00:00without any further delay I will
00:02introduce to you Michael Seibel the CEO
00:04of Y Combinator the founder of companies
00:07like justin.tv and twitch and Socialcam
00:10to begin what is going to be a deep dive
00:13into product over the next several
00:17Michael so before I begin its kind of
00:30had a conversation with Jeff and I and I
00:32wanted to say a couple of things about
00:35my experience at justin.tv and and
00:38twitch so what I will say is that we
00:42broke many if not all of the rules that
00:46are about to tell you at various points
00:49during our company the things that
00:54allowed us to survive were one our
00:58founding team was extremely technical
01:02Justin Emmett and Kyle all were amazing
01:08to work with and basically what I found
01:11amazing about them is they were not
01:13intimidated by any technical challenge I
01:17think that I would not be standing here
01:21privileged to work with them and so I
01:24think this is something that a lot of
01:25companies a lot of startups a lot of
01:28sort of founders don't truly understand
01:30like that fact allowed us to break a lot
01:33of rules the second is we didn't spend a
01:36lot of money we moved out when we were
01:3921 22 22 and 23 we lived in a
01:43two-bedroom apartment that apartment
01:48cost $2,500 a month we were each given
01:51$500 a month walking around money which
01:54technically is against the law because
01:55it was below minimum wage but who cares
01:59about laws and that was it that that was
02:04Emmett got his own bedroom Kyle and
02:07Justin slept in bunk beds I slept in the
02:09living room and sometimes on the balcony
02:11we just didn't spend
02:13much money that gave us a lot of ability
02:16to screw up and make mistakes and then I
02:18say the last thing that was kind of
02:20interesting I only realized later is
02:21that our ego was highly tied to our
02:25startup we were not doing a startup to
02:29have a cool resume item it was really
02:32the only thing we had done on our own
02:33and so I think at various points during
02:36the company when it looked like we would
02:37fail basically our startup failing was
02:40our life failing right it was like well
02:43this is the only thing you've done so
02:44far and so if it fails you get F on life
02:47and I think that we all had that feeling
02:50very internally and therefore we just
02:54couldn't really conceive of giving up so
02:56I think more than anything I want to say
02:58in the rest is presentation those were
03:00the three things that saved our company
03:02made our company work and strangely I
03:04don't even think if you take one of
03:07those things away any one of them we
03:09would have died so this isn't one of
03:11those things where it's like oh you can
03:12grab for one or two and that's pretty
03:13good we needed all three or else game
03:16over so as I get into product I'm gonna
03:19tell stories from justin.tv from really
03:24early days at which when I was still
03:25there and then also from a YC company
03:29from a couple batches ago named poppy
03:31it's a company that I've advised since
03:33so invested in did YC great founder name
03:36off knee and weirdly I just feel like I
03:40needed to do a case study outside of my
03:41own story somehow it's gonna help share
03:44these lessons a little better so I
03:48always like to start with what problem
03:51are you solving because when I'm pitched
03:53by founders most often they just want to
03:55tell me what their idea is what they're
03:58gonna do what their product does I think
04:00what's interesting is that like
04:03oftentimes they don't even know why they
04:07don't know what's the problem that they
04:09expect to be solved at the end of what
04:11they're doing now I think that for some
04:13businesses is totally fine
04:15right I think that especially if you're
04:19early on especially if things are still
04:20in project phase whatever but I think at
04:23some point pretty early on you have to
04:27what are we doing and what do we expect
04:30the result to be so at justin.tv the
04:34first thing the problem we were solving
04:35was entertainment we were making TV
04:37shows Justin was the first one broadcast
04:39his live 24/7 this was to be a TV show
04:41so actually pretty easy for us to
04:43understand whether that was working or
04:44not is anyone watching right that's the
04:46problem we were solving people watch TV
04:48shows no one was watching so we didn't
04:50solve the problem then when we pivoted
04:52to an open platform the problem became
04:54can we let anyone broadcast live but
04:59that was problems trying to solve anyone
05:00can broadcast live on the Internet and
05:03once again once we understood that it
05:05was very easy for us to judge whether or
05:08not someone could do it we had this open
05:09platform as anyone using it but I think
05:11that like that was key to what we were
05:13doing and then sometimes when I talked
05:14to founders there's something they want
05:16to do in the world there's a problem
05:17that they're kind of vaguely interested
05:19in or there's an idea that are vey
05:20interesting but they really haven't
05:22nailed down what's the actual problem
05:24we're solving if you don't know the
05:25problem you can't know whether you
05:28the first thing I asked founders can you
05:30state the problem clearly in two
05:31sentences if you can't you don't know
05:34the problem right in fact it's really
05:35only take you one sentence so if someone
05:38asks you a problem you're solving and
05:40you find yourself delivering an essay
05:42you're doing it wrong
05:44- have you experienced the problem
05:46yourself this is not always required but
05:50is certainly helpful I've met a lot of
05:53founders who are trying to solve a
05:55problem for someone else who they've
05:57never met never talked to and don't
05:59truly know whether that person exists in
06:01the world and so all things being equal
06:04this is a great hand that you're on to
06:06something well at least one person has
06:08had this problem before the next one's
06:12can you define this problem narrowly
06:14what's interesting is when you get
06:16started you can't really solve this
06:20problem for everyone who has it
06:21so when justin.tv first started we
06:24couldn't let anyone broadcast live video
06:26you had to have a laptop you had to have
06:28go to internet connection you had to
06:30have a webcam there are all these kind
06:33of things you needed and so can we
06:36actually now talk about all right we
06:38want to make live video for everyone
06:40but let's talk about the people that we
06:42can address first who can we help first
06:44and I think oftentimes founders kind of
06:47want to skip that step they want to
06:49solve the mega problem like I want to
06:51cure cancer I'm only talking about when
06:53everyone's cured as opposed to like what
06:56can we address immediately how do we get
06:58the first indication that this thing is
07:00working and then the last one is the
07:04problem solvable so here's what I'll
07:07bring up with poppy so poppy is a
07:09company that's essentially uber for
07:11babysitting they make it really easy for
07:13babysitters I'm sorry for parents you
07:15need babysitter's to get babysitters
07:20poppy is a very interesting company
07:22because you need babysitter's for a lot
07:23of different types of things some people
07:26need a babysitter five days a week while
07:30the parents at work right that looks a
07:32little more like a nanny some people
07:33need a babysitter whether it's an
07:35oh I have you know I have a medical
07:40emergency and I need a babysitter right
07:42now because I need to go to the hospital
07:43some people need it because there was
07:45miss planning oh I thought well housing
07:48was gonna be at this time and it wasn't
07:49I thought I was gonna be here this time
07:50it wasn't I need a babysitter some
07:52people need a babysitter because they
07:54have an infant right and so this baby
07:56students have a bunch of skills some
07:58people need a babysitter to watch their
07:5915 year old to make sure they don't get
08:00you know out of the house different
08:02skills and so what's interesting is that
08:05if you just start with oh we're gonna
08:07help people get babysitters it's not
08:11really good enough to understand what
08:13you can address right away right which
08:16one of those use cases do you want to
08:17address if you were to state the problem
08:19more narrowly though let's say we wanted
08:22to start out with infants right we want
08:25to make it easy for parents to get
08:26babysitters for infants then we can
08:28really ask the question is the problem
08:30solvable I think one of the things that
08:32poppy discovered when operating their
08:34business is that the level of skill that
08:36you need for a parent to trust you with
08:39their infant when they haven't met you
08:41is very very high so the idea that
08:44you're going to have a rotating set of
08:46people you haven't met watch your little
08:48baby hard very very hard they have to be
08:52very skilled and then on the flip side
08:54uber model only works because there's a
08:57whole bunch of basically replaceable
08:59people with a common skill well it turns
09:03out the people who've got the skills to
09:04watch infants and make parents
09:06comfortable with that tend to have nanny
09:09jobs where they work lots of hours and
09:11tend to be paid fairly well especially
09:13in up-and-coming cities and so now we
09:16have this disconnect where it's like
09:18well we want to solve the problem of
09:19infant watching for moms but that talent
09:23pool who can solve the problem the
09:25supply of babysitters they might not
09:27exist problem might not be solvable and
09:30so going through this exercise in real
09:34time like with your products out in the
09:35world you should be thinking about these
09:37things you should be thinking how have I
09:39narrowly defined the problem I want to
09:40solve first and you should be always
09:41asking yourself is it actually solvable
09:44I think a lot of founders just don't
09:47want to think about this because it's
09:49hard it's hard to think about who you
09:50want to talk to first it's hard to
09:52understand oh maybe I can't solve that
09:54problem I have to move on Abney was a
09:57mother of two kids and she was really
09:59pissed she couldn't solve this problem
10:00because it was her problem but it's
10:03turned out to be very very hard problem
10:05to solve you know young infants
10:07on-demand babysitters all right the next
10:11question I always ask is who is your
10:14customer and really you don't understand
10:19the problem you're solving until you
10:20understand who you're solving it for a
10:22lot of times people just want to say
10:25everyone everyone's the customer right
10:27and that seems like it makes sense in
10:29some cases right if you're building a
10:31social network or a search engine right
10:33everyone uses those things now what I
10:35will say is that in almost all of the
10:37products that everyone uses now there
10:39was a time when almost no one used them
10:41and the creators of those products had
10:44to figure out who is the ideal first
10:46customer and so if you don't have a good
10:49answer to this question you're gonna be
10:52lost do you have no idea who you should
10:54talk to to ask them whether this problem
10:56has been solved and you have no idea who
10:58to talk to to figure out who this
11:01product is for and I'd be surprised the
11:04number of founders who are just building
11:08they were writing a creative novel where
11:11it's just a product of their own brain
11:13and no interaction with anyone on the
11:15outside and it's not even problem that
11:18they're solved that they've experienced
11:19themselves don't do that don't be one of
11:22those founders like you can talk to your
11:24users you just have to figure out who
11:27they are the next question I often ask
11:30is how often does your user have the
11:34problem what's so surprising is when you
11:37talk through startups with people
11:39sometimes they choose problems without
11:42quite understanding who the user is or
11:46the frequency of the problem so give you
11:48an example a lot of people will come to
11:50YC and a popular idea back in the day
11:53was to build a car shopping website now
11:57if you guys have been on car shopping
11:58websites especially about five years ago
12:00they all basically suck they're they're
12:03hard to use they're not very transparent
12:05you kind of want to have this almost
12:07Tesla experience of just buying a car
12:09but they never actually work out that
12:11and what's interesting is that like a
12:13lot of founders come back to this
12:14over and over and over again and they
12:16always think that their customer is the
12:18person buying a car now the reality is
12:21is that when you go buy a car assuming
12:23it's not a complete lemon you typically
12:25keep that car for seven years so what
12:28happens if I told you I'm going to
12:29create a start-up and if I home run with
12:31my customer if my customer loves
12:34me they're gonna come back seven years
12:35from now that's hard it's very hard it
12:41turns out a lot of the car buying
12:42websites are not built for the person
12:45who's shopping for car because that
12:47person doesn't have a problem very often
12:49they're actually built for the person
12:51who's selling a car that person has a
12:54problem every day every day the
12:57dealership has to hit their numbers and
12:59so you don't see as a customer how that
13:03product helps the real customer the
13:07person who's trying to sell cars and so
13:09by doing this analysis I really try to
13:12push founders in understanding who is
13:14getting the most value out of this
13:16product and it's really helpful if
13:18you're trying to help someone with the
13:19problem they have frequently
13:21if you think about the products that you
13:23use on a daily basis they tend to be on
13:26the front screen of your phone you tend
13:28to use them without even thinking they
13:30become almost extensions of you if you
13:32think about apps that you've installed
13:34and then they're kind of on the third
13:35page in the back or they're buried on
13:37the second page is some folder those
13:40tend to be the ones you don't use very
13:41often hopefully they don't need you to
13:45use them very often or else they're
13:47probably not very good businesses the
13:50next question I always ask is how
13:51intense is the problem I find a lot of
13:55founders think they have a good idea but
13:57they don't do this frequency and
13:59intensity analysis and so if you have
14:02both an infrequent and low intensity
14:05problem that you're trying to solve
14:07you're gonna have a problem getting a
14:09lot of customers even interested in
14:10talking to you um all things we go if
14:13you graph problems it's nicer for them
14:15to be higher intensity higher frequency
14:17let's think about a company like uber
14:19for example usually when you are
14:22somewhere and you need to go somewhere
14:24else it's a pretty intense problem I
14:26have to go to work I have to go to the
14:28doctor I have to go pick up my kids
14:30they're so intense you might have bought
14:33like a twenty thousand dollar car to
14:36help you do those things right so it's
14:38an intense problem and then when you
14:40think about frequency how often do you
14:42move more than a mile more than walking
14:45distance every day happens a lot and so
14:48if you think about that even though you
14:51look at the taxi market and you say ah
14:53taxis unit before Hoover taxis that's
14:55not that big of a market if you just
14:57look at the customer you say high
14:59intensity problem that happens very
15:00often there's probably a good business
15:02here the last one is are they're willing
15:07to pay so many founders who come into YC
15:10their first thought is completely wrong
15:13on this front their first thought is I
15:15need to give it away for free because
15:17that's the only way I'm gonna get users
15:18one of the things that I always push
15:21them to do is think about it this way if
15:23you want to know whether you have a good
15:25product it's a lot easier to make it a
15:28little bit harder for your users to use
15:30it and then see if they use it anyways
15:35really intense problem and you say well
15:37it's gonna cost 100 bucks for the person
15:40with extremely tense problem they
15:41probably think that's a deal if you have
15:44extremely and if you don't have
15:45extremely tense product problem you
15:48charge zero dollars you can have a bunch
15:49of users who come in who don't really
15:51have the problem but they're just trying
15:53something out if you try to learn from
15:55them on how to improve your product
15:57oftentimes they'll lead you astray so
15:59strangely starting with a higher price
16:01or a price is almost always better than
16:04starting free almost always better and
16:08if you have to start free you need to do
16:11this analysis of how do you talk to the
16:13users where the problem is actually
16:14intense I talk to the users who are
16:16using a product frequently in production
16:18for real-world purposes as opposed to
16:21the hobbyists talking your customers is
16:23good but talking to the wrong customers
16:25very very very bad I've seen a lot of
16:29companies that are basically hijacked by
16:31bad customers especially companies where
16:35there's real-world costs so for example
16:40you know if you have a company like
16:41poppy there's real-world costs in
16:43recruiting managing and working with all
16:46of these baby sitters and if you have a
16:49company if you have a customers trying
16:51to basically take advantage of that
16:52system being laid being non-responsive
16:55being rude to the baby sitters that's
16:58not going to help you run your business
16:59and you'd be surprised at how many
17:01hijack customers there are out there the
17:04last question I always ask people is how
17:05easy is it for your customers to find
17:07because inevitably you're gonna need to
17:09reach them and what's interesting I look
17:11at the the last batch there are two b2b
17:14companies one b2b company was here in
17:16America and it was very easy for them to
17:18find customers they could basically go
17:20to LinkedIn find their customers on
17:21LinkedIn find the name addresses email
17:23them they get email thousand customers a
17:25week another company doing b2b was in
17:28China and interesting enough reaching
17:32out over email and the b2b context in
17:34China just isn't - well done practice
17:36getting access to people's email
17:37addresses is actually not very hard not
17:39very easy and so strangely they had this
17:42challenge of a relatively simple
17:44business to explain a real intense
17:46problem that happened often if they had
17:48no way to reach their customers
17:49and to basically invent new ways to do
17:51it and so I often want to ask this
17:54question because if your customers are
17:56ridiculously hard to find
17:58you better have a solution for that
18:00upfront you can't build the whole thing
18:02and expect for them to find you and so
18:06often times you had a situation where
18:08someone's trying to build a product for
18:11either an imaginary customer or a
18:12customer who can't really hope to use
18:14the product I'm trying to get water to
18:18people in the middle of a desert in the
18:22Sahara all you need to do is download my
18:24app and go online and then they can put
18:28their GPS location and then we'll
18:30deliver water to them that's not gonna
18:32work but you'd be surprised at how many
18:34people just don't think through those
18:36logical steps all right next up does
18:40your MVP actually solve the problem that
18:42you want to solve this one is so
18:45hilarious how often it comes up because
18:49in the process of building an MVP things
18:52just go weird and squirrely so you had
18:54this problem and then you started
18:55building it and then you talked to other
18:57users and then before long you're
18:59launching something and then you realize
19:02it doesn't actually do the thing that
19:03you promised and even the thing that you
19:04want to do so part of your process of
19:07building the MVP it's really helpful to
19:10do these pre steps first it's really
19:13really really helpful because then you
19:15can always gut-check yourself on am I
19:17actually solving the problem the other
19:19thing is that it's really helpful to
19:21build your MVP quickly typically the
19:24longer it takes the more you're gonna
19:25have MVP and problem drift or customer
19:28drift if you decide to only build your
19:30MVP in two weeks it's a lot easier to
19:33stay on task and make sure actually
19:34solving that problem for that customer
19:36the way you test this by the way is you
19:39give your product to customers like you
19:43you have to do that that is a required
19:46step what I find interesting is that a
19:48lot of people think of their product as
19:51a painting as something that could be
19:54appreciated as a piece of art as
19:56something that even if it's appreciated
19:58by one person is special that's not what
20:01you're making products are not paint
20:03they're not art if users don't find
20:06products useful then the products are by
20:09definition not useful and they're a
20:11waste of your time to build and I think
20:13a lot of people want to be artists the
20:17startup world is very unforgiving to
20:20artists and I think that interestingly
20:22after the fact a lot of people are
20:24painted as artists right like Steve Jobs
20:26is painted as this like magical artist
20:28right and the other day he had to figure
20:30out how to get make a phone that
20:32millions of people would buy if only one
20:35person bought the iPhone he would be
20:36seen as a failure so the definition of
20:40art is it only has to be appreciated by
20:42one or maybe even none that's not the
20:45appreciate that maybe just the Creator
20:46that's not the definition of a
20:48successful product so this is what you
20:51should always be got checking does every
20:52piece solve the problem the number one
20:54problem with this question is that it
20:56hurts the answer hurts you're gonna find
20:58that a lot in startups where the answer
21:00hurts you know doesn't solve the problem
21:02but as long as we don't talk about it
21:03maybe nobody knows it doesn't solve the
21:05problem a lot of the answers inside of
21:08startups are feel that way which
21:10customers should you go after first a
21:12lot of founders are very confused by
21:16this question what I find interesting is
21:20just like the instinct is to go after
21:23customers by making the product free for
21:26some reason I find a lot of people think
21:29that their instinct should be to go
21:30after the hardest customers first almost
21:33as if it's like a proof like if I can
21:35get this impossible person to use
21:36something then like it'll be easier I
21:39know that I've made something good I
21:42like to start from a different point
21:44it's an MVP you know you've made
21:46something bad like that's the definition
21:49of MP it's bad so the real question is
21:51like how do you find people who are
21:53willing to use a bad product right there
21:57have to be the most desperate the most
22:00desperate and so a lot of times I talk
22:03to founders I really pushed them towards
22:05who are the most desperate customers and
22:08how do you talk to them first that's
22:10what I define as easy desperate if
22:13you're having like a you know if you're
22:15trying to sell a simple piece of
22:17thousand dollars a month and you're
22:19engaged in a six month conversation with
22:21a company that's not a desperate company
22:24move on in fact when you're doing
22:26enterprise sales early as a start-up
22:29like you're looking for even more
22:31desperate customers just because
22:33literally takes so long to sell them so
22:35if you don't feel like you're dealing
22:37with desperate people if you feel like
22:38you've dealt you're you are trying to
22:40get impressive customers who aren't
22:42desperate you're probably doing it wrong
22:45literally the number one thing I often
22:47tell found is just like whose business
22:50is gonna go out of business without
22:51using you which people out there are not
22:54going to be able to get to work or to
22:56watch their kids how do you find the
22:58people who are just literally are
22:59screaming for something like this and
23:03then how do you talk to them and not
23:04talk to your friends I had a bunch of
23:08friends who were using social cam right
23:11my company was doing video for for
23:14sharing with friends and they weren't
23:15really using it they were using it
23:16because it was like my app and there
23:19were friends with me I literally had one
23:20friend it was like super honest about
23:21this Steve the CEO of Reddit when we
23:24sold Socialcam he literally said thank
23:25God now I can delete this app from my
23:27phone so the perfect definition of
23:33someone you should not be trying to get
23:35product feedback from right and so he
23:38didn't have the problem we were solving
23:40many of your friends won't have the
23:42problem that you're solving make sure
23:44you find D and by the way the kind of
23:47community of start-up people and/or
23:49investors usually don't have the problem
23:52that you're solving so if you're using
23:53investors as a trigger for am i solving
23:56the right problem or like do they find
23:59this useful it's almost never the case
24:01almost never the user of a product that
24:04comes into I see and so ignore your
24:07investors ignore your friends like they
24:09will lead you 100 percent astray out of
24:12good intentions they'll try to be
24:14helpful well at you know I've never
24:16lived in the Sahara I've never been
24:17thirsty but maybe it should work like
24:21this right live like horrible run away
24:24run away once you start having customers
24:26I think it's a very helpful exercise to
24:29try early to identify
24:31by bad customers these are people who
24:35are blasting your support these are
24:37people who are constantly
24:40constantly complaining
24:42my co-founder Justin he had a company
24:45that was basically on demand personal
24:47assistant and he was the first one who I
24:50met who actively fired a customer is
24:53basically uber for personally support
24:55systems called exact and literally a
24:57customer would have the exact do
25:00something like like crazy like something
25:03you couldn't do right like we organized
25:04my house the way I want things organized
25:07and I'm not gonna tell you I want them
25:09organized or go shopping for me but I'm
25:12gonna critique every single like piece
25:14of fruit and vegetable that you picked
25:15out she was like completely unrealistic
25:17expectations and so after refunding the
25:19person four times for four different
25:20tasks the person did a fifth task on the
25:24product right because he's getting a
25:25bunch of value for free and Justin Cosmo
25:27says you're fired you can't ever use
25:28product again like look for those people
25:32because if you aren't delivering
25:33anything of value there will be people
25:35trying to exploit that value and some
25:39might be doing it not out of the
25:41goodness of their heart so don't let
25:44these people lead you astray we talked
25:47about this don't discount now here's a
25:51caveat on discounting Parker from
25:55zenefits came to I see a couple years
25:57ago and he gave this great talk about
26:00enterprise sales and zenefits his
26:03product that's given away for free so
26:04it's actually kind of a interesting
26:05enterprise sale and one of the things he
26:09said that really got to me was that
26:12there are ways to convince organizations
26:17basically you can structure discounts
26:20and incentives into your sales pitch if
26:23you basically understand what value are
26:25getting back so his example was he would
26:30try to sell to a company to switch on to
26:32zenefits for their health care and he
26:34would say look because of this third
26:37party let's just say AWS has given us a
26:40discount who knows why right we just
26:43bought dedicated instances so now
26:45we have 40 percent lower AWS bills so we
26:48can actually pass on some benefit to you
26:50but only for the next 30 days now I feel
26:54horrible even telling you this because I
26:56want you to take as much time as you
26:58need to buy my product I would just hate
27:00if you bought it on the 31st day and I
27:02couldn't give you this discount now this
27:04isn't a let me give this away for free
27:07like because I'm afraid people won't use
27:09it this is a very structured process
27:13that he did he basically incorporated a
27:15deadline based on some third party
27:18providing a benefit to the customer and
27:22he knew that when he said this to the
27:24customer every time that this was talked
27:26about internally the deadline would be
27:28brought up and the discount would be
27:30brought up and suddenly this has now
27:32become not a way to be afraid right - oh
27:36I'm not sure how many customers minutes
27:38make it free became a way to speed up
27:40the process and his discount was just
27:42baked in like he just priced the product
27:4415% higher so it's like literally like
27:47that is the way to do it
27:50the way not to do it is I'm afraid no
27:52one's gonna use it so I have to talk not
27:53charge any money okay
27:56the next is how to set up metrics how
27:59many of you are using Google Analytics
28:02as your primary metrics products raise
28:05your hands okay you are doing it wrong
28:10yes so setting up metrics is something
28:15that's like super important very early
28:19in your company because it's how you
28:20know whether your product is being used
28:22or not and it's one of the number-one
28:24sources of new product ideas and
28:26inspiration so Google Analytics I would
28:29say is a great product for knowing how
28:31many people came to your website today
28:32and how many pages they viewed which
28:35used to be relevant and it's not really
28:37relevant anymore and where they came
28:39um what it's not a great product for is
28:42identifying what people's actions were
28:45when they were using your product did
28:47they click this button did they see this
28:48screen how long were they on the page
28:50for before they did something else did
28:52they leave something in their cart for
28:54all of those things you want an events
28:56based metrics product
28:58Mixpanel amplitude heap I think we've
29:01funded like 50 of them they're like 100
29:03of them out there you should be using
29:04one of them if you're not you can't be
29:07sophisticated at building your product
29:10this is just kind of a prerequisite so
29:12get on it and this goes back to the
29:15early thing that I mentioned which is
29:16technical teams for a technical team
29:18implementing Mixpanel is ridiculously
29:20easy for a non-technical team it's
29:22basically impossible this is just one of
29:26the many advantages of having a highly
29:27technical team you actually know what
29:29users are doing without this you're just
29:31missing a huge part of what you need to
29:34know the next thing and Suhail from
29:37Mixpanel gave a great talk about how do
29:39you set up mix panel one of the
29:41challenges of setting up mix panel is
29:43the second that you're sitting there
29:44saying I want to track what my users are
29:45doing you can come up with like 150
29:48things your users can do with your
29:49product and you want to track all of
29:50them that's often a mistake if your
29:54analytics product is got too many
29:56analytic sitting in it in the beginning
29:58it will be hard to use and part of what
30:01you're doing if you've never used
30:02product like links panel before is
30:03learning how to use it and most
30:05importantly teaching your employees and
30:08your co-founders how to use it because
30:10this product should be a product that
30:11everyone in your company understands how
30:13to use is everyone your company should
30:14understand how the product is
30:16functioning this is not something that
30:18like the CTO uses and creates reports
30:20from this is interactive product that
30:22so to start pick five to ten simple
30:26let's take Instagram as an example right
30:29if I went to pick five to ten simple
30:31stats for Instagram let's say open the
30:33app trade in an account took a photo
30:36applied any effects share the photo
30:40that's probably all I need in the
30:42beginning right I mean the number one
30:44mechanism for Instagram is taking a
30:46photo and sharing it I can track that
30:48I'm pretty happy the other thing that I
30:52will warn you about is that if your
30:54product is good the naming conventions
30:57for these stats are going to become very
30:58important because one day there will be
31:00a hundred or even a thousand stats you
31:02track so think a little bit ahead of
31:04time and don't name something something
31:06that only you'll understand make sure
31:09that you if your company's good many
31:11many people have to look at the
31:12stats make measurement a part of your
31:15product spec oftentimes when I talk to
31:18founders they say we built it on this
31:20release and we'll add the measurements
31:21some point in the future I don't
31:24understand how that works you build
31:26something you want people to use but
31:28you're not incorporating the measurement
31:29that tells you whether people are using
31:30it that doesn't work building
31:35measurement is part of a product spec so
31:37when you spec out a product you better
31:39spec the stats you expect to be tracking
31:42and you should also spec the stats that
31:44you think are going to improve when
31:48you're building that product that should
31:49be part of the spec it should be part of
31:51the first release otherwise you're
31:53flying behind and this is just countless
31:57times a justin.tv this is screwed us
32:01part development cycle so justin.tv was
32:04and twitch was three yell kids and one a
32:09mighty kid yell probably the most
32:11productive skill you're taught is how to
32:13argue with other yell kids and so the
32:16number one way to get products developed
32:18at justin.tv was to win an argument with
32:21kyle disliked this so much that he
32:24actually switched his sleeping schedule
32:26so that he wouldn't have to be involved
32:28in these arguments so we were awake from
32:30about 8 a.m. to about 12 midnight he
32:33would wake up around 11:00 midnight and
32:35write code all night and then go to
32:39sleep in the morning so he wouldn't have
32:40to argue with us on what stupid thing to
32:42build one of the classic arguments at
32:45justin.tv that lasted approximately
32:47three months was the background color
32:51for the original site so the original
32:53site is just one page Justin wanted a
32:56black background I wanted a wood grained
32:59background three months of bait we
33:04settled on changeable backgrounds so
33:09there were five background options
33:12clearly idiotic like I said we made many
33:15of these mistakes um we didn't actually
33:17really learn how to do product
33:19development cycle until we failed at it
33:21for about five or so years
33:25and during that time this is what bad
33:29product-moment cycle looks like one we
33:31would release every three months for a
33:34web only product that is horrible second
33:39we would have a product meeting and we
33:41wouldn't write anything down right it
33:44was just four of us can't you remember
33:46you're an idiot if you can't remember a
33:47conversation before people had right and
33:50if you forget something just ask one of
33:52the other four people in the room right
33:53No so as a results during the first
33:57month of the dev cycle we'd all go off
34:00working on slightly different versions
34:02of the thing we wanted to build because
34:03we didn't write down this back then at
34:06the end of that month we'd come together
34:07and we'd be like oh wait this isn't
34:10we're not really building all the same
34:12thing and then we'd have another product
34:14meeting where we didn't write anything
34:15down and go off and build again for
34:19another month at this point right two
34:22months in we probably have about three
34:23weeks of productivity and about five
34:26weeks of just stuff that's gonna have to
34:27be thrown away at this point in we kind
34:31of come back together and realize that
34:32we're not 2/3 of the way done through
34:34this sprint were less than 1/3 of the
34:37way done and we're starting to get sick
34:39and tired of this feature that we're
34:40building so then we basically say
34:42alright slash-and-burn
34:43let's just make a shitty version of it
34:46and we take another month to do that now
34:49we've worked on this product for three
34:50months if you had any good or new or
34:52interesting ideas during that three
34:54month period of time you were told we're
34:57already working on something else so
34:58your ideas are worthless just write them
35:00whatever we're working on this thing
35:02right now at the end of the three months
35:03instead of wanting to iterate we were
35:06sick of the da TM feature we just spent
35:08three months building poorly so we would
35:11launch it and if it wasn't used right
35:12away we would come up with some new
35:14brainstorm on some brand new feature
35:16that would rescue the company this is
35:18the wrong way to run a company it was
35:21absolutely horrible I was talking to
35:23Jeff earlier the major product decisions
35:25that justin.tv made that carried through
35:27to twitch - today was chat on the right
35:30video on the left we decided that in
35:332006 it is the same way 2018 the vast
35:37majority of the products issues we made
35:39and never saw the light of day because
35:41they went through a process like this so
35:43if your process revolves around arguing
35:47revolves around not writing speck
35:49revolves around long dev cycles you are
35:52doing it wrong you are 100% doing it
35:55wrong what I'm gonna give you is a model
35:57of how we figured out how to solve the
35:59problem steal as much or as little of
36:02this as you want but understand that if
36:04you have any of the symptoms I'm talking
36:05about you need to solve them or else
36:07your company is just going to be much
36:09less productive than it could be first
36:12you need to actually have a number that
36:14you track that reflects how good your
36:17company is doing almost always if you
36:21ever are going to charge money to your
36:23customers this number should be revenue
36:26almost always if you are never going to
36:30charge your customers money like
36:32Facebook then maybe it should be a usage
36:35based metric like how often do your
36:37customers come back every day like DEA
36:39use it is almost always one of those two
36:43usage if you will never charge the
36:45customers money if you will charge the
36:48customers many people will invent
36:50reasons why these two metrics don't
36:52apply to their business one percent of
36:55them might be right 99 percent of them
36:58whatever this KPI goal is make sure
37:01you're measuring it make sure that
37:02everyone in your company knows what this
37:04goal is every day helpful to put it on
37:06some screen somewhere if an investor
37:08asked you what your KPI is you not only
37:09should be able to say what it is
37:11sheíll to say what the metric is you say
37:13where the metric is now where it was
37:15three months ago where it was when you
37:16started this is kind of table stakes the
37:21next thing that we would do is as the
37:25kind of product person I would come into
37:27the meeting and I would say this is the
37:28KPI we're looking to improve this this
37:31cycle at Socialcam the top level KPI was
37:35da use and the three ways that we
37:38thought we contributed to da use was
37:39either new users retention of users and
37:43new content created those three things
37:45so every cycle we ran one of those three
37:48numbers moving that number so that the
37:50right direction was the goal and we'd
37:54brainstorm for us would take a couple
37:55hours and it was a real brainstorm it
37:58wasn't the brainstorm we're like you say
38:00what about this and your co-founder says
38:02that's a dumb idea that's not a
38:04the real brainstorm is that any idea
38:07that's stated is written on the board
38:09the cool thing about these brainstorms
38:11is that everyone's computers were always
38:14so if you had an idea or you had a
38:17thought you could always just go in and
38:19check the metrics and see like oh is
38:20that right or is that wrong you'd be
38:24surprised at how much value there is and
38:26seeing your idea on the board not
38:28everyone's gonna get to have built what
38:31they want to have built but the fact
38:33that your idea was considered and added
38:34to the board actually makes people feel
38:36a lot better than otherwise people feel
38:39horrible when their ideas are shot down
38:41CEOs their job is to make their
38:44employees not feel horrible all the time
38:45sometimes I think CEOs think their job
38:48is to shoot down ideas it's not it's not
38:52gonna help you at all I mean everyone by
38:54the way in our company participated in
38:55this brainstorm at that point that was
38:57for people so easy to do the next thing
39:00was we did what's called easy medium
39:03so our brainstorm was actually typically
39:05split up into three categories new
39:07features or iterations on existing ones
39:09bug fixes and/or other maintenance and
39:13tests a B tests we want to run we have a
39:16whole board filled out with ideas on
39:18these three categories and then we go
39:20through and do what's called easy medium
39:21hard for us hard mint it would take one
39:24engineer most of the dev cycle to build
39:26medium typically meant to take a day two
39:28days easy means we could do multiple in
39:31a day this is extremely important how
39:34many of you in this room do not know how
39:35to write code raise your hand there we
39:38go I am one of you it's extremely hard
39:41if you don't know how to write code to
39:43figure out whether your idea is easy to
39:46build or hard to build that's something
39:48that you actually learn as a skill over
39:49time and this process basically is the
39:53process that can help educate you it
39:55turns out that easy ideas get built way
39:58faster way more quickly than hard ideas
40:00and it turns out that most hard ideas
40:02can be restated as an easy idea if you
40:06just understand what
40:07bits of your heart idea are both useless
40:09and hard and most of the time there are
40:12useless and hard bits and hard ideas
40:14that can just be removed and so for us
40:17this was educating everyone the team and
40:19for us we had a cross-functional team so
40:21someone might not realize that this is
40:23really hard on the video system it might
40:25be an easy web feature but hard on the
40:26video system or vice versa so this was
40:28basically educating everyone a team on
40:29what's easy medium hard it also created
40:31an objective standard by which to start
40:33thinking about these ideas instead of
40:35just based on the argument ability of
40:37the person delivering them it was like
40:39well your idea is like really freaking
40:41hard and seems like it wouldn't move the
40:43numbers that much based on Mixpanel
40:44whereas like this other person's idea is
40:46super easy and probably can move numbers
40:48a lot the next thing we would do is we
40:52decide hard first so we look at all the
40:55Harz and we say which hard is gonna
40:57impact the KPI the most and then we move
41:01to Easy's and then we move to one with
41:04moved at mediums and then we moved to
41:05Easy's what was interesting is that like
41:07just with the ideas on the board and
41:09with easy medium hard a lot of the ego
41:11was removed from the debate because one
41:14you knew your idea has been considered
41:15and two you'd sent some objective
41:17measure about how hard it was and three
41:20because the board has a bunch of ideas
41:21on it now it's probably pretty easy for
41:24you to find an easy idea that you really
41:26like and so you're just gonna be excited
41:29that that's probably gonna get in and
41:30you're really hard idea that's fine if
41:33it doesn't the next step is you have to
41:37write the spec this is where everyone
41:39 up the meeting might be going on
41:40for four hours now and this is the step
41:43no one likes you actually go through and
41:46you actually write down what do we mean
41:48by we're adding video filters to
41:50Socialcam what do we mean by were
41:53allowing people and justin.tv and twitch
41:56to chat with one another what does that
41:57actually mean how is it going to work
41:59this is really important and once this
42:03is done you can then distribute tasks to
42:05the team now we would run these cycles
42:07every two weeks at Socialcam because
42:09back then summing to the AppStore took
42:12longer if you're doing a pure web
42:14product you can run these cycles once a
42:16week the rule that we had to make the
42:19team not hate these really
42:21meetings is the only meeting we had this
42:23is the only meeting and so it might take
42:25two hours it might take six but for that
42:27to occur period of time there were
42:29northern meetings in fact for me being a
42:32non coder my number one job was to shut
42:34the up because I could create a
42:41we're busy working we have this written
42:42spec everyone knows what they're doing
42:44if I have some brilliant idea during
42:46that two weeks right that'll throw the
42:49whole thing into chaos
42:50suddenly the written specs not important
42:52we're back to the drawing boards or
42:54we're changing things Yeti ID Etta what
42:56I had to realize is that every two weeks
42:58we do this over again so for that
43:00burning idea just wait two weeks
43:01and we're gonna have the meeting again
43:03and then we can get it in and turns out
43:05you're burning ideas probably wrong so
43:07it's totally fine to wait two weeks to
43:08try to kiss people do something wrong is
43:10totally fun as opposed to like having
43:13this cadence meant every two weeks we
43:15had success every two weeks if we built
43:18what we said we were gonna build we felt
43:20good and then that cadence meant that
43:22we'd go into the next cycle and do even
43:24more this cadence is extremely important
43:27because it's going to take you guys a
43:28long time to find product market fit
43:30maybe trying a lot of things we
43:32iterating a lot and if that process
43:34doesn't feel fun you're getting very
43:38this made the process feel fun because
43:40we had goals and accomplish them versus
43:43iterate a lot of YC companies a lot of
43:48founders in general will tell me our
43:51thing isn't working it's been two months
43:54it's time to pivot when I think about
43:57that statement it blows my mind right
44:00you're building a new product for a
44:02customer who might not have ever used
44:03the product before you're oftentimes
44:06exploring a problem that you only know
44:07to some degree or you've only
44:09experienced it personally what makes you
44:11think two months is enough time to know
44:14whether you figured something out what
44:16impressive thing only took two months to
44:18build so if you're not thinking that the
44:22process of coming up with a solution for
44:24this problem is probably the more like a
44:26two-year process you're doing it wrong
44:28if you are unsatisfied with significant
44:31progress in under two years you're
44:34probably doing it wrong it's going to
44:37take time you're doing something hard if
44:39it was really easy someone else would
44:41have done it so I define pivot as
44:46changing the customer or changing the
44:48problem this should be rare this should
44:51happen infrequently many times this
44:55means you should start a new company i
44:58define iterate as changing the solution
45:03it turns out you had the right customer
45:06you had the right problem your MVP was
45:08shitty and it didn't work
45:09we need a new solution it turns out
45:12maybe your MVP was great but it didn't
45:14solve the problem with a new solution it
45:16turns out you showed the product to your
45:18customers and they didn't want to use it
45:19even though they have burning problems
45:20we need a new solution often times I see
45:23this in Reverse people think solution
45:25first and when cut the customers they
45:28thought didn't like their product they
45:30try to find some other random customer
45:31who does he might even have a completely
45:34different problem and they tried
45:35shopping around their solution because
45:37they think their solution is the genius
45:39part I think the problem is the genius
45:42part I think identifying a problem that
45:44other people haven't figured out is
45:46worth working on is the genius part
45:48right Facebook wasn't first a social
45:50networking and Google wasn't first a
45:52search engines their genius was
45:54understanding that the people who came
45:56before them hadn't solved the problem
45:58and if they could solve the problem
46:00better they'd built huge companies their
46:03genius wasn't oh we built this cool
46:06thing let's just figure out who might
46:07want to use it wrapping up a little bit
46:10here I always tell the story about fake
46:13Steve Jobs versus real Steve Jobs a lot
46:16of people think that Steve Jobs is this
46:19person they should emulate but they have
46:21a false picture in their heads of what
46:22Steve Jobs was they think that like he
46:25dreamed perfect ideas out of his head
46:28and into the world and what's funny is
46:30that I think often times people look at
46:32the iPhone as perfect example of this
46:33but they look at their iPhone today your
46:38iPhone today is magical the
46:40first iPhone sucked in almost every way
46:44and they don't realize that Steve Jobs
46:48was just not iterating who just imagine
46:50your perfection minute one Steve Jobs
46:53was iterating at every step so I like to
46:56remind people what the first iPhone did
46:58first iPhone no 3G back when 3G was a
47:02standard feature so you have this great
47:04internet browser but you can only use it
47:06on edge which means it sucks
47:08right one carrier oh you don't have this
47:11carrier sorry switch carriers figure
47:15horrible battery life screen cracked all
47:19no App Store you can't even download
47:22other apps that was the first iPhone
47:24everyone forgets that iPhone so if you
47:27are the person in your company who is
47:29being fake Steve Jobs is saying the
47:32product has to be this way because what
47:34I said the customers everyone
47:36else you make the product the way I
47:38want it to be you're being fake Steve
47:41Jobs real Steve Jobs released a shitty
47:44MVP that was revolutionary but still
47:46fairly shitty and every year iterated it
47:49until you have the thing in your pocket
47:51right now which is pretty damn good real
47:54Steve Jobs iterates and talks to
47:56customers fake Steve Jobs just dreams
47:59and creates art don't be fake Steve Jobs
48:03okay so with all this I want to go back
48:07to the beginning what I said the
48:10beginning still holds justin.tv the only
48:12reason why I actually even know any of
48:14these rules is because we broke all of
48:16them the one thing that justin.tv and
48:18twitch had was a really strong technical
48:21team with high ego in the product and lo
48:24burn when we started figuring things out
48:28with twitch it was very interesting
48:30gamers had been on our product the whole
48:34time gamers had been streaming on
48:36justin.tv since almost the beginning at
48:38any given time there were twenty percent
48:40of our traffic for years we ignored them
48:43we ignored them we ignored them we
48:46ignored them ignored them they still use
48:48the product we didn't build features for
48:49them they still use the product they
48:51must been pretty desperate
48:52because they still use the product year
48:55after year the number one thing that
48:58changed when we started working on
49:01we started talking to them and what's
49:03weird was it's not like we were talking
49:04to other users and the only reason we
49:08didn't talk to any users we had this
49:09like crazy part development cycles we
49:11couldn't do that with talking users too
49:13so what we did in the beginning was we
49:15literally just sat down with these
49:17gamers and we said what did you want and
49:19what's funny is we didn't build them
49:21anything very special they were like oh
49:22like lag socks or the couple they wanted
49:25like little things what was great about
49:27it was they realized we were now going
49:29to build something for them and no one
49:32on the internet was building things for
49:33these gamers and they realized that when
49:35we said we're gonna build something it
49:37came out when was the last time that you
49:39talked to someone building a product
49:41that you like and you said can you do
49:44this and they did it was last time you
49:46suggested a feature to mark at Facebook
49:48and then the feature came out never like
49:51it's one of the magical things you can
49:53deliver as a startup is you can talk to
49:55a passionate user and then you can build
49:56what they want and then you can say here
49:58it is and they will fall in love with
49:59you even if those features are
50:01relatively mundane because let's switch
50:03today chat on the right video on the
50:05left the same product what was great
50:08about this process was by talking to
50:10them they realized that we were on their
50:12side we realized they were building
50:13something for them so they tell their
50:15friends that was the major change if we
50:19didn't have the technical team if we
50:22weren't sheep if our ego wasn't involved
50:24I never would have gotten to that point
50:27and if you look at the history of
50:29justin.tv in the first five years it
50:32went from being worth nothing to being
50:34worth about twenty four million dollars
50:36in the next three years it went being
50:38worth twenty four million dollars
50:39touring worth a billion like that's what
50:42software can do when you when you hit
50:43the right customer let's do a couple
50:46questions in the back
50:48so you've mentioned that product so the
51:05question is put it more generally should
51:10you be going free if your final idea for
51:12a product is to be free what I would say
51:14is this if your users are users who you
51:18never plan to charge then it's totally
51:21fine free to be free but if you do plan
51:24to charge them in some way it's really
51:26helpful to charge them as soon as
51:27possible because you want to know
51:29whether or not they're willing to pay
51:31and certainly if their business depends
51:33on it it's especially helpful to charge
51:36them so that's the measure that I would
51:39use and they're all kinds of little
51:40tweaks and and so and so forth but at a
51:43high level do you ever plan to charge
51:45them I charge them if you never plan to
51:48charge them you can plan to monetize
51:49based on ads which is really usually the
51:51way that you never plan charges you can
51:53monetize with ads if you're not going to
51:54monetize with ads you probably should
51:56start charging them all right next
52:18revenue if you're metric if your KPI is
52:22metric do that if your KPI is revenue
52:28and the number is zero should you still
52:30be tracking that as your top-line KPI
52:32the short answer is yes you should be
52:35depressed looking at that number every
52:36week too but that's that's the answer
52:39now let's be clear like I said with
52:40Socialcam right there are contributing
52:42numbers to that right and so whereas
52:45like DEA use was our top-line metric
52:47right we also the things we thought
52:49contributed to that new content new
52:52users retained users those numbers we
52:55can move so if you're in a sales type
52:57business your KPI number one revenue if
52:59it's zero that should you in face
53:01zero zero zero horrible but then you
53:04should ask yourself maybe your three
53:05other metrics are how many conversations
53:07did we have this week how many people
53:10are in contracting how people aren't
53:11onboarding right and those can be your
53:13three numbers that will if those numbers
53:16move you expect the revenue number to
53:18move and they can keep you motivated but
53:20your your top-line KPIs no like
53:24absolute no good we're a
53:27hardware company pre-launch and so our
53:31users our target market experience our
53:33problem five to nine hundred times a day
53:35and the intensity is can we pay our rent
53:37or not but we'd like to offer pre-sales
53:41as a way of getting the product to
53:43market do you guys have any tips on
53:45doing that do we have any tips on
53:51pre-sales what I would say is that there
53:52are many many tips on pre-sales I would
53:54email some founders who've done it
53:56before that's one of my best tip is to
53:58email them and ask them what they did
54:00the number one mistake I see with
54:02pre-sales is discounting the number one
54:05mistake is that basically especially
54:07hardware founders will misunderstand how
54:10much they need to charge so they don't
54:12lose money and so their presale becomes
54:14their death so I'd avoid that all right
54:17two more quickly and the way back what
54:29was the hardest part of having a slow
54:30burn well we were young so it wasn't
54:33hard at all we were living like we lived
54:35in dorms I think it's a lot harder now
54:39right I've got a kid I got a wife got an
54:41apartment got a car it'd be really hard
54:44now and so I think really the hard part
54:48about slow-burn is can you adjust your
54:51lifestyle if you've already leveled it
54:53up and if you haven't leveled up your
54:55lifestyle yet and you're still working
54:57you know if you're young and you're
54:58still working at a company that's paying
54:59you well not leveling up your wife style
55:01is a big way to stay ready to do a
55:04startup adding on the mortgages and the
55:06cars and the vacations makes it a lot
55:08harder I just know a lot of people who
55:10never could come back from that
55:14her back going from beta how do you go
55:26from beta to early MVP I don't really
55:28know what those distinctions are like
55:30all I know is are people using your
55:32product yes no like if people are not
55:35using your product get to the point
55:36where people are using your product
55:37extremely quickly once people are using
55:39your product there's all different
55:40labels for it beta pre-launch alpha blah
55:44blah who cares right it's just that's
55:47the dividing line like are people using
55:48a product the next question is are you
55:49actually solving their problem that's
55:51the next question not like are we
55:54following this line of launching things
55:56most YC companies will launch many many
55:59many many times so that progression
56:02isn't really that important are people
56:04using your product great BAM
56:08you're launched congratulations call
56:11whatever you want all right this is a
56:44great question and I have I'll have an
56:47unsatisfying answer so the question is
56:49basically how do we figure out what to
56:51build next here's my answer the reason
56:56why you have a part development cycle is
56:58that you can work on multiple things
57:00usually there isn't a right answer
57:02usually all of the things that you want
57:04to build won't work so what you need to
57:07do is you need to create a process in
57:09your company to build things quickly so
57:12that you can actually see whether
57:13they're work or not and then you can
57:14iterate them from there so it's far more
57:17important to have attack tactically
57:20talented team that can build MVPs
57:22quickly in a non frustrating way and
57:25then measure the results than it is to
57:28a super genius who can imagine what's
57:30going to happen in the future without
57:31actually knowing now I'm the big picture
57:34you have to have that imagination for
57:36your vision for where it's gonna be ten
57:37years now you have to have that
57:38imagination for the little technical
57:40like tactical move in the next three
57:42months like it's really hard to nail
57:46those if you have a process that can rip
57:48out things quickly and then only iterate
57:51the things that are working that'll
57:52serve you far better our mistake was
57:56it was thinking every time we've got the
57:58home run let's only swing for home runs
58:01and of course it would take three months
58:03to do it because we got to make it
58:04perfect right and then the whole spike
58:06spiral of death alright um the last
58:09thing I'll say is my email address is
58:10Michael at Y Combinator comm strangely I
58:13tell people that I answer every email
58:15and people mostly don't believe me and
58:18the ones who do email me and I reply and
58:21so everyone I talk to you and everyone
58:24online there's really only two
58:25categories of people the people who
58:26don't believe me in which case great
58:27continue your lives and the people who
58:30will believe me and if you need help
58:31they'll email and I will try to help you
58:34it's really that simple you don't need
58:36to have networked with me Y Combinator
58:39is fairly easy to spell and so is
58:40Michael so you should be able to figure
![How To Succeed As A NEW & YOUNG Realtor [Deals Every Month + Luxury Listings]](/_next/image?url=https%3A%2F%2Fi.ytimg.com%2Fvi%2FrtrRghgTayc%2Fhqdefault.jpg&w=3840&q=75)
