Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Hannah Clayton-Langton (00:04):
Hello
world, welcome to the Tech
Overflow Podcast.
As always, I'm Hannah CaitlinLangton.
Hugh Williams (00:10):
And I'm Hugh
Williams.
Hannah Clayton-Langton (00:11):
And we
are the podcast that teaches
technical concepts to smartpeople.
How are you, Hugh?
Hugh Williams (00:16):
I'm well, Hannah.
I'm well, Hannah.
I'm coming to you live from myuh my little studio on my farm
today.
So that's uh that's prettyexciting.
How are you?
How's uh life in Ohio?
Hannah Clayton-Langton (00:26):
Well, I
am coming to you from inside of
my walk-in wardrobe, which uhwas always a dream of mine to
have one.
And I never thought I'd beusing it for podcast recording,
but the acoustics are prettygood in here.
So today, listeners, we havegot a real treat for you, which
is our first ever tech overflowinterview.
We've got a really awesomeguest coming up, and he's gonna
(00:47):
tell us a little bit more.
Hugh Williams (00:48):
Today we have
Jonathan Bedeen, who is one of
the co-founders of Tinder.
For those of you who don'tknow, Jonathan, he's a pretty
well-known guy, but Jonathanreally is the product brains
behind Tinder.
So he he has the claim to fameof having invented the swipe
right, which I guess is just ametaphor these days for saying
yes to things.
Jonathan today is gonna reallytalk through the early days of
(01:10):
Tinder and gonna really dig intothat story, both from sort of a
visual design side and productside, right through to some of
the tech that was behind it whenJonathan first built the
original Tinder interface.
Hannah Clayton-Langton (01:23):
Yeah,
it's super cool episode.
You guys are gonna love it.
And there's two things I I loveabout it.
One is that I was a Tinder userback in sort of 2014, the early
days.
And I think I mentioned this onthe episode, but if not, I met
my husband on a dating app.
So, you know, I'm verypersonally invested in the
course.
And uh secondly, and probablymore importantly, what I loved
(01:44):
about our interview withJonathan is that the way he
speaks about the evolution ofdeveloping the technology really
resonates with a lot of thethemes that we talk about in our
previous episodes.
So if you've not listened toepisodes about sort of coding
and product and apps and how itall hangs together, it's super
useful foundations for thisepisode.
And it was a really good testcase of what content we pulled
(02:05):
together so far, which was whichwas nice.
Hugh Williams (02:07):
Yeah, I I'd say
to any of our listeners, if you
haven't listened to thoseepisodes, go back and listen to
them.
They're a great foundation forthe conversation that we're
gonna have with Jonathan.
So should we get him on theshow?
Hannah Clayton-Langton (02:16):
Yeah,
let's do it.
Hugh Williams (02:18):
Jonathan Badin,
welcome to Tech Overflow.
It's so, so good to see you, myfriend.
Nice to see you.
Thanks for having me.
Yeah, it's been a long time.
Uh obviously, we workedtogether at Tinder in I want to
say 2015, somewhere aroundthere.
You're making me feel old.
Hannah Clayton-Langton (02:34):
And I
can say that I was a user of
Tinder in 2015, so we were allpart of the story.
Hugh Williams (02:39):
Oh, it's all
connected.
And did you meet anybody onTinder, Hannah?
Hannah Clayton-Langton (02:43):
Not to
take our podcast in any strange
place, but I could regale youwith some tales off mic, but the
uh sum total of my dating appuse was a success story, as I'm
now married to someone I met ona dating app.
Not on Tinder, but I feelstrongly that Tinder changed the
face of dating, which I'm surewe'll get into.
So I still give Tinder largecredit for my happy marriage.
Jonathan Badeen (03:03):
I'll swipe
right on that.
Hugh Williams (03:05):
Look, Jonathan,
you had many, many claims to
fame, I'm sure, but I think thething that you are best known
for is actually inventing theswipe right, which we obviously
want to get deeply into in thisstory and talk about the genesis
of that and the technologybehind it.
But I thought perhaps a reallygood place to start is right at
the beginning.
So, how did you actually getinto tech in the first place?
Jonathan Badeen (03:26):
Uh, you know, a
long time ago, I don't know, it
was middle school, I startedkind of getting into computers
and all.
And then during the dot-comboom, I and a friend uh created
a partnership.
We were making websites.
I didn't even know HTML,though.
I was using like Microsoftfront page, you know, it was
more of just like a hobby, it'sjust something I just kind of
always did, sort of side gigkind of things, and then uh went
to school for other things,went to LA to act, which is a
(03:50):
little bit out of left field orwhatever.
And then I ended up working ata uh company and got a job as a
web designer uh because of awebsite I made for somebody else
that they liked, and was sortof cognizant of conflict between
designer and programmer, youknow, designers trying to ask
for the world, programmerssaying you don't understand how
(04:10):
this works and that's reallyhard and gonna take too long or
whatever.
So I kind of decided that totry to learn some of the stuff
so I could help out, you know,communicate, how I was gonna uh
talk to them and all.
And it was much easier at thattime to do it because before
it's so daunting.
If you're especially in the webworld, it's like, okay, I need
to know HTML.
Well, that doesn't do much ofanything.
I need to know CSS.
(04:31):
Okay, well, now I've got astatic site that does nothing,
and no, I need to know uh somesort of back-end sort of thing,
and it's you know, C sharp oruh, you know, Ruby on Rails or
PHP or whatever, and you know, abillion different things to
choose from, and I have nocontext for what the right thing
is to choose, and you know,then of course a database and
all is you know, having to learnlike five different
(04:52):
technologies just to do onesimple thing, JavaScript.
Sorry, I forgot about that.
But when I worked at thatcompany, it was this opportunity
where I could just take onepart of the stack and learn
that, and so that kind of got meinto the programming aspect of
it, and uh obviously kind offocused more on the front end of
things, coming from sort of adesign product side, because
(05:13):
that was really the part that Igot really excited about the
sort of the animation, theinteractivity, the uh the
visuals and all.
And then ultimately that endedup leading me to Mac
programming, which ispractically the same thing as
iPhone programming and fastforward, you know, a little bit.
I'm writing Tinder.
Hannah Clayton-Langton (05:33):
How did
you get from there to working at
Tinder?
Or was it not even calledTinder when you started working
there?
Jonathan Badeen (05:37):
Well, was it?
Yeah.
So uh so originally, you know,I was working at this uh thing
called LA Casting CastingNetworks.
That's where I'd learned to dothe Mac programming, and I I
taught myself to do that.
Uh from there I went to workingat a company called Cheg, who
had just acquired a company inPasadena called Cramster.
So I was working there and kindof hired there as sort of like
an incubator uh of sorts to tryto use this API that they were
(06:01):
um working on, which they neverfinished, at least while I was
there.
I ended up uh being introducedto Sean, my co-founder, to join
him at Hatch Labs for somethinghaving nothing to do with
Tinder.
It was we were working gonnabuild this thing called Cardify,
which is like a loyalty rewardsprogram that you'd get points
off of your credit card atrestaurants and stuff like that.
Although when I first did meetSean, he was they were working
on a hackathon project withinHatch Labs, which was a um
(06:25):
something he was callingMatchbox, which ultimately would
become Tinder.
So we made Cardify and we hadsome issues getting it out there
and knew we had some downtime.
And that downtime, Sean's like,you know, what do you what do
you think about like making thismatchbox thing for real?
And uh basically built Tinderin like six to eight weeks and
and launched it, went back toCardify right afterwards,
(06:47):
rebuilt that and pivoted withthat, and then decided, eh,
we're not really so interestedin that.
Let's let's focus on Tinder.
Hannah Clayton-Langton (06:53):
Okay, so
the the hackathon for
non-engineers, like that was anevent that was like a quick and
dirty for fun concept definitiontype thing, and and that was
where the idea was born, andthen it turned out to be.
And who who was in thehackathon?
Like, do they get credited forTinder or so like yeah?
Jonathan Badeen (07:12):
Sean Sean was
in that there, and then another
guy uh named Joe Muniz, uh, whowas early on at uh at Tinder,
they did it.
I don't even know what if theyactually built anything, but it
was there was the presentationfor sure, sort of this concept
in general.
Not every element was the same.
There's like, you know, it's itwas more focused on like uh it
was going to be used in a bar ora restaurant or something, and
(07:33):
not the restaurant next door,but in this restaurant or
whatever, uh, as opposed toTinder, which you know was a
little bit a wider radius andusing some different
technologies, you know, to tryto accomplish that.
But uh yeah, it was uh it waspretty close to it.
So they did that, and I don'tknow, it was like a weekend or
the week or something like that,kind of pitched that.
And then a few months laterthat we actually sat down and
(07:54):
decided, let's let's build thisthing.
Hannah Clayton-Langton (07:57):
I
presume online dating on like
match.com or plenty of fish waswell established at this point.
And we were just, I think I gotmy first iPhone in like 2012,
which may have been a little bitlate, but that was that was the
year that you launched, right?
So was it the first use case?
I presume it was the first usecase of a so-called dating app.
And was the original matchboxan app or was it a website or
(08:19):
something?
Jonathan Badeen (08:19):
So the original
matchbox was to be an app, it
wasn't it never really existedbecause Hatch Lab's focus was
actually on creating uhdisruptive mobile apps, so
native mobile apps and all.
And so that that was the focus.
And then uh, of course, Tinder,when we were making Tinder, uh,
we were gonna call it Matchbox,but because of issues that
(08:41):
could arise with match.com andall, started looking for
different ways of uh differentnames and all.
And Tinder was one of thosekind of went along with the fire
theme that we were kind ofgoing with.
And so we ended up ultimatelylanding on Tinder.
Hugh Williams (08:53):
I'm super
interested in sort of the the
creation of the of the product.
I mean, I'm dying to get intothe swipe right itself, but how
did you sort of decide orrealize that it should be a you
know a photo-based, app-basedtool?
I mean, did that did that sortof come out of some user studies
and and and data, or was itjust a moment of inspiration?
So how did you actually sort ofend up where you ended up?
Jonathan Badeen (09:13):
You know,
there's a whole bunch of things
that kind of go into it.
You know, as you kind ofpointed out, too, I think was
actually a really good pointthat like the iPhone came out in
like 2007, but you didn't getit until 2012, which is
coincidentally when Tinder wasreleased.
And so it really, even thoughit's a five-year time period,
it's not like everybody got aniPhone on day one.
So we're really still early onin that time frame.
(09:35):
But you know, we were lookingat some of the things that
existed out there in the world,your match.com, your okay, as
you mentioned, plenty of fish,also eHarmony, and all of those
things, they were verydesktop-based.
They've been around for awhile, they had a lot of cruft
and extra billion differentfeatures in there.
And a lot of them, they yeah,they had a mobile app, but they
were secondary.
(09:55):
It was like they they had onebecause they had to have an app,
you know, they didn't, itwasn't like their main vector or
anything like that.
In fact, I think I remembertalking to Match group at one
point or the uh CEO and waslike, you know, you can't even
actually create an account inyour app, it's completely
broken.
They really weren't giving itmuch love.
And so there's between thatwhere it's half baked to try to
cram everything onto it.
(10:16):
And you have to kind of realizeback in 2012, as you remember,
the you know, we were talkingabout an iPhone 4S, I think was
the phone at that time.
So on really small screens andall.
So part of when you're decidingto make a uh dating app for a
mobile platform, a very smallscreen and all, you're gonna be
prioritizing very differentthings.
Instead of having maybe a gridof a whole bunch of photos, uh,
(10:39):
that's gonna be pretty small.
It's going to start pushing youin that direction of one person
at a time.
That was sort of something thatSean had kind of envisioned.
I know when I saw it, I verymuch kind of agreed with that
sort of thing.
It reminded me of the hot ornot, which was a thing when I
was in college.
And I remember how addictivethat was going through one at a
time and and sort of rating andthen just always like, well,
(11:01):
I'll just I'll I'll stop after Ifind one more nine or whatever
it is, you know.
And so I think that that reallylent it to it.
But there were a lot ofdecisions that got made because
of the platform.
You know, obviously your swipecouldn't exist without gesture
technology, which comes from atouch screen.
Uh, we originally had sort ofthe a quick and easy way to log
in or to create an accountbecause you don't want to be
typing profile for an hour on asmall little uh so Facebook
(11:25):
login, uh, which was more of athing then than it is now, but
it allowed us to create theseprofiles real easy.
We've crafted the communicationpart of it more off of texting
as opposed to the previousthings, which were more
email-centric.
You know, you've got this smallscreen, you're gonna put less
information about the personright up front.
And so the the very firstversion had a photo, although
(11:47):
not quite as large as itultimately ended up becoming.
First name, age, and it had uhnumber of shared friends and
shared interests.
And then you'd tap into theprofile and you could get all of
that plus a little writtenblurb.
But uh, you know, I think it'suh a lot of people think, oh,
it's only photos.
Never was only photos.
However, it does turn out to beactually one of the most
important things.
(12:07):
And I think okay, Cupid didsome interesting tests that
really proved that too.
Hannah Clayton-Langton (12:12):
And so
let's start with the most
defining feature of Tinder,which is the swipe right.
How was that born?
Why wasn't it swipe left?
Like what insights can you talkus through?
Jonathan Badeen (12:22):
So there's a
couple different stories out
there about the uh origins ofthe swipe, and one is a um makes
for a great scene in a movie orsomething.
The other one's the real story.
And the real story is that theorigins of the swipe actually
predate Tinder.
I had been working at Check,and somebody uh within my team
(12:43):
brought up the idea of likemaking a flashcards app.
I was pretty dismissive of thisat first.
I was like, there's a billionflashcards apps.
What are we gonna do that'sgonna be interesting?
Hannah Clayton-Langton (12:54):
Um, this
is um flashcards for like
studying, right?
Because you said exactly.
Jonathan Badeen (12:59):
Yes.
So like you're talking likecollege students and stuff like
that, you know, learning, uh,you know, studying for the exams
and things like that.
Um, and so I started playingwith some of them, and I did
actually start to notice somethings I really hated about
them.
So for instance, I, you know,one, I kind of a I was a
procrastinator, I was alwaysgonna be studying last second.
I would be probably if I was,you know, walking with one hand
(13:22):
with my coffee in it, and if Ihad a if the iPhone existed back
then, I'd have the other handcarrying my iPhone, the class
studying last second for mytest.
And most of these apps or allof them would be very
button-centric.
And if you ever watch somebodywhen they're walking and
texting, for instance, they slowto a crawl as they're texting,
(13:43):
right?
And that's because when you'rewalking, you know, one, you're
distracted, but two, it's harderto hit things.
You are moving, so it's harderto hit little uh targets on the
screen.
So I realized I was like, thesebuttons, one, they're visually
in the way.
I want to study a focus on thecontent, but two, they're hard
to hit when I'm walking theclass.
(14:04):
Uh, not that I was walking theclass.
So it kind of pushed me towardsthis like, this should be
gestural.
And there were a few that didhave one basic gesture to it.
You swipe in one direction tokind of like flip the card over
to see the other side.
But in order to say, like, yes,I got it right, or no, I got it
wrong, that required pushingsome button.
(14:24):
The swipe on the flipping wasalways horrible anyway.
They they did a horrible jobimplementing it.
But I woke up one morning withthis epiphany, just like
literally woke up and just gotreally excited about how I
thought you would make theperfect flashcards out.
And that was you would swipe inone direction for flipping the
card over and you'd make anactually good swipe, but then
(14:46):
you'd swipe in the otherdirections for saying I got the
card right or I got the cardwrong.
And I kind of came up with thiswith the idea of like how I
would use real flashcards, likein real life, not going to
class, but like if I was sittingthere, I'd start out with a
stack of cards.
And I would take that card andI'd put it into one pile if I
got it right.
And that's the the cards Idon't need to study anymore.
(15:08):
Uh, you know, oh, I got thisone wrong.
I'm gonna put it over here intothis other pile.
It's the one that I need tostudy more.
And so now you got three pilesof cards until you whittle it
down to two.
And so I envisioned those twostacks of cards, the right and
the wrong ones, right off thescreen of the iPhone, because
your screen's small, right?
And basically that's where thegesture comes from is dragging
(15:29):
that card to the wrong or rightstack that's just off screen.
So ultimately, when we ended upmaking Tinder, we had already
landed on this sort of one at atime sort of card.
We're also in this very skewmorphic period.
Hannah Clayton-Langton (15:41):
So we
had sorry, skew morphic.
Jonathan Badeen (15:44):
No worries.
So skew morphic is like theoriginal, uh, you know, if you
looked at the old iPhone, iPad,and all of that sort of stuff,
and the calendar app looked likea physical calendar.
The notes app looked like, youknow, had the little tears in
the sheet.
It looked like a physicalnotepad.
That's skew morphism.
It's using something from thereal world, and and it's it's
useful for making somebody, youknow, comfortable in a digital
(16:06):
setting, try to help them feelmore familiar with it.
And it's it kind of went awayafter like iOS 6, it uh it was
got pushed out.
But so everything felt likereal objects back then.
So we did have this sort ofstack of cards.
Uh, then we thought, well,let's use my swipe from the
flashcards app and throw that inthere.
And the very first version, wedidn't actually implement it.
(16:27):
Uh, I thought it was going totake me a week to implement, and
Sean was like, eh, we don'tneed it.
But we had the big heart and Xbutton and all, so it wasn't
necessary.
But it one one day, uh a coupleweeks after we released the
very first version of the app, Ithought, well, let me see one
night, you know.
And I ended up being able to doit within a few hours.
And I don't know, maybe a weeklater, I don't know.
(16:47):
I was like, oh, by the way, youguys can swipe in here, but
never really thought this waslike the defining feature of
Tinder.
It was just sort of like akeyboard shortcut, like, hey, if
you don't learn, figure it out,great.
Um, but it was plannedoriginally, like the very first
version, when you hit the heart,you know, the light stamp would
come on it, the card would flyoff to the right.
And if you hit the X, it wouldfly off to the left, basically
(17:09):
with the intention of being ableto actually drag those off or
swipe those off uh later.
Hannah Clayton-Langton (17:14):
And why
was right the yes?
Are you right-handed?
Like why was right the goodfile and left the bad?
Jonathan Badeen (17:20):
It was simple
as in our design, the heart was
on the right and the X was onthe left, and making it match
that.
Uh, however, I think it workedout nicely because in many
versions of Tinder, you wouldactually get to your match list.
It was kind of positioned offto the right of the screen.
So it kind of worked out nicelybecause it's like, yes, I want
(17:40):
to put this into my match list.
Hugh Williams (17:42):
It must have been
one heck of a night of coding
to go home and create that.
And obviously, you got itworking and the rest history.
But I'm super interested in thesubtleties of swiping right and
swiping left.
And knowing you, Jonathan, Iknow you cared a lot about those
details.
But maybe you can just sort oftake us one level down into sort
of a later version of Tinderthat you built and tell us just
a little bit about thesubtleties of swiping left and
(18:03):
swiping right and the animationfrom a from a design
perspective.
Jonathan Badeen (18:07):
Sure.
So, first off, I you know, kindof going back to the flashcards
thing, within iOS, there's thisthing called the swipe gesture
recognizer.
That thing is evil, it'shorrible.
That's what was wrong withthose other flashcards apps
because it recognizes it, ittriggers a swipe after you
finish the swipe.
So that means you go across thescreen, it says, Oh, there's a
(18:28):
swipe, and now somethinganimates based on that.
You know, I was like, well,that's horrible.
So, in order to do a properswipe, basically allows you to
just track your finger on thescreen.
So it that way I can animatebased upon or just move things
around based upon the movementof your finger.
But I wanted it to feel real, Iwanted it to feel physical and
(18:51):
react to you as you know,similar to a real card.
So to do that, you know, I kindof thought about like, well,
cards don't move perfectly whenyou drag them on a on a surface.
You know, it's you depending onwhere you put your finger on a
card and drag it, it's going toangle itself because the drag on
it, right?
So I kind of split the cardinto quadrants based upon where
(19:13):
you're starting your thing, yourtouch on the card, it will
angle different depending onwhich quadrant that you started
swiping from.
So then the card will kind ofdrag in one direction or the
other.
So there was that part andtrying to get that uh to feel
right, get that rotation to feelnatural.
And then on top of that, therewas the release.
(19:34):
Uh, you want to make it feelnatural when you release that
swipe.
There are a couple differentfactors I'd look at.
It's like one, did it match acertain velocity, or did they
stop swiping completely?
So if the gesture ended with acertain velocity, then it would
go off the screen.
If not, it would bounce backtowards the center.
Then I'd use that velocity andI think I'd create essentially
(19:57):
like a fictional distance offthe screen and then calculate,
use that velocity to uh createan animation with a time uh
based upon all of that to get itoff the screen so it felt
natural.
Because you wouldn't thinkabout these things normally, but
like you know, if you'redragging really fast and you let
go and there's a standardanimation that says, oh, this
thing flies off in one second,all of a sudden the card goes
(20:20):
faster, slower than what youwere dragging.
To make it feel natural, youactually need to take into
consideration all of thesethings.
Hugh Williams (20:26):
And did you do
all that up front before you you
released it and and and put itinto public consciousness?
Yeah.
Jonathan Badeen (20:33):
I'm sure it was
finessed a little bit, you
know, here or there and atcertain points, but yeah.
And did you write all of thiscode yourself?
I did.
I did.
Hannah Clayton-Langton (20:42):
And you
mentioned a phrase earlier on
gesture technology, which iskind of self-evident, but could
you just talk the listenersthrough that in a bit more
detail?
Jonathan Badeen (20:50):
Sure.
I mean, you look at previousthe iPhone, I suppose there were
sort of gestures when you'redragging a mouse or whatever and
all that sort of thing.
But you know, it's uh for themost part, things were sort of
point and click.
And then Apple, you know, comesout with the iPhone and it's
got this touch capacitivescreen.
So you're actually it'srecognizing your your your
(21:11):
finger input and it can actuallyrecognize multiple fingers and
all.
And that just really opened upa whole new world of
possibilities.
Uh and I think really excitingpossibilities because it makes
you feel far more connected tothese objects on the screen
because you're you're able toactually touch, manipulate them,
um, and and it feels um verylifelike.
(21:33):
And so that that was justsomething that that was really
new.
It's you know, it's it's almostminority report-esque, you
know, instead of having thisextra device like a keyboard or
mouse between you and thetechnology, you're touching the
technology.
Hugh Williams (21:47):
And has the swipe
right changed much over time,
Jonathan?
I mean, is the version thatthat folks are using today?
I mean, it's so ubiquitous,particularly in dating apps.
Is that is that effectivelywhat you build, or is have some
things changed over time?
Jonathan Badeen (22:01):
I'd say there's
uh one big thing that is
different now.
There was one thing that we didexperiment with, but it never
made the never saw the light ofday.
At one point, Apple came outwith this pseudophysics
framework, and we thought, oh,let's use this thing.
This thing is going to make theswipe even more lifelike and
(22:21):
more uh, you know, feel morerealistic.
And so we implemented it, andthe damn thing felt horrible.
It was so realistic, though,that you could like put your
finger on the corner of the cardand flip the card around in a
circle, you know, on the screenand make circles with it and
all.
But like it just felt reallycumbersome.
It felt like you were having todo a lot of work to swipe and
(22:42):
realize that our sort ofpseudo-physics um that we had
come up with, which may notcompletely match the real world,
but they felt way better andjust uh just much more fluid.
So that was one thing thatalmost changed, but didn't
change.
But the other big change, andand Hugh, you were you were
there at that time, was uh weended up adding a swipe up for
(23:04):
super like, which I did not likeand I do not like to this day,
but uh that obviously changed alot and did require a lot of uh
adjustments in the code to tryto figure out which direction
they're trying to swipe in.
Because before it's prettyeasy, it's either it's left or
it's right.
You can't be a little left anda little right, it's one or the
other.
Later on, you're is it a littleup or is it a little right?
(23:26):
Uh is it I it it could be both,you know, which a little bit
ambiguous as to where where thatswipe is going.
And then besides that, therewere other elements that weren't
necessarily swipe specific thatchanged over time, but more of
just the the stack, whattechnology we would be using to
present those cards.
But I would say that the swipesort of physics and in general
(23:48):
that kind of code is has stayedfairly uh uh similar throughout
time.
Hannah Clayton-Langton (23:52):
And the
other apps that utilize the
swipe, right, of which there aremany, even like beyond dating
apps at this point.
Is that all built around thesame technical principles, or
was there like an IP thing therethat you guys owned?
Jonathan Badeen (24:04):
Sure.
So we did we did um patent theswipe.
There's been a couple differentlawsuits and stuff over it.
People have come up with theirown things.
You can go on GitHub and findall sorts of different
implementations for it, somebetter than others.
I've seen a lot of them thatthey don't factor in like the
quadrants uh touch.
It's just it's a standard.
They'll do it uh rotation inorder to give it the similar
(24:27):
sort of feeling, but I call ituh swiping on rails because it
doesn't matter how you swipe it,it's always going to be in the
exact same place, the sameanimation.
And that's okay, but you know,I take pride in what we made.
Hannah Clayton-Langton (24:40):
Did you
learn anything unexpected when
you got user testing on theoriginal swipe right out there?
Was there anything that liketook you by surprise?
Jonathan Badeen (24:48):
Well, you know,
the first thing I suppose was
people started using it a littlebit, but they didn't
necessarily know what it was.
Now there were some indicatorsin there, you know, as you
swipe, that's mimicking whathappens with the buttons, but as
you swipe also uh the like orstamp or the nope stamp are
coming up to try to indicate.
But some people just thoughtthey were like skipping a
(25:09):
person, they didn't realize theywere actually taking an action.
So that ultimately led us, Idon't know, a month or two later
or something, to uh we didn'twant to educate people
beforehand because we didn'tthink it was important.
It's like we don't want to slowthem, we want them to get into
the app, start using it, startsaying yes or no to people.
Um, that's the fun part.
So we didn't want to have likesome tutorial when you started
(25:30):
up the app, like here's how touse Tinder.
So what we opted, instead ofhaving to teach people to use
the swipe, because again, wedidn't even think it was like a
defining feature, was after youswiped, we'd say, Hey, you just
swiped right.
Swipe right means this.
Did you mean to do that?
And you'd say yes or no.
And if you said yes, great.
And if you said no, the cardwould come back and allow you to
(25:53):
do do it again.
Um, so the education would onlyhappen if you tried to swipe.
Hugh Williams (25:58):
That's super
smart.
I love that.
Jonathan Badeen (26:00):
But then I
think also to answer your
question, I I I think there's aninteresting story.
Like I said, not supposed to bethe defining feature of Tinder,
but there was um a few monthsin a college student emailed us,
and he said that he and hisfriends, when they're walking
down the street, would say swiperight or swipe left as code,
(26:21):
uh, because you know, peopledidn't know what it meant back
then, as to uh an attractivelady passing by or something
like that, which I guess thedownside is they were doing a
little bit of objectobjectifying, but he even
suggested using that in ourmarketing.
And it was at that moment thatit made me realize I was like,
oh my God, we we really tappedinto something or swiped into
something, I guess.
(26:41):
And and so that was it was notlong after that that we actually
started to market the swipe anduse that as one of the defining
features of Tinder.
Hannah Clayton-Langton (26:50):
Tinder
is one of the few things that
like you'll say, like Tinder forfill in the blank as like a
different use case for thedefining feature, but that
defining feature is the swipe,like so.
It's kind of crazy that therewas a Tinder without the swipe,
and it was just uh aninteresting idea that you coded
in one night because now I meanI think you said it up front in
the episode, right?
Like swipe right to that meansI like it.
(27:12):
And I think people don't haveto be dating app users or have
ever used Tinder to know whatthat means.
So it's um it just goes to showit's not necessarily the things
that you think will land thebest that do.
Jonathan Badeen (27:22):
Exactly.
You gotta roll with it.
Hannah Clayton-Langton (27:25):
Hughes
taught me about product as being
like the intersection betweenthe technology, the user, and
the business, as sort of likeVenn diagrams.
And we talked a lot about theuser and we've talked a lot
about the technology, but whatwas the commercial use case that
you were building to?
Was it like at some point thiswill be a paid-for subscription
app, or were you you mentionedthe super like?
Like, were you already thinkingabout paid for features, or
(27:48):
were you just trying to getsomething off the ground and
popular and not too worriedabout the commercials of it?
Jonathan Badeen (27:52):
Both.
Uh, we weren't too worriedabout that aspect.
We were um, it was one of thosethings like we come up with
things like, oh, maybe we couldcharge for this in the future,
you know, or have ads orwhatever, you know, things like
that.
But we were focused wholly ongrowing the thing.
Uh, we wanted to grow that userbase, you know.
We'd looked at them as, youknow, the Facebooks and the
(28:13):
Googles as the model of, youknow, get the users first, then
figure out how to make somemoney off of it.
Um, so we certainly did.
In fact, even in that uhpresentation, there's mention of
like being able to givesomebody a rose.
And I think the super like forit was actually the evolution of
that rose idea.
Hugh Williams (28:32):
Just coming back
to the the super like, Jonathan,
I remember that very, verywell.
I was uh I was there at Tinderat the time, and I remember you
stopping that release.
I remember us, you know, we'regonna go for it, and then you
said it's just not quite right.
You know, we need to we need tospend some more time on this
before this goes out the door.
And ultimately it did go outthe door.
I think I know you didn't loveit, I probably didn't either.
Um but tell tell us that storyof sort of stopping the release.
(28:53):
So I want to know what's inyour sort of your design and
product mind when you know thatsomething's not ready and that
it needs more work.
Jonathan Badeen (29:00):
I don't
remember the specifics of
different things, but I I knowthat there were a few elements.
Obviously, the swipe itself, Ithink was so important to us at
that time.
We needed to get that right.
We needed to get the velocitiesright.
We needed to try to make it asaccurate as to the intentions of
the user as possible.
And especially if you weregoing to sell something too, I
think we really wanted to makeit feel like you're getting
(29:20):
something.
With the super like, we also wewe added a whole bunch of
flourish to it, you know, withextra special animations when
you pressed the super likebutton and it would move up.
I think it kind of compressedand kind of flew up like a
superhero kind of would.
We kind of gamified it withputting like the star, like when
Mario hits the block and thecoin flies up, kind of like
(29:41):
that, and it would twist around,and I think it would show you
how many you had left.
So I think there was just thosethose elements that we were
really, I think, just reallytrying to nail.
But I think uh, if I remembercorrectly, I think it was really
focused on that swipe.
Hannah Clayton-Langton (29:53):
Okay, so
before we go, is there anything
you can share with thelisteners that, particularly for
the non engineering listener?
That was technically like moredifficult or surprising or
complicated versus what we asend users would have expected.
Jonathan Badeen (30:08):
Oh, absolutely.
There's a billion.
But I think kind of going alongwith the theme of the swipe and
all, there is a lot going on uhbehind the scenes to try to
make the experience seamless.
We wanted you to uh not have toworry about your internet
connection and all and thingslike that.
So we would we would download acertain number, you know, we
(30:31):
played a lot with those numbers,whether it was we're gonna
download three or download 20different cards at one time.
And some of that has differentconsequences of like, okay,
well, you're trying to get thefreshest people, people who are
using things or most pertinentto you.
So you want to do smallerbatches instead of larger
batches.
So we fiddle with thosenumbers.
But then on top of that, it'slike, okay, well, now you're
(30:52):
swiping on people and yourinternet goes down.
Well, we don't want it to juststop working.
So what we'd be doing is we'dessentially be caching that on
the client and storing all ofthose things.
So you swipe right on thisperson, you swipe left on that
person.
Once we get an internetconnection back up, we'll send
all of that stuff up to theserver.
But then other interestingthings happened at certain
(31:13):
points.
So, like uh, especially earlyon, we had all of these issues
with just our servers justoverloaded.
We needed things to be muchmore efficient, and they
weren't.
And you we'd have multipleservers, but they weren't
necessarily fully synced.
So you'd be asking a server,you know, give me some new
(31:33):
cards, and then you'd get thesenew cards or to swipe, and then
you're done with them, and youask another server, you don't
know which one, it's just kindof automatically routing into
ones, and it might give you thesame cards because it doesn't
know that you already had them.
And we don't want to show youduplicates, and it got really
bad to the point that if wedidn't do anything, you're just
(31:54):
swiping on the same people overand over and over and over
again.
So I had to build a whole bunchof protection into the client
where I started to create anextra cache of everything, every
person that you saw.
And so that if a card came downfrom the server of somebody
that you already saw, we wouldnot show it to you.
And then we would, you know,keep asking for more cards until
(32:16):
we had enough to do everything,and we filter those out and
then come up with some rulesbecause we didn't want your you
know, your local database tojust balloon with every person
that you ever saw on Tinder.
So it was sort of like if youhadn't seen a duplicate within
24 hours or something like that,we would toss out everything
just in feel because we'd feellike okay, the servers are
probably in sync at this point.
So there's all sorts of littlethings that we do to try to make
(32:39):
it so that it feels like it'salways working, even when it's
not.
That's such a great story.
Hugh Williams (32:44):
I got one more,
Jonathan, maybe, and then we'll
wrap up, set you free.
You created one of the greatstartups of our time.
I mean, I think everybodywishes they'd created Tinder.
There's probably people rightnow pitching to a VC saying I've
got Tinder for whatever.
Um, and I know you advisestartups.
I'd love just a final word fromyou about, you know, what
advice would you give tosomebody who wants to be an
(33:06):
inventor or wants to createtheir own startup?
Jonathan Badeen (33:08):
You know, I
mean, I think it's really just
to go for it and to meet people,to learn and expand your
knowledge and and play withthings.
I to me, I I could never tellyou the secret to things, but I
feel like there's a certainelement of luck, but luck, I
think, is based upon it's kindof where opportunity and
(33:29):
preparation coincide, where theythey meet.
So it's you know, it's aboutlearning the things and when an
opportunity arises that you'reuh equipped to take advantage of
it.
But you got to get yourself outthere and say yes to things,
you got to meet people, you gotto try things because you know,
the more times you try, morethings you expose yourself to,
(33:50):
that's where you're going tohave that opportunity to apply
your learnings.
Hugh Williams (33:54):
So it's partly
inspiration, a lot of
perspiration.
Jonathan Badeen (33:58):
Absolutely.
You know, I find it um funny.
A lot of times I'll have uhpeople pitch things and they're
very worried about somebodystealing their idea.
And I, you know, you shouldalways, you know, be cautious or
whatever.
But at the same time, most ofthe good ideas that have blown
up in the world are things thateverybody thought they were
crazy.
Nobody was going to steal itbecause they thought they were
(34:19):
crazy to do it.
And so it's really about theimplementation.
A lot of people might have evenhad your idea.
It's about getting it right.
And that takes a lot of work,uh, takes insight that is, you
know, may come naturally or itmay come from just putting in
the work to learn the insightsof how to get things done and uh
make a product that's actuallygoing to work for people.
Hannah Clayton-Langton (34:39):
Thank
you so much, Jonathan.
That's been very insightful.
I've definitely learned a lot,especially around what skew
morphia or skew morphic means,uh, which felt very relevant
actually when we were talkingabout the user experience.
So I'm glad I asked thatquestion.
Been great to chat, and I'msure the users will learn a
thing or two.
And I need to go and have alook for that supposed pitch
deck, which you say is somewhereout there on the internet.
Hugh Williams (35:00):
Thanks, Jonathan.
Great to see you.
Nice to see you.
Hannah Clayton-Langton (35:03):
So that
was really awesome.
He's such a cool guy.
Hugh Williams (35:06):
Yeah, wasn't that
amazing?
I I love talking to Jonathan.
I could talk to the guy forhours.
I hope all of our listenersenjoyed it as much as we did.
Hannah Clayton-Langto (35:13):
Jonathan,
if you're listening, thank you
again.
Um and unfortunately, I didn'tmanage to bring up the Tinder
jet, but listeners comment onthe episode if you want a photo
of Hugh next to the Tinder jet.
Hugh Williams (35:25):
The only thing
I'll tell you, Hannah, is it's
sort of a, I guess it's a realjet, but I think it didn't have
engines.
It was kind of like a it waskind of like a prop for one of
the best parties I've ever beento.
Um something else.
Hannah Clayton-Langton (35:37):
Hang on
a minute.
So basically, what you'retelling me is when you hire only
software engineers, nothardware engineers, you get a
jet that doesn't fly.
Hugh Williams (35:44):
That's pretty
much it.
That's pretty much it.
But you do get good parties.
Hannah Clayton-Langton (35:47):
Oh dear.
I think that's probably thefirst time someone's ever
recommended inviting a bunch ofsoftware engineers to a party,
but we won't get into thatbecause um we need to wrap the
episode.
So, listeners, if you like whatyou heard, you can like and
subscribe to this podcastwherever you find your podcast.
You can find us attechoverflowpodcast.com.
Hugh Williams (36:05):
We're also
available on LinkedIn where
we're posting tons of reallyinteresting content and also on
X and Instagram.
Hannah Clayton-Langton (36:12):
So we'll
see you next time for another
episode.
Uh, see you then.
Bye you.
Bye Helen