and they run.

They run npm and it doesn't work.
And you're just like bro, likecome on, like I have to do stuff
, you know, yeah, there arebetter places to ask about it or
figure that out.

But uh, yeah, it's hard.
It's hard unless you like spenda lot of time and I guess there
is um probably a disproportionalamount of value that you can
get from something like that inthe very early days.
Because, you know, I thinkthere is something really
important about, uh, your story,the story of cal, is that it

(23:28):
was a need that you had rightand that gives you insights into
the what makes the productsuccessful.
That it's much harder to get ifyou're building something for a
target audience that you're notpart of.
But even when you are part ofyour target audience, there is
always variability and I thinkbeing able to talk to them in a
very direct way, um, it's superimportant.
I I've had to go through thepainful process of having to

(23:50):
like try to schedule calls withpeople for interviews, um in in
maybe moments where I didn'tspend as much time building that
community, and it's so muchharder to engage people and you
can try to offer, you know,money, vouchers, discount stuff,
but it feels so different whenyou had invested in the
community and they just want toreciprocate the value they're

(24:13):
getting from being part of thecommunity.
But just spending time with youand telling you stuff, like, I
think, if you can generate that,that's amazing and I can
clearly see that as one of theadvantages of open source and I
would love to get a little bitdeeper into open source.
I think that's somethingfascinating that you bring into
the world of how to buildcompanies.
I think probably for a lot ofpeople, and especially like for

(24:35):
some entrepreneurs, they mightthink wait, but if I give away,
you know my code, the wholething, how can I really make
money with this?
And I wanted to read a quotefrom a quote that you wrote.
I wouldn't be surprised.
And it said like going head tohead as a SaaS company against
existing market leaders is afool's errand.

(24:55):
Hence we are doing somethingsimilar.
We are doing similar things ina fundamentally different
category open scheduling andthere is a concept like the open
source, the open sort of that Iwould love to go a little bit
deeper into with you.
We started talking about it,but in your experiences, what
are the, the, the advantages andthe disadvantages?
Why going open source?

yeah, um, so I actually I don't know know why I
removed my blog on my personalpage, but I used to have a blog
post that said like to opensource or not to open source?
And I had, like this what isthis again from Shakespeare, the
posture.
And I don't agree with the ideaof just looking at Cal.

(25:42):
com and applying everything wedo to an entire different
product, slash industry, becauseit's not like, let's pretend
open source is a technology,right, whenever a new technology
comes around.
There's two types ofentrepreneurs, right, the ones
that have an existing problemand now they're like oh, what if

(26:04):
this technology can solve this?
Right?
And then there's the other typeof entrepreneur and I don't
want to be rude, but most ofthem are less successful with
whatever they're doing, and theytake a look at a technology and
they and they try to come upwith a problem like what can I
solve with this technology?
And so you and once you this iscourtly, you know like as a, as

(26:29):
a hammer, everything's a nailand um, and you start to let's
say, oh, I like I get dms ofpeople saying I want to start an
open source company, but Idon't know what to work on.
I'm like that's not how youapproach anything.
That's the same problem youhave with crypto, the same
problem you have with AI.
It's like I want to build an AIcompany but I don't know what

(26:50):
to work on any ideas.
I'm like you're clearly lookingat the wrong picture and the
thing always goes back to theproblem, right, like I had a
problem and I could only solveit with open source.
If you have the same problemthat you can only work, solve
with open source, you're in agreat business and so when and

(27:13):
also if that's a great market,it's a big market and it's
probably also commercial marketthen you're in a great place.
The projects that I find hard tomonetize and even open source
is kind of famous for lack offunding, which I partly disagree
with but there are projectsthat are just really really hard

(27:34):
to monetize.
A programming language is veryhard to monetize because you and
me, we would never pay forReact, we would never pay for
JavaScript, we would never payfor Nextjs, we would never pay
for a more advanced router.
Developers fucking suck when itcomes to paying.
They make six figures, butpaying five bucks a month for

(27:57):
React to use them soundsridiculous.
For good reasons you wouldn forReact to use them sounds
ridiculous, for good reasons,right?
You wouldn't pay to use English.
It's just such a core thing,fundamental thing, and so I
think if you head out and youbuild a better React, you better
know what you're doing.
I think Tailwind is an amazingexample how they've probably

(28:19):
made a million dollar businessby now because they understood
very early on we can nevermonetize the fundamentals.
We need to build on top ofthese fundamentals and package
them up and sell them to peoplewho really care.
And you know, tailwind UI isall of their pre-built

(28:39):
components for a one-time fee.
Maybe they have a subscriptionby now, but to me that's clever.
And so there are ways tomonetize these more fundamental,
lower level things.
I'm pretty sure if Facebookcared, they could find a
business model for React.
I'm happy they don't.
Please don't commercializeReact, really.

(29:01):
But so long story short, aslong as you know what problem
you're solving, I think you willbe able to monetize it, and as
long as you have an addedservice or added product on top
that uses your fundamentals, Ithink you're in a good place.

Why will someone pay for using CAL instead of
just using the free open sourceversion like self-hosted?

Well, that probably goes back to every
self-hosted product.
Running software haslimitations and issues and it's
time consuming keeping things upto date.
Secure your own servers I mean,most developers nowadays don't
even run their own serversanymore Like we use the cloud or
we use serverless or we usewhatever comes.

(29:55):
Whatever you know gets us tosuccess.
I think a lot of peopleunderestimate the saving power
of self-hosting.
I typically say self-hosting ismore expensive.
Even the enterprise license wesell for self-hosting.
So Calicom is an open core.

(30:17):
We have a 95-ish percent opensource github repo and then
there is, I'd say, five percentof the code base is enterprise
only and these are like auditlogs and and fcim and and all of
that like workspacing and stuff.
You as a hobby engineer or evenas a small startup or agency,

(30:38):
never need you.
You will never need theseproducts.
You wouldn't even know theyexist really.
So these 5% we monetize.
We require you to have alicense, but you can have it on
the SaaS plan for free not forfree without setup or you can
self-host, but the self-hostinglicense is actually kind of
pricey.
So if people reach out to usand say, hey, I want to save

(31:01):
money, but I want yourenterprise license, we typically
tell them like, I don't thinkyou're the right customer and so
self-hosting is never really away to save money really ever.
It's more about being in chargeof your own data, making
changes, changing every line ofcode, being on your own upgrade
schedule.

(31:21):
Some people don't want toupgrade.
That's perfectly fine, as longas they patch the
vulnerabilities, security issues, and so I think, instead of
taking open source as somethingthat's cheaper than other
companies, you change theperspective on like this is more
than other companies give youso if let let me try to

(31:45):
summarize this to see if I if Iget it right what you're saying
is that by making it open source, you get way more people to
contribute to making the productawesome.

You have this community of highly engaged
users that you can learn from.
That can help improve theproduct.
Then for many of them, in theearly stages, it's expensive to
run their own servers or to evenhost the open source version on
the cloud, because you have tomaintain that and that takes
focus away from your corebusiness, which is not

(32:19):
scheduling.
However, as the companies growand maybe they do have, you know
, a larger team, they can afford, like you know, and then it's
compliance, so it's actually aneed.

At that point you say, okay, we are open
source, but we will reserve, uh,strategically, a series of
features, uh, that you will haveto pay a license for, so that
we don't get the largest clients, who are the ones that can pay
the most, actually getting thisfor free, completely correct yes
, 100 and um, if you look, ifyou take a look back at the, the

(32:55):
dark side of open source thishas happened so many times where
, um, you see this landing pageof an open source library that
does something, converts a to b,and it says used by facebook,
netflix, amazon, right, and thenyou scroll down and you see
donations by steve and anna andyou're like what, like how Steve

(33:20):
and Anna donated and, andFacebook and Amazon and Meta and
everyone's using this for freeand no one donates back.
There's no commercialengagement whatsoever.
Um, to me, that's that's theopposite of of the strategy of
open source.
To me, I think the top onepercent is really like.
It's almost like it's like asocial welfare system where the

(33:44):
top 1% of customers should payfor the bottom 99%.
And if you can figure out a wayto do that, not only do you
have a amazing free product,right, like, our free product
will always be more of more highquality than any one of our
competitors, because we havethis big, strong enterprise
pipeline that just subsidizeseverything.
Right.
And it's a vicious circlebecause now you have a free

(34:06):
product, people love it, theystart using it, they share it
with others, they share it withenterprise companies.
Enterprise companies keeppaying free, product gets better
and as long as you can get thisvicious cycle cycle going, I
think you're always in a bettershape than trying to monetize
the bottom 99, which is reallyhard makes a lot of sense.

Otherwise you're like failing to capture most of
the of the lessons there rightand have you had any case or
maybe do you know of anyone thatthat did this wrong?
And by trying to charge theyend up, like you know, getting
their community mad anddestroying that precious
community.
All of all the time, of course.

yeah, I mean, um , I think, uh, I I big fan of
docker and its founder.
The founder was our yc partnerand solomon and they had and
they had a similar issue wherethey went after consumers and
enterprise and I thinkeventually I don't remember
fully, maybe I'm wrong, maybeyou need to correct me in the
show notes, but I do think thatthey sold part of the enterprise

(35:12):
business to another company todo all of that stuff and then
they doubled down on theconsumer facing and they tried
to charge a monthly fee fordevelopers.
And every developer was like,well, fuck you.
And again, that's the perfectexample why you would probably
not pay for React per month touse React, because for them,

(35:34):
docker is just a scriptinglanguage to run containers.
They don't understand what itmeans to you, know what the
value is behind it.
So, yeah, that definitelyhappened.
I don't know if Docker wentback.
I think they went back andfocused on enterprise again.
To me, that just makes so muchmore sense because every
engineer eventually works at alarger corporation anyway.

(35:54):
So like, if you can win theengineer, you win the enterprise
.
Like that's just how it works.
There's other examples there'stons of open source companies
that change the license and tryto bully you into whatever
getting a license.
I'm really not a fan of thesedelayed open source licenses

(36:17):
that have popped up.
You know where they say it's acommercial license, but in two
years uh, versions two yearsolder are open source.
Like me, as a maintainer, Inever want my community or any
customer to be on a versionthat's two years ago.
Two years ago we launched liketwo years ago is fucking dark

(36:38):
shit.
Like if we had this license, wewould not have been open source
up until this point.
And and then now people areallowed to use the two years old
code base that's full of bugs,full of vulnerabilities, full of
issues.
They would spam my discordchannel about bugs that we've
already fixed.

(36:59):
Like I don't know who came upwith that type of concept, but
you're basically saying like abig fuck you to the whole
community saying likeeverything's commercial, you can
still contribute.
And then in two years, you knowwhen everything's, when we're
already, you know, far away inpuppy land, you can use the
leftover crumbs, I suspect thatis a solution to a different

(37:23):
problem.

I'm speaking out of intuition here, but that once
you are far ahead but you havecustomers that don't want to
migrate to newer versions forwhatever reason, and you can't
afford or don't want to, likeyou know, bother maintaining the
older versions or like keepingthe backwards compatibility, you
have to say, okay, sorry Ican't help you, but at the least

(37:45):
I can open source this.
I look good on paper and youget to maintain it for me.
And who maintains this will?
also help me with the otherclients that have this old
version.
Um, otherwise I like it soundslike a really bad way of
creating an open source companyif you start just with that
version.

None of these companies, in my opinion, should
call themselves open sourcecompanies, because they're not
open source first.
They are open source second,and that's also fine, but I
think you want clarity.
Also, how do you even keeptrack of what two years has gone
?
Do you need to check in?
I need to wait one more weekand then something gets open

(38:24):
source the feature that I needthat was shipped in one week,
like, come on, like, nobodykeeps track of that.
Yeah, my, my philosophy has beenum, everything's, even the
commercial parts, are sourceavailable, right, like, you can
look at them, you can copy them,you can change them For a
commercial use.
You need a license.
Now what happens?
We have like a little bit of aDRM check that like pings one of

(38:46):
our endpoints and sees if youhave a license key inside your
environment variables and ifit's a valid one, it returns
true.
If it returns true, yourcomponent gets unlocked and you
can use it.
So, obviously, any engineerthat was born in 2000 or in the
90s knows how to remove drm.
That's like the easiest stuffever.
You just remove that function.

(39:07):
Um, and we specifically havecomments about, like, if you
remove this without a license,you're violating the license.
It's a copyright issue, yada,yada, yada.
Of all these three years.
This has happened once that II'm aware of, and I reached out
to the founder.
Um, the founder wasn't evenaware.
They hired a third-partycontractor who clearly didn't

(39:28):
give a shit, um, and theyrectified rectify that issue.
And um, my point of view is,the types of people who violate
your license are the ones whowill never pay for your product
in the first place, and so ifyou give it to them for free,
you just need to be okay from anego point of view.
People are using yourcommercial stuff, but don't be

(39:51):
offended by the fact that you'remaking less revenue, because
you would not make that revenuein the first place right, you're
not losing.

You are not losing business, you're not losing
anything.

I have never seen a well never say never.
But it's very, very rare that alarge corporation such as the
companies we sell to actually gointo the code base, remove drm,
open them up to millions oflitigations, potentially, like
that engineer gets fucking firedif the cfo or this, the, the gc
, finds out about that.

(40:23):
Absolutely so you're in a hotmess and it's way easier for you
to ask procurement to get alicense key than to spend an
entire day removing drm anddoing all of this illegal,
literally copyright illegalstuff.
Um, and so I never woke up andbe like oh what if Amazon takes
out our DRM?
I don't think that's ever goingto happen really no, no, no.

It doesn't make sense for those large companies
to do something like that.
I completely agree, and I thinkthe good news for anyone who is
considering starting open sourceis that you have now a lot of
like companies.
I can I can think right now,like github, for example, having
you know publicly tradedcompany that you know is has an

(41:08):
open source model from thebeginning, super successful, um,
but I'm curious about and I'venever had to go through this
myself.
I'm curious if you have somekind of like framework.
Maybe we can come up with ittogether.
For how do you make thedecision?
Like you said before, you don'tstart open source and then
figure out what to build that isopen source.
You find the problem and thenyou decide does open source make

(41:31):
sense for this product and thissolution?
How do you make that decision?
How do you like evaluate, forexample, will I make more money
by being open source and bybeing, you know, not non-open
source?
Like will I be more successfulby any definition like?
How do you make that thatdecision?

yeah, again, I think it really boils down back
to your, your industry, um.
We and knock on wood were veryfortunate that we're working on
a product that a lot ofenterprise and highly regulated

(42:09):
industries and and um, justlarger corporations like
desperately need um, and sowe've always had a good sales um
habit and muscle of sellinglike six-figure contracts early
on um and and and so, and theseare deals that we are winning
because we're open source right.

Why is that?
Why open source allows you towin contracts.

For compliance, for compliance, for they need it
on-prem.
They want to self-host.
They also want to look throughthe code.
They have their own independentcontractors that they hire for
code audits and securityvulnerabilities.
And open source Even Elon Muskhas said it publicly in an
interview is known to be themost secure technology because

(42:48):
you have a million developers.
I don't know how many times ourREADME has been opened, but it
must be over a million times andwe have like 500 contributors
and a dozen third-partyresearchers that try to find
holes in our code base.
It's one way to reverseengineer an API and try to find

(43:09):
vulnerabilities, but if youactually see how people have
written code, it's so mucheasier to be like okay, there's
a check missing, I can get intothe system.
So after three years, we've hadno breach of anything.
We even have people who startedimplementing a more advanced

(43:32):
encryption authorization systemthan before.
We've had a communitycontributor adding 2FA factor
login just because he was like,hey, I think this could be fun.
And I'm like, yeah, let's do it.
And so you know, when I say opensource makes better products, I
firmly stand by it, becausethere's just so many things that

(43:55):
an engineering team can buildand all of the rest typically
gets thrown into a roadmap thatyou will never look at again.
The classic backlog problem.
When the backlog is is publicand and you may even start
incentivizing certain thingswith bounties, with algoraio and
some other companies andproducts, you start to see where

(44:17):
you know the the Pareto of like20, 80 80 20 rule.
Right, like your team works onthe 80 percent of the most
important things, but the last20 percent will never be done.
Well, now you have a communitythat does the last 20 percent,
and these are small PRs, bugfixes, vulnerabilities, even
small features that are notreally of high priority, but

(44:39):
it's.
It's nice to have them for abunch of people, and so I would
say open source builds thebetter products, but also the
more secure ones as a sideeffect.

Really, I want to go back to the, this idea of
like being able to work on stuffthat you wouldn't work
otherwise.
I'll get back to that in amoment.
I want to maybe recap orsummarize something you said
before so that it's clear.
Maybe there is more that youwant to add to this.
I asked you like why willsomeone decide to open source?
And there are a lot of benefits, right, and but one you said

(45:11):
makes a lot of sense alsobecause connected to the
business side of things is thatif you have companies that are
highly regulated, uh, or, youknow, massive in size, which
also are have companies that arehighly regulated or massive in
size, which also are thecompanies that can also afford
paying the most, they will bemore likely to need, for
compliance reasons and also forauditing processes, to have the
self-hosted version, and opensource facilitates or actually

(45:35):
makes that possible, makes thatpossible right.
So that is a great, I think,way to think about it, because
it's aligned with the type ofcustomer, enterprise, customer
that will allow you to capture alot of the business value, so
it's terrific.
Is there any like principle orindicator that you should not go

(45:56):
open source?
Like going open source will bea big mistake.

Yeah, I'd say there are a couple.
I think to really make abusiness work in open source.
It's a bit of a I wrote aboutthis at Cal.
com slash about as well.
When you're, let's say, whenthe SaaS business captures 50%

(46:23):
of revenue, the open sourceversion captures 1% of revenue.
I said this before 1% pays forthe 99%.
If your market is too small toand you multiply your market by
1% and you come up with a totalrevenue of, I don't know, 500K
or a million, it's really hardto go beyond that Because you

(46:45):
reach your total market capreally quickly if you only
monetize 1%.
I mean, I don't know how muchthe total market cap of Tailwind
CSS is.
It's every JavaScript developerreally, every developer, web
developer, really.
So that's a huge time, it's ahuge market and I don't think
they have 1%, even Probably ahalf a percent maybe.

(47:07):
So be aware that when you buildsomething in the open source,
it cannot be an open sourcemarketplace to buy and sell fish
.
It's not a market where you canmonetize 1% of enterprise sales
.
There's no enterprise companythat sells fish as a marketplace

(47:27):
that needs a highly customizedopen source marketplace to buy
and sell fish, right.
So, of course, just an example.
But be aware, when you chooseto be open source, try to find
the largest market of yourinterest, because it's going to
be hard to monetize Not hard,but you will probably disregard

(47:50):
99% of the market to makerevenue.
And also, if you don't have a,we actually introduced a free
plan after we felt comfortablewith our product and comfortable
with our support, because youalso want a free SaaS tier
because otherwise thealternatives people are

(48:10):
self-hosting and you don't wantto cannibalize your own traffic.
Really you want to make it thesame offering.
People have self-hosted.
You should have hosted just forease of use, and I don't like
it when people say it's paidonly on hosted and then it's
like here's the red lead to hostyourself and no one ever does
that really.

(48:32):
So my recommendation is if youare interested in open source,
well, first you've got to have areally really big problem that
you can only solve with opensource.
Best um examples are emails.
Everyone uses email, um, maybevideo calls maybe you want to
build the largest videoconferencing tool?
Um, maybe you want to build umthe next uh, spotify, I don't

(48:57):
know, doesn't make much sense tobe open source, but maybe with
AI you have a new way ofgenerating.
Maybe it's an open sourceSpotify that's powered by AI,
where you have open prompts andpeople can share their prompts
publicly and you can make musictogether and stuff like that.
So at any given time, Cal.
c om is very different to all ofits pre-processors.

(49:20):
Right, it's, it's a successor,it's different, it's, it's, it's
um, it's better in my opinion.
So if you build something opensource, have a big market and
try to be better, you have to bebetter.
If you're just, if you're justa free, open source alternative
to something, you'll always be afree, open source alternative
that fucking sucks.
You won't raise funding, youwon't hire the best team, you

(49:43):
will never build the bestproduct and I think that's kind
of like the open source 1.0.
That's just been a like.
Compare LibreDocs with MicrosoftDocs or Microsoft Office it's a
clear loss.
Or Thunderbird Mail knock onwood has been great back in the

(50:04):
days, but it's just not a goodclient if you compare
Thunderbird with Superhuman oreven Gmail, to be honest.
So you kind of really want todouble down on building the best
product, regardless whetheryou're open source or not, and
by building an enlarged industry, by capturing 1% of the revenue
, you end up product, regardlesswhether you're open source or
not, and by by building in alarge industry, by capturing one
percent of the revenue, you youend up building a good business
and that leads you to to buildthe best open source project

(50:27):
awesome thanks for sharing allof that and it's interesting,
and probably somehow connected,that uh, you're not just an open
source company but also an openstartup and you have this like
uh, you have a lot of data aboutthe company publicly available
in the website.

How did you get to that?
And, by the way, I'll make itwell anyway.
No, go ahead.
I would love to hear yourperspective on this.

How do we start?
Yeah, pretty much day onereally.
Me and my co-founder, we sattogether we said, well, we're
already sharing code.
Of course, it's hugely inspiredby other companies.
Uh, open startup is nothingwe've invented.
There's a.
It was actually more popular inthe indie hacker bootstrap
space, which is kind ofinteresting because all of them

(51:13):
went back to like removing theiropen pages.
If you take a look at all thefamous bootstrappers on Twitter
or Xcom, they all, kind of likeremoved their data.
I've always been a big fan.
It made me feel closer to thecompany.
I felt more invested.
If you're into sports, you careabout all the statistics of how

(51:33):
many defense a person had, howmany scores, how many assists.
Right, like, that's justinteresting.
If you're into sports, you wantto know everything about your
team.
To me, having an open sourceproject, you're already sharing
code.
You're already sharing yourissue backlog, your
communications, your Slack.

(51:54):
Everything's open, everything'spublic.
It feels more alien to then keepstuff for yourself, right?
Especially when it comes toshared success.
Right, you want to showcaseyour community?
Hey, this is how it's going.
Here's where all your energy.
You know like, remember peoplewho are working on open source.

(52:14):
They do it on a weekend,because they really want to work
with you.
They really want to improve andimprove their own product.
So what's better than goingback to them and telling them
hey guys, hey, everyone, listenwe've grown 40% month over month
.
Here's all the data.
I think there's nothing moreexciting about that than

(52:36):
contributing to a project butalso seeing its success in real
time.
And then the open startup, opensalary policy that's something,
I'd say, that is even morecontroversial than sharing
numbers.
That is also something wehaven't invented.
I mean, nothing's reallyinvented in 2024.
Some of it is still with AI, soBuffer and some other companies

(53:02):
were sharing the salaries for along time.
I've always been a huge fan oftransparency within companies.
We could have easily just madethis internally right, and that
would have been already adramatic step for most companies
.

That's what we've done for like five, six years.
And it's a huge game changerinternally.
Appspot I think it's anothercompany that does that, but they
have an interesting philosophyon why not sharing salaries
particularly, and I think theprinciple is we can share
everything except stuff that isnot just ours to share, and if I

(53:38):
share someone else's salary,I'm sharing information that is
not just ours to share, and if Ishare someone else's salary,
I'm sharing information that isabout the person and it can have
implications Like I want tomake this up, you know, like
family members or friends,knowing they make a lot of money
going to them and asking

for something and I and I always found it very
interesting To interrupt youhere We've never forced people
to do that.
We, we've clearly said it is anopt-in policy.
Uh, there's no peer pressure.
Even if you feel like it,there's no negative connotations
if you don't do it, nobodyreally cares like within the
team.
But if you, if you want to bepart of this movement and you're

(54:15):
fine with it.
I think so far we've only hadthree people who did not share
their salary for likeundisclosed reasons.

Whatever reason, yeah, yeah, exactly.
Do you still share itinternally in those three cases?
No, no, no, not even internally.
Fair enough, yeah.

Okay, and so it's.
I mean, the manager knows whohired them, right?
Yes, of course, but no, I thinkit's people and I got a lot of
negativity from investors orother founders that said like,
oh yeah, this is cute in theearly days, but this will not
scale with you.
That doesn't even go into thewhole global policy, global

(54:56):
salary policy, which is evenmore fringe and controversial in
my opinion.
So what I've learned over thepast three years is global
policy, global salaries, publicglobal salaries, um, or, the
best way to keep fairness.
It's the best way to fightracism, gender, gender um,

(55:17):
what's it called?
Uh, pay gaps, wage gender, paygaps is that what it's called?
yeah, yeah, um discriminationamong whatever, yeah, any, just
it.
Really it's impossible for meto hire a junior engineer at
this rate and another juniorengineer at another grade.
Um, but there's no way I can beracist to one or the other.

(55:42):
Uh, because it's all the same.
Like for you to make more money, you need to become a
mid-engineer, senior engineerand, uh, lead engineer
eventually.
Um, it's, it's more fair.
When it comes to work equalsmoney.
You know, we all have thisconcept of like if I work, I
make money, if I work more, Imake more money.

(56:03):
If I work better, I make moremoney.
How is it that remote companiesand it?
I understand that not everysalary works in every country,
but to me, the fact that ajunior, fresh software engineer
who's written the first line ofReact comes straight out of uni
in San Francisco gets aninternship for a hundred grand,
and then you have a seniorengineer in India making maybe

(56:28):
75 or 50, half of the revenue,and you all come together in a
retreat in Portugal, on bothsides of the world, and one
person makes twice your salary,just because they were born in a
high expense state.
Cost of living yeah, as amanager, I should not care what

(56:49):
your cost of living is If youcannot make it work, don't apply
for my company.
And like this is where this is.
This is all public.
Like sure, we discriminateagainst people in the Bay Area,
but also, truth be told, there'swork outside of the Bay Area,
that's fantastic, and that is.

That is because the benchmark salary use is too low
for the Bay Area, but it'sprobably very high in almost
every other place.
Exactly which market do you useas a reference?

We kind of have a combined, mostly because me
and my co-founder are spendingtime between New York and Europe
, so we kind of took a New Yorkand Europe salary and like took
the average really.

So it's like between in on the ocean kind of,
and so but you use your the twofounders kind of reality to
then set the benchmark foreveryone else in the team.
Yeah, yeah, okay.

Okay, okay, perfect, and we also use pave to
like make sure it's notoutrageously low, outrageously
high.
But, truth be told, probably 60to 80 percent of our team make
way more than their market rateand what does that mean?
You attract the highest qualityof people in those markets,
right?
Like, compensation is a keyfactor for incentivizing the

(58:09):
team, but also hiring qualitypeople is a key factor for
incentivizing the team, but alsohiring quality people.
If I have the choice, let's sayyou take 150K or maybe 300K
total comm, including equity.
If you get a junior engineer inthe Bay Area and you get the
most senior, diehard engineer inanother country, like, who do

(58:30):
you hire as a manager?
You probably hire the bestperson, right?
Um, and so that's been, that'sbeen something where, uh, we've
had just amazing success andalso from an ethics point of
view, um, that person the bayarea will find a job, that they
will always be fine, they, theywill find a job, they will find
find an internship, they arefine.

(58:52):
And also, even if you end uphiring them at like 250, they
are like six months away fromgetting a better offer from
Netflix.
So it's just not an interestingmarket for me, and same thing
probably also true for New Yorka little bit different, but that

(59:23):
person you hire at 150, that'sfour times the salary of the
local market is incredibly loyal, incredibly incentivized,
incredibly senior, incrediblysmart, incredibly interested in
your company, and that's thetalent we want to attract.

really, one of the common arguments that I've heard
, for example, from the founderof GitLab, against having this
global salary is that it goesagainst and I think this is a
very subjective thing and Iwould love to hear your thoughts
on this business purely as abusiness because you are

(59:51):
overpaying on one side, meaningthat that's money that you
cannot spend on hiring otherpeople that could allow the
company to grow faster, getbigger, capture more business
value, and so it's alsodetrimental for everyone in the
company because at the end ofthe day, especially if you're
giving equity to people, peoplewill end up owning a chunk that

(01:00:13):
is much smaller.
Like how do you rationalize, uh, your decision against an
argument?

Peer Richelsen (01:00:20):
do they also give less equity to people in
cheaper, in low-cost countries?
I don't think so um, wow, okaybut then, then, what the fuck is
that even?
What does that even mean?
Like you get compensated foryour work, right, and that's the
principle of work.
Like if, if, um, if you work ingermany and you hire a job, you

(01:00:42):
get hired for a job, your nextperson should earn the same
money, right, like yeah, I think, by the way, it's job, right it
is.

Ariel Camus (01:00:51):
is it's the business side versus the
principles, right?
No, I understand.

But to build a business is just a combination
of people you hire, really Likeit's just a team of people you
hire, and if you try to see yourteam from a cost analysis, not
from an impact analysis, I thinkyou're already in the wrong
boat.
Just be honest For me, to savemoney, I hire one person less.

(01:01:24):
It doesn't make sense.
Let's say you have 10 people.
They all make 100 grand justfor math Simplicity.
To cut everyone's salary by 10%.
It's much easier, grand, justfor for math simplicity, to to
cut everyone's salary by 10.
It's much easier to just letone person go right.
So much easier.
Like get, I think my futurecompany reduces the amount of of

(01:01:47):
c players.
That that's the one thatdragged down the company, not
the mental model of like.
Oh uh, you know that that yousaid of like it drags everyone
down.
It drags everyone'scompensation down.
No, get rid of bad performersfirst and have more room for
high performers.
And you know, I don't thinkthis global policy speaks

(01:02:08):
against high performers.

Really, if anything , you attract high performance
in outside of the bay arealocations no, I agree with that,
but I think the argument is notso much that you attract bad
performers, but then you,because you can hire fewer
people, you end up with, uh,suboptimal capacity to grow the
business, which means why?

would you be able to hire less people, Like
if I have a global salary thatis below SF market?
I have friends who run in-houseSF companies with four
engineers and they spend amillion dollars a month in costs
.

And I'm like why would you do that?
And I think that is theunmeasurable upside of building
businesses outside of the BayArea right.
In today's world.
If you adopt a more remote,friendly policy, little policy,
you're going to attract amazingpeople who are around the world
without having to compete inthis crazy market in the Bay
Area.
Of course there are otherdisadvantages of not being in

(01:03:05):
the Bay Area.
I've lived there for like sixyears and there are advantages,
right.

It's amazing to be there?
Yeah, absolutely, but it'samazing for founders.

Let's be honest, it's amazing for founders to be
there.
Yes, it's a horrible place tohire people.

But also to live .
Listen if I hire an engineer,look at our team.
People live in Spain, italy,argentina or some of the most
beautiful places.
We also have a bit of a nomadstigma to us but, like people,
are able to choose in ourcompany where they want to be

(01:03:44):
without a pay cut.
If someone moves from and thishappened many times I think
we've had nine people in totalthat have moved countries many
times.
I think we've had nine peoplein total that have moved
countries from London to NewYork, from San Francisco to
Manila, you know all the way onthe other side of the cost
spectrum.
I've moved five times.
Since then.
We've had a team member whowent from India to Dubai, which

(01:04:05):
is a really, really starkcontrast.
And all of this because we havea steady global salary income.
It empowers people to be incharge of their life, and life
also means where you live.
Right?
If we had a localized salaryfor India, we would always lock

(01:04:26):
people into that local market.
They can move a city, maybe,but they will never move from
India to Dubai.

Why wouldn't they move to Dubai?

Because the salary just doesn't empower you
to make that move.
What if?

you adjusted the salary to Dubai, if they moved
to Dubai?

Well, okay, then this person has to call me as a
manager and the typical managersays sorry, that's out of
budget, like maybe next year,Okay, so it gives you less
control.
It's different.
Empowering people means there'sno gatekeeper empowering
empowering people means you'rein charge, you can.
You can live your life.
Here's how.
Here's how.
Like guardrails.

(01:05:03):
Like you, you probably have atough time.
It's not impossible, but youhave a tough time to live on 150
in new york or in san francisco, right, that's just.
You know it, but it's not great.
But if you fly to Argentina nowand you have a US salary income
, you're going to have a greatlife.
So make a choice, really.

So you are accepting the trade-off of like
there are places that areoutside of our.
You can still move to the BayArea, but I'm not going to be
able to like pay you a salary,then you will get a more
competitive salary.
You, that means you're willingto say no to people that want
that, or you or not.
I cannot adjust yourself inthat way and you're willing to
lose people because of thatreason.
But that's fine because of allthe other advantages that you

(01:05:45):
gain on the other side bear inmind.

Bear in mind we we don't have the.
We have this global policy forfor pay bands for junior
engineer, mid-engineer, seniorengineer.
Um, maybe junior designer,mid-designer, senior engineer.
We've never hired more than onedesigner, but our team is
mostly engineers.
When we hire a head ofsomething, now that's a
different tier, right, that's aand.

(01:06:11):
And the truth is now, your, yoursalary comp changes.
Because when there's greatupside when it comes to
engineers, let's say, in ineastern europe or india or latin
america, chances are the vp ofsales, the head of engineering's
the uh, you know, vp oftechnology.
They are in other parts of theworld, they are closer to where

(01:06:35):
the economic innovation happens.
They are probably in the BayArea, probably New York.
So now you need to readjustyour thinking of okay, what's my
total time of employees andwhat's the best salary for that
one?
So, yes, I'm not saying we'recheaping out, but will be, we
are and we will be payingsalaries that make sense in

(01:06:55):
other geographics, right, sothere is a chance that a vp of
engineering makes more moneythan someone else because the
total time of where we can hirethis person is much smaller,
right, but is that?

adjusted by the tier, the seniority, or also do
you adjust there like if youfind the perfect vp of
engineering in san francisco,will you pay higher than because
that's in san francisco?

well, we would still look at the total time of
where we can hire a cfo and it'sit's kind of like all right, if
the time is big, the average isin the middle, right.
If the time gets smaller, youprobably get smaller to the
higher end of the market.
So your average is in themiddle, but at a higher rate,
yeah.
So if suddenly your time ofhiring is just SF and New York,

(01:07:47):
you cannot go below New York,really Right.
So now you have to adjust yoursalary band for this type of
region you can hire from.

we still don't care about the region, but it's just
very unlikely that you hire aVP of finance in whatever
Denmark possible, just rareright, I've been there and I
understand that, like we also atMicros have like open salaries
just internally Everyone knowsthem we do adjust partially to

(01:08:20):
cost of living, to adjust forpersons in power.
It's a different philosophythat we're discussing, but one
of the first things that Inoticed as we're starting to
scale was that the more seniorside, like it again, as you said
, it's a smaller uh pool ofpeople in the world.
Yeah, exactly, and and that alsogives them more, uh,

(01:08:41):
negotiating power, and we do notnegotiate right because these
aren't the salaries in ourcalculator.
Uh, that means we need to like,proactively adjust for that
reality.
So you, it is what it is.
Do you use PAVE also for seniorcompensation?

We do PAVE.
We do use PAVE quite a lot.
We never take it as face value,right?
Because if everyone startsusing them then it kind of
defeats the point, because PAVEis a market index and if
everyone indexes into PAVE it'skind of like you know.
I mean, that's typically how.

Yeah, vicious circle.

Yeah, vicious circle, yeah yeah, yeah, no, I
mean we always make a realitycheck, but PAVE is a good
estimate for sure, especiallybecause PAVE has different
regions.
So a lot of times we just takean average out of all regions
and so if SAP is 250, UK iswhatever and 250, New York is

(01:09:40):
150, and then the United Kingdomis maybe 100.
And you merge it all together,you get a pretty good grasp of
what you can do as a total comp,Gotcha Okay?

perfect, it's up.

Yeah, well, go ahead, go ahead.
Yeah, I think I think we will.
I'm not cheaping out on this.
I think we will have you know,vp of finance, and eventually,
as the company grows, that Ijust have to be in the Bay Area
or maybe have to be in New Yorkif you go public at the New York
Stock Exchange.
But again, this is notcommunism.
This is not like everyone makesthe same money.

(01:10:15):
This is clearly not how itworks.
This is pure capitalism.
But fair right, where you paidmoney, it sounds so funny how
paying for work for, like, equalwork, equal pay you know that
used to be a banner when peoplewent to the streets, right?
I don't think that'scontroversial.
If you pay equal work withequal pay, I don't think that's
controversial.

If you pay equal work with equal pay, I don't
think that should becontroversial at all no, I, I
think even you can argue uh, infavor against it from a business
point of view, and I don'tthink there's like a perfect
definition, but at the end ofthe day also, the cool thing
about being a founder is thatyou get to build the kind of
company and culture that thatyou want, and you don't need to
get everything right to besuccessful.

(01:10:53):
You need to get a couple ofthings right at any given point,
and the rest you get to decidehow you want to shape it, what's
the kind of company and peopleyou want to work with, and I
think that's as you said.
You're still within theboundaries of capitalism, of
supply and demand.
Even if you pay here, but themarket is demanding this, you're
not going to get anyone, butwithin that you can then make up

(01:11:16):
your own rules, and I thinkthat's a pretty powerful,
amazing thing.

It's really cool .
Yeah, I think a lot more aboutmy company building than product
building.
I mean, I'm a product personlike a diehard product person,
but also, you know, I think alot more about how does it
impact culture, how can we keepour culture and how can we grow
as a team and win together.
And all of these ideas I thinka lot of people initially don't

(01:11:46):
think about, and so if you canbuy yourself a little bit of
headspace, and think about that,I think it takes you very far.

I've had this observation that companies that
go remote, like fully remote,fully distributed, especially
across different, very differenttime zones there's a need for
them to be much more structuredon their company building
efforts, Because if you don'testablish some of those best

(01:12:12):
practices from day one it's purechaos, right, and you get out
of companies like like, get labas well.
I'm seeing in your company, I'msending my company and it's a
difficult trade off because atthe beginning what you need to
be doing is, you know, executing, finding product market fed,
talking to the customers.
And it's very easy to forget oncrafting, not just the writing

(01:12:39):
of it but the essence behind thevalues of your company, the
mission of your company, theguiding principles of how you
will make decisions, how youwill compensate, how you will
hire.
But because you don't get thechance to be in person with the
team and have this really tightfeedback loop of how the company
culture is evolving, If youdon't base your decisions on

(01:12:59):
these more establishedprinciples, things can go really
bad pretty quickly.
I don't know how it's been yourexperience about it.

I think I do agree, when you hire people
who've never worked remotely, Ido agree, when you hire people
who've never worked remotely,you need to.
There's a lot of like anupskilling curve just to like
understand how to workasynchronous and work in a
remote environment.
People I hear a lot of times onTwitter when there's many ways
to do a remote company wrong.

(01:13:27):
Right, like, if you try tomimic the office experience, if
you try, if you think we've hadX amount of meetings in person,
let's do them in zoom.
We'd never do zoom calls, whichsounds odd, but like we not
never, but like we try to reallyreduce it down and maybe a 10
minute standup versus a one hourlong brainstorming session.

(01:13:50):
Like empower people to writegood spec, think on their own,
come up with a good plan, post aplan and then have other people
comment on it has been so muchmore efficient and healthier for
your like, work-life balancethan being in a eight hour long
zoom call grind fest.
And I think that's why peoplehate remote work.

(01:14:13):
It's not the remote work per se, it's the constant Zoom calls
that kill you.
I couldn't do that either.
And then there's the other side,where managers are not being
taught how to work remotely andthey try to micromanage and they
don't understand the dynamicsbetween different time zones and
how to work asynchronously andthere's all kinds of different

(01:14:38):
people think that team buildingthings that you need like Friday
games on a Zoom call.
No one wants that really.
I think what we've seen workfor us and we check our company
morale religiously.
We have a HR person who doesinterviews of how people feel,

(01:14:59):
how stressed they are, how theyfeel like the morale is in the
company, how they if theexcitement is, are they excited
about the work they do and dothey get recognition?
So all of these points.
We also publish every surveyevery month on telecomsopen.
So if you want to see the data,it's really cool to see it.

We'll definitely put the link in the episode
notes.

Yeah, and so that helps for me as a founder,
because you're not only buildinga product, you're also building
a company, and if the companymorale tanks, the product
quality tanks, and so that'sbeen great and so that's been
great.
But I think the way to solvethat morale is to work on more
exciting things, to give peopleroom for it, to not micromanage
them, to I mean, of course,everything about like mental

(01:15:42):
health and not burning out andhaving off time, deload and all
that stuff.
That's common knowledge.
But then also to choose yourown friends.
I think that's a really, reallyimportant one, because when you
have an in-office company, yourco-workers are in person and
you have to get along with them.
If there's a co-worker you don'tlike personally, you'll have a

(01:16:05):
bad time.
If I work remotely, I justdon't interact as much with
people, and that's good andthat's also bad.
It's good if it's aprofessional engagement.
It's it's um, it's bad when youthink you have to be a friend,
like a close friend, with yourremote worker, because you you
will never be, you'll never beas close to a remote colleague

(01:16:26):
on a personal, social level thanto your neighbor who's across
the street or your best friend.
And so we have the policy makeyour own friends.
Hang out with your friends,take the time.
I don't play video games withyour friends.
Invite them over, um, becausethat's the ones you choose.
I me, as a founder, I shouldnot be choosing who you should

(01:16:46):
be friends with, right I.

I think that's the.
The double edge of the doubleedge, sort of remote work, is
that it gives you so muchfreedom to control how you work,
if done well, that it's risky,if you don't know how to work
remotely, to spend the whole dayworking, be in the house alone
and then continue to expect,because you come from like a

(01:17:10):
non-remote work, that you willget your socialization from the
job interactions yeah, so Icompletely agree with that.
The only thing I would say butit's not about friday, you know
virtual beers, it's um.
There's a lot of research thatshows that teams that have high
levels of like that allow highlevels of vulnerability, for
people feel the psychologicalsafety to you know, express how

(01:17:33):
they're feeling, that they canconnect emotionally with each
other, that they can givenegative feedback to each other
without you know being hurt orreacting in a defensive way, are
much more likely to make youknow better decisions, to drive
better innovation and so on.
So as long as you are creatingintentionally elements of the
culture that will foster thatvulnerability, I think you're

(01:17:54):
fine.
I agree that that doesn't meanit has to be a zoom call.
Where you like, play videogames yeah, yeah, yeah, no for
sure.

I.
I mean again, that's also wherethe hr person comes in and
we've hired that person.
When we were 20 people, um,like 27 also, it's like not that
crazy, but, um, I think, as youapproach 20, I think someone
said like every, like every timeyou go, and intervals of seven,
so like seven people, 14 people, 21 people, like something

(01:18:23):
breaks, um, I find that veryinteresting.
And so, at 20, 20, 21 people,we, we made the decision to
bring on someone in full time tonot only deal with onboarding,
offboarding and sending peoplelaptops and all that stuff, but
also just to be more of a coach.
This person has a bachelor, Idon't know master in computer,

(01:18:46):
not computer science, in sportsscience, not computer science,
sports science.
He's been a personal trainerfor many years, so he knows the
athlete point of view right, the, you know training, you need to
train, you need to rest, youneed to, you know be the best
version, and.
But that person also offerspersonal training to our team

(01:19:06):
members.
So if you want to do sports, ifyou want a training plan, if
you want a workout plan, he willdo check-ins with you, he will
write the plan, he'll look atthe diet.
I like it it's amazing, uh, Ithink at least half of the
company is using it, um, and,and that gives you a much closer
connection also to your, uh,your yourself, but also to the,

(01:19:28):
the coach you have, um, and sofor me, you know, it's all
connected right, like work-lifebalance, mental health, workout,
good food, good diet, good um,workouts, um.
I think if you, if you caningrain that into your company
culture, it's much easier tosustain the loneliness and pain

(01:19:48):
of remote work that can happenas well, and and the the grind
that they grind to build acompany which is super, super
hard, right absolutely um, yeah,I

you're sharing a lot of things about your company
that to maybe someone new orsomeone less familiar might seem
like very kind of like, um,kind of counterintuitive, like
uncommon.
Like you are remote, globalsalaries, open source, open
startup.
You share a lot of data and atthe same time, you have been

(01:20:23):
very successful at raisingcapital.
You raised, I think, 7.1 onyour seed.
You raised 25 on your.
Your series say amazinginvestors behind that.
Has any of them like complain,not complain as in, like hey,
this might give you troubles inthe future?
I was like hey, you need tochange.
Has someone challenged fromthese investors?

(01:20:44):
You know, group challenge someof these like unique things
about, about cal I don't thinkso.

We were very diligent when we raised funding
to attract people who boughtinto the vision.
We've had a investment memo butwe also had a company values
and visions memo.

separate have you published that one.
By the way, I've seen the memobut not the.

It should be on our handbook.

If you want, we can share them in the notes that
would be terrific.

It's handbook.
cal.
com.
And then on the left sidethere's a vision statement
Perfect, like filtering beforeis always easier than convincing
people after.
So, like, get good people onyour team, I think, is the hard
part.
The rest comes after.
I don't think anyone has everchallenged that.
If anything, alexis Ohanian hasbeen a fantastic board member,

(01:21:40):
really supporting us withanything we do.
We both share kind of like thefuture outlook of less people in
an organization that areempowered by good leadership and
AI.
As funny as it sounds, but thetypical software engineer today
gets probably five times moredone than a couple years ago,

(01:22:02):
not because of AI, but alsobetter tooling and better tools,
better software, betterframeworks and all that stuff.
So I think I find it veryunlikely for us to have 200
engineers anytime soon.
It just doesn't sound right.
And even if we are a series B,series C, series D company, I

(01:22:27):
think I'd rather have a smallteam that gets a lot of shit
done by having this culture thatwe have today.
Again, remember we've gone fromzero to version 4.0.
Small team that gets a lot ofshit done by having this culture
that we have today.
Again, remember we, we've gonefrom zero to version 4.0, with
like almost a quarter million inbookings per month, with a team
of like 12 engineers, right, um, that's, if you can get there,
you can get to 100 million, likeyou.

You just can absolutely at the end of the
year, showing the investors thatwhatever you're doing is
working right and, of course, ifsomething is not working in the
future, they will adjust.
You will cross that bridge whenthe time comes.
But I think, a very importantlesson for anyone especially
think a lot about like newentrepreneurs that might be
wondering, like you know, isthis weird will, like this,

(01:23:10):
break the business and and thelesson there is like wait, you
do get to choose which kind ofcompany you want to build, which
kind of culture, which kind ofpeople you want to attract, and
not just your team members butalso your investors.
And saying no to investors whoare not aligned with your vision
for how to run the company.
It's an important step and it'shard to do, especially if you

(01:23:32):
feel like tight on capital, butI personally believe that it's
better to say no to capital.
That will be very painful inthe people because it's really
hard to get a divorce from aninvestor.

No matter how bad your burn is, do not raise
from the mafia.
Like, just don't you reallywant to keep your kneecaps,
really and this is, of course, amore visual picture but there
are people who you just don'twant on your team and, and you
know, no matter the desperation,I think I've learned in the

(01:24:07):
hard way that like it's betterto shut down a business than to
to keep it alive with the wrongpeople, because every day you,
every day, you'll not be happyabout that decision and from
that on it only gets worse,because now you have new
expectations, right, likeraising more money from bad
people.
Not only you bring on badpeople, but you also raise the
bar of expectations because nowthey have expectations and they

(01:24:28):
want to.
You know a thousand X theirmoney.
So I would rather really justand it sounds hard, but I would
rather just shut down thebusiness than raise for bad
people.

I'm of the same opinion.

And then start a new.
And then it's as funny as itsounds, as sad as it sounds
start a new business, goodpeople, right, it is very it's a
very privileged thing to say.
But also, you know, I've beenat a point where you have no
funding and you just need to putin the hours and bootstrap and,
you know, get get to a pointwhere you are in a good position

(01:25:01):
to make good decisions.
Um, a good product attractsgood people, right?
Good, good, good businessesattract good people that's such
a common mistake, right?

It's like thinking I need to attract capital so
that I can attract, so that Ican build a business, instead of
like I need to build a businessso that, I then gain the
privilege of needing the moneyto build the business further
right.
But it's the other way around.

It's always the other way around.

It's been fascinating to get to know Cal
and your approach to buildingthe company and the culture.
It's fresh, it's different inmany ways and I think that's the
key here.
It's like to expose people tomany different ways of building
companies and buildingbusinesses to show that there's

(01:25:53):
just no one single path and youget to choose them, and I'm
really grateful for you sharingyour journey and your approach
and also excited for otherpeople to learn about it.
So thank you so much fordedicating us this time, thank
you for having me Adil, thankyou so much for tuning in.
Your support means the world tome.

(01:26:13):
If you enjoyed today's episode,please consider subscribing and
leaving a review.
It's one of the best ways youcan help this podcast get off
the ground and help moreentrepreneurs like you.
Thank you.

All Episodes

Episode Transcript

Popular Podcasts

Dateline NBC

Stuff You Should Know

My Favorite Murder with Karen Kilgariff and Georgia Hardstark

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Monetizing Open Source with Peer Richelsen (Cal.com)

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}Dateline NBC

Stuff You Should Know

My Favorite Murder with Karen Kilgariff and Georgia Hardstark

All Episodes

Monetizing Open Source with Peer Richelsen (Cal.com)

Dateline NBC