00:00 okay we're live hi my name is Eric I'm
00:03 the course facilitator at startup school
00:05 hi I'm Michael I'm a partner at Y
00:07 Combinator and help her yeah our
00:11 volunteer today to help answer your
00:12 questions so thanks very much for
00:16 joining in we've got a bunch of
00:17 questions that people asked about MVPs
00:20 on the forum we're planning to spend a
00:24 couple minutes just answering each one
00:26 and really trying to dive in and give
00:28 some give some advice tailored
00:31 specifically to the syrup so it's just
00:35 let's do it yeah first question from
00:38 Luke at Zeff Zephyrus zephyrus is an
00:43 analytics platform for competitive
00:45 gamers that provides users with game
00:47 specific tools and data to help analyze
00:49 track improve their performance well
00:51 let's start with the one-liner I I
00:54 thought it was pretty clear I kind of
00:56 knew what it was I'm from the space it's
00:57 not the sexiest one layer in the world
01:00 but it definitely accomplished the goal
01:02 of me having a good picture yeah it does
01:04 I kind of nodding my head and I
01:05 understand where this is going that's
01:07 like a a - for sure yeah
01:11 it currently is focused on Starcraft -
01:15 consumers expect more than a scrappy MVP
01:17 if there are existing products in the
01:19 space - advice for MVPs entering the
01:22 space with existing developed
01:25 competitors so that that was the funny
01:27 bit because I felt like the first
01:28 sentence in that should have been the
01:29 paragraph do consumers expect more than
01:32 a scrappy MVP if their other example
01:34 that's the question and the answer is
01:39 hopefully no because it's a start-up
01:42 your ability to build more than a
01:44 scrappy MVP is very challenging I would
01:48 say that if I go back to the to the
01:51 justin.tv days like if you can find a
01:56 small population of people that have
01:58 enough of a problem that the other folks
02:00 aren't solving that they're willing to
02:01 use something that's pretty bad
02:03 that's a very good sign if the only way
02:07 you can find any group of users even ten
02:09 even 100 who love you
02:11 to have a very polished product that's
02:13 not a good sign that's actually a good
02:17 sign that the other folks are doing a
02:19 really good job yeah that the products
02:20 work yeah so one of the hacks that I
02:23 like to use at this stage is to actually
02:25 layer the MVP on to the existing product
02:28 so for example if there's already some
02:30 software that your users know you're
02:33 like the people in the community are
02:34 already using to track their competitive
02:36 Starcraft 2 analysis what you could
02:39 consider doing is actually like building
02:41 say like a Chrome extension that sits on
02:43 top of the other products and just
02:45 overlays the extra data because you're
02:47 really like the goal of any good MVP is
02:49 to fist it is to test your hypothesis is
02:53 what you're making like important enough
02:57 for people to care about like will they
02:58 actually care that you're building it
03:00 and so you want to do that with like the
03:02 least amount of effort and if there's
03:03 already existing good products like it's
03:06 possible that you may find the users of
03:08 those products pretty easily and you can
03:09 like build that little little bit that's
03:12 that's special to you there you go yeah
03:15 number two cool next up is from plumbing
03:20 at flat row or flat row we automate
03:24 billing customer care and maintenance
03:25 for property managers ok the only thing
03:30 that's confusing to me is that I thought
03:31 that property managers do billing
03:33 customer care maintenance so are you
03:36 replacing property managers is it
03:37 software for property managers yeah or
03:41 is it like the billing of property
03:42 managers yeah so maybe a bit of
03:45 maybe a bit like I would I would
03:48 probably just start off by we make
03:49 software for product map for property
03:51 managers we help them automate billing
03:54 customer care and maintenance of their
03:56 properties something yeah yeah if it's a
03:58 tool part managers are using to be more
04:00 efficient than not to be more clear in
04:02 the up front and maybe even speaking to
04:04 how much more efficient does that mean
04:06 the typical property manager can go from
04:07 managing one property to five or from
04:10 five to fifteen yeah like what that
04:12 what's the value prop that you're
04:13 offering to them yeah I'm not a property
04:17 we didn't get it right off though I mean
04:19 like I can't visualize the pain though
04:21 I'm sure there's Rangers so we are
04:26 currently starting with a few clients
04:28 and are basically tailoring our service
04:29 then cool how do we avoid the trap of
04:32 making our service too specific and how
04:34 do we balance between the requirements
04:36 of a few clients with the need to build
04:37 a platform that is more Universal so I
04:41 kind of think that like when I see a
04:44 start-up like this I really have this
04:46 basic question which is are you going
04:48 into this space with knowledge of the
04:51 space like were you a previous property
04:52 manager if so you should have a lot of
04:55 insights and the types of problems that
04:56 are general that can be applied across
04:58 the industry and so you should have a
05:00 lot of insights on what your initial
05:02 product should be sub a set B if you're
05:05 coming into this with little personal
05:07 experience you think there's an
05:09 opportunity here but you kind of don't
05:10 know what it is I don't know any I this
05:13 is your problem and like I don't know a
05:15 clever way of avoiding it just learning
05:20 the business means you're gonna be
05:21 building a lot of features that probably
05:23 are not generally applicable and that's
05:25 the cost of learning about this business
05:26 so my mind like if you already know
05:29 about the business use your gut if
05:30 someone asks you to build something that
05:31 like your experience tells you is not
05:33 general great don't do it if not
05:37 consider the first five customers as you
05:39 just getting an education on the
05:40 industry and it might be a very
05:41 expensive one that's fine for the case
05:43 of someone who's coming into the
05:45 industry that they're not already a
05:47 participant of some of the like some of
05:49 the best founders that I've seen they
05:50 just immerse themselves in it like for
05:52 example they may actually become
05:53 absolutely manager like they may
05:55 volunteer or they just like go and say
05:57 I'm gonna help you run your business for
05:59 like a week or two I think like that
06:04 yeah even if I mean if you are a
06:06 property manager yourself you don't
06:08 necessarily have to do that because
06:09 hopefully that's yeah why we're starting
06:11 this company that seems like an even
06:12 better offer than giving them software
06:14 it'd be like hey I'll just do some of
06:16 your work for you so I can learn one of
06:18 the things that I learned from kind of
06:21 like to this to this question of like
06:25 how do you prioritize prioritize
06:27 features I always like try to
06:30 as I'm doing user interviews or talking
06:32 to initial users I really try to find
06:35 pain points that people are already
06:37 trying to solve so for example like if
06:40 you're talking to property managers and
06:42 they're totally comfortable doing all
06:44 the books by themselves like they just
06:46 have like their weekly like hour that
06:48 they spend doing all this that might not
06:52 actually be the right place to start
06:54 with maybe maybe you could start with
06:57 like are they are they actively
06:58 experimenting and trying other tools are
07:00 they have they tried like a competitor
07:02 product those are usually the problems
07:04 that I like to start with because like
07:07 the person's already trying to find a
07:08 solution so they're primed when you come
07:09 to talk to them like to be excited and
07:12 to try it versus you having to like
07:13 pitch them hard and like changing their
07:16 entire business workflow or whatever
07:17 yeah I think we see the opposite of a
07:19 lot where somebody says if something's
07:21 being undone on paper there is a
07:24 business in putting it into the computer
07:25 and like by nature of the fact that it's
07:29 being done on paper you could build a
07:30 billion dollar company bringing to the
07:31 computer and that's actually not true
07:33 like there are some things that done on
07:35 paper that are not broken
07:36 that's like build it and they'll come
07:37 then yeah yeah so really works all right
07:40 what's next how do you okay we're asking
07:48 and answering questions from sky who has
07:51 started fridge worthy let's see what's
07:53 fridge where should we use first they
07:54 did with young maybe will be your
07:56 initial users ok fridge worthy a social
07:59 network where people share their
08:00 accomplishments well sounds good for Ivy
08:02 Leaguers you mean like LinkedIn people
08:08 are using LinkedIn more as a social
08:10 network yeah things were they there's
08:13 just an urban I'm trying to see if
08:14 there's like a website no website yeah
08:15 okay how do you launch an MVP of user
08:20 generated a user-generated content
08:22 platform if I make an MVP with minimal
08:26 content users will probably leave since
08:28 there's not too much to consume so the
08:33 classic answer to this is one does your
08:36 platform have a lot of replay value is
08:39 there any reason why I should come back
08:40 daily or weekly I think that
08:43 that's a really important question ask
08:44 because even if you get a lot of people
08:46 if there's no reason for them to come
08:47 back it doesn't really matter I think -
08:49 is there a smaller community that if
08:52 that community was using a product they
08:54 would find it useful even if others
08:55 weren't the classic I guess now old
08:58 example of this was Facebook at Harvard
09:00 like as long as the Harvard kids were
09:02 using it or even a class even just one
09:04 of the grades in Harvard was using the
09:05 product it was useful and so therefore
09:07 they didn't need to have millions of
09:10 literally users of the order of like a
09:12 couple thousand was very useful so the
09:16 way that I'd be thinking about this
09:18 question is one how frequent do I get
09:20 value out of it if it's not that
09:21 frequent is there any value on creating
09:23 - is there a smaller population of
09:26 people who can find this useful even if
09:27 all our friends aren't using it and
09:28 those people might use it like have the
09:31 problem or whatever on a more regular
09:32 basis so they'll be like participating
09:35 more so start with like a focus smaller
09:37 group I think there's always you know
09:40 tricks of the trade like generate them
09:42 yourself a lot of the best user
09:45 generated content networks like 9gag
09:47 which is now this like massive meme meme
09:51 website was started by like scraping
09:53 memes from other sites that's not that
09:56 shouldn't be something too too crazy to
09:58 think about reddit same story in the
10:00 early days Sachs Amster um so like in
10:02 this case if you're talking like if
10:04 you're building a network for people
10:06 sharing their accomplishments you could
10:08 look to where people are already sharing
10:10 this and like prime prime it by grabbing
10:13 the accomplishments like for example if
10:15 someone's tweet like if you're trying to
10:17 like get an someone who's already on
10:19 Twitter on to your platform you could
10:21 potentially like go through each
10:23 person's Twitter account and scrape
10:25 their the things that they're proud of
10:26 crunch page is actually interesting
10:28 example of this before people cared
10:30 about their approach phrase profile
10:32 TechCrunch just built the profiles on
10:34 CrunchBase yeah if I'm media reports and
10:36 it was media reports and so and so forth
10:37 and after a while people cared about
10:38 their profile and so they tried to keep
10:40 it updated and when I'm thinking more
10:42 about this problem like it's true like
10:44 we have a lot like one github we're on
10:46 like dribble there's all these different
10:50 platforms that people may participate in
10:52 so that's us I mean that's a service
10:54 right there like not how
10:55 generate your own page yeah cool
11:00 next up Dennis at swish swish swish
11:07 swish technologies we're building the
11:10 self-driving cars for ride-sharing and
11:12 retirement communities and for schools
11:13 in the US okay so I'm imagining like
11:17 club cars or whatever got down please
11:20 share some of your advice for founders
11:22 like us with a product that can take a
11:24 very long time to launch how do we get
11:26 users even before we launch how can we
11:28 generate sales even before we launch so
11:32 I deal with a lot of companies you seen
11:35 a lot of this we hope they like 25
11:39 companies that are all you know working
11:40 their way through the FDA process and
11:42 we've also funded a lot of like
11:45 supersonic aircraft companies that will
11:49 probably take a little while for launch
11:50 so this is like this hands-on I mean
11:51 Guardian self-driving car companies in
11:53 fact probably we follow this exact idea
11:55 yeah so there's a couple different
11:59 there's a couple different ways that you
12:00 can go with this I'd say that in in
12:07 areas where there's like the possibility
12:09 for harm or there's like safety concerns
12:13 you have to be extremely careful about
12:15 what you consider an MVP in any way that
12:19 can actively like harm somebody so that
12:25 having been said we have had companies
12:29 test smaller versions of what they're
12:32 like smaller versions of their larger
12:35 vision with kind of off-the-shelf
12:38 technology some of the best hardware
12:41 companies that I've worked with even
12:43 though they now make their own hardware
12:45 and they they build a bunch of custom
12:49 stuff they started using they started by
12:51 using off-the-shelf systems like
12:53 raspberry pies cameras just existing
12:56 technology and put them together in
12:57 interesting kind of normal ways so in
13:01 this case like there is
13:04 now you know more and more kind of
13:07 off-the-shelf tech that you could use to
13:08 craft together a like a rudimentary
13:12 I think we're getting back to like I
13:16 guess like why we work on MVPs one
13:20 reason we do is to actually validate
13:22 that people want this
13:23 the second reason we work on MVPs is to
13:26 demonstrate that we as founders have the
13:28 skills and capabilities to like to that
13:33 kind of like take the problem from just
13:36 an idea to an actual implementation even
13:38 if that implementation is not the like
13:40 long vision of this so in this case it's
13:43 very easy to have the idea to say I want
13:45 to build like a self-driving car or
13:46 something like that but it's but there's
13:48 like a huge gap there's a huge discrete
13:50 bump between people who have the idea
13:52 for self-driving car and people who have
13:53 like built something that resembles it
13:56 so in this case like building something
14:01 that demonstrates your capability could
14:04 separate you dramatically from the pack
14:06 so we've had companies that are now off
14:09 building like embark which builds
14:11 self-driving trucks like literally
14:13 they're running on the streets in the
14:15 highways in Nevada they started with a
14:18 cop car yeah I think like at demo day
14:20 they had a golf cart that they rigged up
14:24 with like six cameras I don't even know
14:26 if it was autonomous it might have just
14:27 been remotely driven or something like
14:29 that there's a great there's a great
14:34 like comic on how to build an MVP and it
14:37 literally is someone building a car but
14:39 the first step is building like a
14:40 skateboard and then you build a bicycle
14:42 and then you build a car so you can
14:44 apply the same logic here that like if
14:46 you're trying to figure out how to build
14:47 something that's autonomous that works
14:48 in these communities maybe you start
14:50 with like an autonomous wheelchair right
14:52 like you just create like or one of
14:54 those personal mover things right you
14:58 start with that and it still kind of
15:00 accomplishes the first the overall goal
15:02 of making something that's for
15:04 retirement communities for example but
15:06 much safer much cheaper and that
15:10 actually sounds like kind of a cool idea
15:11 maybe Segen or not at all
15:14 the couple of parts of this question how
15:17 do we generate sales before we launch I
15:20 actually think that that is sales are
15:24 hard generating letters of intent or
15:27 agreements to pilot you shouldn't be
15:29 very hard if there's a lot of demand for
15:31 this sort of thing so being able just to
15:33 get your customer to sign anything even
15:35 if doesn't have a price on it is
15:36 evidence that you can sell to an
15:38 investor and is often seen as more
15:42 valuable than not having it so some
15:45 other things that people have done in
15:46 this way is if your plan is to sell cars
15:48 maybe the first thing you do is you sell
15:51 the service so you don't have you don't
15:53 have autonomous you know self-driving
15:56 cars yet but you do have yourself so
15:58 what if you ran a car sharing service in
16:00 this retirement community to start that
16:02 would demonstrate to people that you
16:04 know there are people who want to use
16:05 smartphones to like request cars not
16:07 kind of thing and you could you could
16:09 literally get started like tomorrow
16:10 doing that potentially all right what's
16:15 we've got Ethan at volley word volley we
16:20 facilitate the servicing of users
16:22 vehicles they do not need to leave the
16:24 home or office servicing that seems like
16:27 an awkward way of saying we're a car
16:30 mechanic that yeah anywhere you are I
16:32 think this is a perfect example of like
16:34 the description requiring examples
16:36 because servicing does not put a picture
16:40 in my head as opposed to refuelling
16:42 changing the tires doing an annual
16:46 checkup all those things create pictures
16:48 in my head and also there's probably
16:49 something that you're starting with I
16:51 imagine that you're not like servicing
16:53 every end-to-end thing in the car to
16:55 start so some of the best pitches like
16:58 like Michael said they cause you to like
17:00 visualize what the service does but
17:02 here's actually a problem that a lot of
17:03 founders run into is they think that
17:05 their one-liner or their short
17:06 description has to encompass like
17:08 everything that they'll ever do under
17:10 the ends of the earth because they won't
17:12 they think that like investors only will
17:14 get excited by these like big dreams but
17:16 in reality like we're actually just
17:17 trying to understand what your company
17:19 is so that we could talk about something
17:20 else and so your description doesn't
17:23 need to be this all-encompassing future
17:25 vision it's just like we won't
17:26 I know what you do quickly like get to
17:29 the next step what's while getting all
17:31 of the legal taken care of to protect
17:32 from liability so Ethan the first
17:34 question I would ask you before you even
17:36 think about this is have you studied
17:37 every startup that has failed trying to
17:40 do this idea and do you understand why
17:43 if not you are doomed to repeat their
17:45 failure so you can start by looking in
17:48 the YC database we have funded this so I
17:52 really think that the real number one
17:54 questions do you understand something
17:56 that everyone else doesn't know about
17:57 this space the other thing that I think
17:59 is tricky is that like the population
18:05 that has the most problem with their
18:07 cars tends to also be in a population
18:10 that is extremely price sensitive about
18:13 fixing their cars and so make sure that
18:16 you have some insight on how you can be
18:18 offering the cheapest option because if
18:20 you're if you're the kind of issue here
18:22 is a lot of people like to go with
18:23 luxury and high price first but the
18:26 problem is like you're solving like car
18:27 braking which is way less common if I
18:30 bought it a year ago and so make sure
18:33 that you're actually letting this up so
18:34 you actually have a product that can
18:36 have demand oh yes we've seen in other
18:38 countries as well yes say in terms of
18:42 the liability question like you got to
18:44 be you got to be extremely careful
18:45 especially when you're dealing with like
18:48 large expensive items that people care
18:51 about the way that I like to think about
18:53 this is you know trust your gut you're
18:56 usually not going to make you're usually
18:58 not going to try to put yourself in a
18:59 position where you're taking too much
19:01 risk naturally so like listen listen to
19:04 listen to yourself and listen to other
19:06 people when they when they when they
19:08 kind of bring up problems that you that
19:10 they think you might encounter alright
19:13 okay this one's okay can you pronounce
19:17 that one no are we doing high-five or
19:19 this Oh being being stopped gonna stop
19:24 Louis Louis okay Louis we're a been a
19:28 stop we build a home wellness service
19:30 and products marketplace for people's
19:33 wellness objectives okay no idea I don't
19:37 really know what they do
19:39 home wellness I think I'm like thinking
19:41 of home gym I'm telling you I don't even
19:45 want to stretch my mind to try to figure
19:47 out what emotions because it's so much
19:49 jargon it is a unique name let's see
19:52 let's see if the website has anything
19:56 didn't step Oh best app surfing this oh
20:01 no is it be honest Dennis being a step
20:05 ok services for your well-being at home
20:08 what kind of services what kind of
20:11 service Quest Oh exercises meditation
20:14 sports when you say like yeah like your
20:18 description on your front page is much
20:20 better than scription does so that okay
20:25 back to the question great
20:27 should we launch our MVP for free or is
20:32 it better to charge people let's be
20:34 clear you have a service where you're
20:35 actually like there is real cost meaning
20:39 like your service is helping someone
20:41 exercise there is cost to sending
20:43 someone to that place
20:44 there's costs recruiting them there's a
20:46 cost to their time in any situation
20:48 where you actually have real cost you
20:50 should be charging from day one in a
20:52 pure software play where you don't have
20:54 any costs maybe consider something
20:57 that's freemium but like you don't want
21:00 to anchor like the number one thing with
21:02 these like marketplaces that are
21:03 actually moving humans around is that if
21:05 you can do a situation where you're
21:06 providing a dollar with a value and
21:08 you're only getting 75 years that sets
21:10 back you're gonna get a lot of demand
21:12 but you're gonna lose a lot of money and
21:14 when you actually equalize the the
21:17 demand in the price you'll lose a lot of
21:18 customers especially for something like
21:20 this where it's an existing service that
21:22 people are already buying from other
21:23 vendors yeah so you kind of don't need
21:26 to validate that like people want to
21:27 have yoga classes or people want to have
21:29 dance classes what you want to validate
21:30 is do they want to have them at their
21:32 home like that's the that's the thing
21:34 that's kind of new and I think unique
21:37 so like definitely definitely charge for
21:41 it yeah I think just charge a lot charge
21:43 more than you might feel comfortable so
21:44 you can identify the people are most
21:45 desperate for it here's your crazy idea
21:47 charge more than the existing solutions
21:49 because like when I think about it if
21:52 I'm like I want to have a private yoga
21:54 class that's expensive I would have to
21:56 schlep out to go to the studio and I'm
21:58 having it at home yeah I mean and that
22:02 would actually charging more it sounds a
22:04 little bit weird but it would like force
22:05 it would really force the question of
22:08 like are you building something that
22:09 people really want you would get an
22:11 answer really quickly if you were
22:12 charging more than like the neighborhood
22:14 yoga studio and people are still coming
22:16 to you then you're really on to
22:17 something if like you're constantly
22:19 having to cut the price and like you
22:21 know run discounts and stuff like that
22:23 like maybe they're only coming to you
22:24 because it's slightly cheaper than the
22:25 alternative so there you go yeah okay
22:29 now we're on to high five high five is a
22:32 local news website that helps you get
22:34 invited to activities based on topics
22:36 you enjoy discussing we help people in
22:38 their 20s and 30s meet new friends with
22:40 similar hobbies humor and lifestyle that
22:42 sounds like two one-liners in one or
22:46 three or three yeah so once again Vlad
22:50 something that a lot of people have
22:52 tried before is both local news and
22:55 interest based social networks and
22:57 dating and dating and meeting new people
23:00 yeah so like make sure that you're
23:03 studying those products and you have a
23:05 strong opinion about why they suck it's
23:08 gonna help you two reasons one it'll
23:10 help you figure out what to make is a
23:12 good product and two a very common
23:14 investor question if you ever want to
23:16 raise money is like well what about this
23:18 everyone's always heard of like the
23:20 other app that they funded they may have
23:22 funded like five of them yeah and if
23:24 your answer is like oh I don't know then
23:26 instantly you lose credibility so make
23:29 sure you're an expert on all of these
23:30 various things my hunch here is that
23:32 Vlad this might actually be a problem
23:34 that you're solving for yourself which
23:36 is great like maybe you just moved to a
23:38 new city and like you're experiencing
23:40 this exact problem so like or yeah or
23:45 your friend has so like the way that I
23:48 would start building an MVP is kind of
23:49 like what we mentioned at the beginning
23:50 it's taking what you already do on a
23:54 regular basis like maybe you go to
23:55 meetups maybe you like read it for the
23:58 subreddit event of that city and like
24:00 think clearly about like what are you
24:02 doing today and what's the smallest
24:04 thing that you could do that has an
24:06 outsized impact like you want to you
24:08 want to add the least amount of new kind
24:11 of like process to your weekly workflow
24:14 that just has this like out sized large
24:19 all right okay what do you do after you
24:22 launch an MVP exciting okay next up we
24:25 got a keys from small scope what are
24:30 some steps to launch an MVP that found
24:32 that what are some steps after launch of
24:35 an MVP that fence founders should follow
24:37 in the video there are some examples of
24:39 successful companies MVPs however
24:41 there's no discussion at all on how they
24:43 iterated what they learned why they
24:44 narrowed it down to those features so
24:47 Peter from segment Donahoe actually had
24:52 this week on this subject which i think
24:54 is really interesting so we started with
24:56 like this depressing fact most startups
24:58 are in the process of dying and that's
25:02 offense in fact one and depressing fact
25:04 two is that most startups have convinced
25:07 themselves that they're right on the
25:08 verge of product market fit one of the
25:10 fact they're actually in the process of
25:12 dying so if you set that as the like
25:15 environment that you live in one of the
25:18 things he said that I thought was really
25:19 helpful is take your MVP and be a
25:22 skeptic have someone in your company be
25:25 dyspeptic and basically ask the question
25:28 is this providing any value to my users
25:32 and not like oh well you know if I built
25:36 these 17 features than it would is my
25:38 first MVP what the service that I could
25:41 write people today is it providing any
25:42 value to my users then if the answer is
25:45 no go figure out who's in so much pain
25:48 around this problem that they would use
25:50 your MVP and get them to use it and then
25:53 ask that question again am i providing
25:55 any value to those users it turns out
25:57 that like it's too easy to go down a
25:59 track where step one you're not
26:02 providing any value and just to make it
26:05 so that iteration one you're providing
26:07 no value iteration two you're writing no
26:09 value iteration three of writing value
26:10 if your product if your problem you're
26:13 solving is severe enough there should be
26:15 some over people who are willing to use
26:16 a really bad product to help solve it
26:18 and so go find them
26:20 that's where I think the best folks that
26:23 do MVPs and go and what they do is they
26:26 literally take what they have to
26:29 and they talked to different groups of
26:31 users and they begin to kind of instead
26:34 of iterating on the product like
26:35 building more features they actually
26:36 iterate on the market yeah so they'll
26:38 find like I mean it's it's always a
26:42 great example but the superhuman the
26:45 super hold the founder of superhuman
26:46 board a great blog post on how you
26:48 iterate towards product market fit and
26:50 one of the things that he talks about is
26:52 looking at like the demographics within
26:55 the first hundred users and he realized
26:57 that like founders sales people were
27:00 really like really taking advantage of
27:03 the basic features that super human had
27:04 and then like product managers and some
27:07 of the people who didn't like deal with
27:09 email on the same volume just like did
27:12 not get value and stopped like using it
27:16 another test to kind of your point of
27:19 like do people benefit from this very
27:23 first version one like kind of crazy out
27:25 there test that I sometimes seen people
27:27 run is they kind of they turn off their
27:29 product and they see how complaints like
27:32 you don't really have to do this but you
27:34 could kind of figure out ways to
27:35 simulate this so for example if you had
27:37 an outage if your servers went down like
27:40 last week look back at your email logs
27:43 to see like did anyone care did anyone
27:45 care so you might already have run this
27:48 experiment by accident at some point
27:50 it's a perfect example of doing
27:51 something doesn't scale like a big
27:53 company could never do that and awesome
27:55 all start-up like I mean reddit did that
27:57 from yeah like I mean it Twitter did to
27:59 but like a small start-up but something
28:01 you totally can do and if you think
28:03 about the environment is like the
28:05 greatest percentage chance of what I'm
28:06 doing is death then you should be
28:08 willing to experiment with extreme
28:09 things fairly with death yeah that's
28:12 alright that's next I do remember a lot
28:14 of failed wills funds whatever else and
28:17 I always complained like whenever reddit
28:18 would go down I would complain to my
28:20 friends when we were trying to like oh
28:21 yeah router where I would text Eve it
28:23 would be funny I'd be like Steve your
28:24 website's down I'm sure I'm the first
28:26 person who's telling you
28:28 like that's a massive sign that you're
28:30 onto something if like people can't live
28:32 without your website like five minutes
28:34 yep got something yeah
28:37 next up from posh wha-at local be my
28:42 company's our website for shopping from
28:44 home base and small businesses around us
28:45 after evaluating what I do I've realized
28:48 that I have a small I have I have users
28:50 interaction that we do not have product
28:52 market fit now the question is should I
28:54 continue working on this especially
28:55 since margins are thin or should I quit
28:57 this and beginning in and search for
28:59 market product market fit you know I
29:01 think this is always the hardest
29:02 question to answer because I see
29:04 problems on both sides there's a set of
29:06 founders who if they don't succeed
29:08 immediately their mental model is that
29:10 their company's a failure but then
29:12 there's also a set of founders who even
29:14 after learning two years worth of
29:16 horrible facts they still like will not
29:19 acknowledge that it's not working I
29:22 think the best strategy for this the one
29:26 that's easiest to execute is to
29:27 time-bound it so what I would say is
29:30 like hey we're gonna try everything as
29:33 hard as possible to make this work for
29:35 this period of time probably no less
29:36 than six months and if we haven't
29:39 figured it out then then we're gonna be
29:40 open ourselves up to pivoting I find
29:43 it's easier to time bandit because like
29:45 you can it's easier to be intellectually
29:48 honest with yourself about like have I
29:50 worked on this really hard for six
29:51 months versus three versus have I tried
29:55 all of my ideas like that's a look at
29:58 they're running out of ideas then yeah
30:00 yeah like you like ideas they're kind of
30:02 thing it's it's it's easy to trick
30:03 yourself that you've tried all of your
30:04 ideas it's not easy to trick yourself
30:06 have I been working on this for six
30:07 months so some strategies that you can
30:09 use during that six months are like put
30:11 all the ideas that you still have up on
30:13 a whiteboard and like pick the top one
30:16 and spend a week working really hard on
30:20 that because it's possible that like
30:22 you've been going through this for you
30:24 know a couple months or a year and
30:25 you've gotten into this into this setup
30:28 where you think you've you know tried
30:29 everything or you're not sure like about
30:32 a specific question that you might have
30:33 like am I going after a demographic am i
30:35 and I use it do I have the right stores
30:38 or small businesses on my platform and
30:39 you could test it you
30:41 say like actually we don't have product
30:44 market fit the founder of TaskRabbit was
30:47 just speaking last night and she was
30:48 talking about how like different that
30:51 first every single everything like
30:54 TaskRabbit you could just do anything
30:55 yeah but then they started realizing
30:57 that like some people who really just
30:58 wanted idea furniture set up and so one
31:01 of the things that she said that she was
31:03 looking back and she was like I wish we
31:04 tried this at the beginning was saying
31:05 what if we just made like a landing page
31:07 that was just like get your IKEA
31:09 furniture set up and so I could imagine
31:11 something like this really like forcing
31:14 that issue and the cool thing about that
31:15 is you can do that in a week yeah like
31:16 you already have built all the framework
31:18 for it you could spin up a landing page
31:21 that's like I wanted my hypothesis is
31:24 that people really want to buy you know
31:28 baby stuff on like a really short term
31:31 basis because they can't run out and get
31:32 it so I'm gonna use my existing website
31:35 and the existing like whole thing that
31:36 I've built and just test that one
31:38 demographic with one use case and then
31:42 when to like when to quit
31:44 insert start the search over again what
31:46 did Peter say about that because he did
31:48 like they certainly I think Peter said
31:50 that like if you can convince the critic
31:54 in the company but basically he was like
31:56 identify a critic in the company
31:58 identify a sign that would either that
32:03 would even convince the critic that
32:04 there's something here if you can't hit
32:06 that sign within a period of time you
32:08 should move so like don't trust the
32:11 faith of the optimist like try to assume
32:14 the pessimist role and like try to come
32:17 up with an objective measure of like
32:20 yeah next person would be like holy
32:21 yeah yeah even the skeptic would be like
32:23 yeah okay I guess we have to work on
32:25 this like that's what you're looking for
32:27 and like for him he tells a story
32:29 segment which is really cool which was
32:30 that the idea of segments he didn't
32:33 believe in and his co-founder did and
32:35 his test was like well okay put it on
32:38 Hacker News let's see what like
32:39 engineers think and he thought it was
32:42 like it was just gonna not get any up
32:45 votes and move on to the next topic and
32:46 the next day it's from an age of hacker
32:49 news and he was like oh crap like even
32:51 as a skeptic I must acknowledge that
32:53 is way more demand for this than ever
32:55 thought like that is a good yeah a good
32:58 mental framework you have to get
32:59 yourself out of tricking yourself that
33:01 that it's working when it's not next up
33:05 crystal at site full along our journey
33:08 we've done surveys interviews and had
33:10 many conversations with users once the
33:12 MVP is live and we invite users to use
33:14 and test our tool what are the top
33:16 methods you recommend for capturing and
33:18 measuring feedback I often think that
33:20 startups try to scale this effort way
33:22 too fast the things that I've seen work
33:24 really well are you call the users
33:27 yourself personally you invite the users
33:29 to your office you host a dinner for
33:32 your users you put your initial users in
33:35 a whatsapp group that you can all talk
33:37 with together but like things that
33:39 absolutely don't scale another famous
33:42 one was Airbnb where customer service
33:44 was one of the co-founders Joe's
33:46 cellphone number like this is the
33:48 perfect area we're like you want there
33:50 to be so much information flow from the
33:52 customers to you that like you have to
33:55 do things that don't scale you shouldn't
33:57 be thinking about oh how do I get my
33:59 like ideal intercom setup and my Zen
34:02 desk set up in the other days you do not
34:03 need that stuff but let me unpack
34:05 something there because it's easy to
34:08 overlook I think why we talk about do
34:10 things that don't scale the the reason
34:14 in this case is actually you don't know
34:15 the questions to ask yet and so you want
34:18 a really high bandwidth a really wide
34:20 broad way of collecting feedback because
34:23 you don't know the questions that you
34:24 want to ask it and so by having this
34:26 like open-ended like the example there
34:29 may be very cool because they're
34:30 literally just like talking like
34:32 actually talking to users that have
34:34 problems so you could ask questions like
34:35 when they when they phone up and say
34:37 like hey like take him to my Airbnb you
34:39 solve that and then while they're on the
34:41 phone you say like oh by the way how did
34:43 you find us how's your experience so far
34:45 like your friends you like you basically
34:47 can you basically can like pack the
34:49 system and instead of having to create
34:51 like the perfect survey the perfect you
34:53 just like talk my recommendation is to
34:56 write that all down and maybe on like a
34:58 weekly basis get together with you and
35:00 your co-founders kind of debrief about
35:02 these conversations that you have
35:03 because like some of the times like
35:05 during office hours
35:06 I really like talking to founders that
35:08 have had these conversations and we just
35:10 kind of just talked through I'm like
35:12 tell me about like this users background
35:15 like where do they work how do they and
35:17 and sometimes you can pull interesting
35:19 insights out of this that can help you
35:21 like tune and tweak your MVP to
35:24 basically like kind of to the point that
35:26 we were talking about before about these
35:27 like experiments that you can run you
35:28 could say like well actually like a lot
35:30 of you know singles are using this tool
35:33 and I've talked to like three of them
35:34 maybe we could like try something that
35:36 just goes after this demo yeah
35:40 next up MVP as part of your startup
35:45 pitch well let's see what Spencer has to
35:47 say from Spencer at carnuba work her
35:50 nuba the first online back office and
35:53 field service management platform for
35:54 independent car care professionals Wow
35:57 that is a mouthful and you say mechanics
36:01 Detailers etc field service management
36:05 platform is jargon that's something that
36:07 probably they understand in a normal
36:09 investor will not independent I'm
36:11 imagining like awesome one-off mechanic
36:13 yeah I don't know what's good like a
36:14 nutburger professional is either okay
36:16 here's what I think you do you write
36:18 software for mechanics shops and it
36:22 helps them manage dinner there it like
36:26 what places the paper with them yeah I'm
36:29 still not totally alright so that's the
36:31 question our end goal is to build an
36:33 online marketplace to connect car owners
36:35 with local independent car care
36:37 professionals and get it to go chain
36:38 says but our VP is the back-office a
36:41 pure service management platform to
36:42 build up our supply side of the
36:43 marketplace should we be pitching to our
36:46 investors our end goals or what we're
36:47 working on now classic question so
36:50 here's how I would probably structure
36:52 the pitch step one what you're working
36:54 on now step two the traction step three
36:58 the business model and then step four
37:03 division it's a lot easier to sell me if
37:07 I know what the hell's going on and
37:09 those first three now I kind of know
37:11 okay this is what you're doing today I
37:12 have a base of knowledge so I can
37:14 basically when you tell me the vision I
37:17 have context enough to
37:18 how where you are today connects with
37:20 the vision if you just send me the
37:22 vision I'm gonna be very confused that
37:25 you you know if you're saying this
37:26 marketplace I'm imagining why don't you
37:28 just build the marketplace today and
37:29 then like Oh actually we're doing this
37:31 like b2b product now I'm just confused
37:35 so Star Tours you're doing today then
37:37 talk about the vision here's here's
37:39 another reason why I really like
37:40 confounders do that is I can evaluate
37:43 like I'm mostly at this early stage I'm
37:46 mostly just trying to like understand
37:47 you and like are you are you cool are
37:50 you like you're smart have you have you
37:52 figured out a pain point and by sharing
37:55 like what you're doing and some but if
37:58 you just talk about this vision I have
38:00 no like metrics to evaluate you you you
38:03 you know if you pitch you're short if
38:05 you pitch your short-term goal I can say
38:06 like what you're like is every part of
38:07 the way there's no way there if you
38:09 pitch this like take over the world
38:10 vision you're like no winner yeah I just
38:13 can't really differentiate like are you
38:15 a poser or you just like talking a big
38:17 idea thing where are you actually
38:18 getting deliver on this oh no it's
38:22 questions during 3 what we're gonna go
38:24 into some YouTube questions see how this
38:26 goes autumn at jump dot IO asks says my
38:32 name is autumn and my business is a
38:34 barter service that connects startup
38:35 entrepreneurs with one another ok
38:39 bartering ok I thinking back the barter
38:43 no pictures yeah maybe like an Olympic
38:45 training are they trading desks or the
38:46 trading computers the trading service
38:48 providers software's no idea in
38:52 consideration of getting an MVP up as
38:53 soon as possible what are the pros and
38:55 cons of using white label services so
38:58 this is a good this really brings up
39:00 kind of a classic thing so I would say I
39:02 would say a couple things to this point
39:03 one if you're just trying to test any
39:05 demand at all and a white label service
39:08 between like let you do that great but
39:10 I'd argue that like if you want to test
39:11 any demand at all for this type of
39:12 product probably like a Google sheet
39:15 spreadsheet could do it too and so if
39:18 you're just looking to test is there any
39:19 demand like go as simple as possible
39:22 if you have seen demand and you're
39:25 thinking to yourself ok I want to make
39:27 this a company if you want to make it a
39:30 software company that can
39:32 scale that can raise money it's a lot
39:34 easier if you're building your own
39:37 what's an exit vivec at fleet code I am
39:42 asks says fleet Co provides a business
39:46 assistant for transport and logistics
39:47 companies which helps business owners to
39:50 know how much their vehicles are making
39:52 and profit a profit and loss oh I don't
39:56 know what a business assistant is is
39:57 that like it my consultant is it an
40:01 it sounds like an Indian specific thing
40:03 potentially so oftentimes like we you
40:06 know are investing a lot in India right
40:08 now and I think one of the things that
40:10 I'm learning is that there are certain
40:11 different like just different facts in
40:14 the Indian market but as someone who
40:16 lives in San Francisco I can't like
40:18 visualize that as much so some of the
40:19 best Indian fakirs they actually like
40:21 spend maybe like a sentence or two
40:23 bringing me up to speed on what's
40:25 happening in their market so that I
40:26 could begin to like understand what that
40:28 is it's question about pricing yeah and
40:30 so the same answers we get before price
40:32 expensively make it so that you have a
40:37 high bar for delivering great value and
40:39 that you have a good verification that
40:41 the person has a lot of pain it's more
40:44 often that people price under price than
40:46 over price so lean towards over pricing
40:49 again it forces it forces the issue see
40:56 what else we got let's do this one okay
41:00 okay Deb food is an e-learning platform
41:02 to help people learn software
41:04 development okay we've already launched
41:07 our MVP it's basically a publication
41:09 text plus YouTube we get about a
41:11 thousand users on our channel monthly
41:13 shall we focus on building new features
41:14 and innovations or should we go after
41:16 currently implement a little revenue
41:20 sources like subscription court oh ok so
41:22 should we start charging for things that
41:23 we're already offering or should we
41:25 build new free features this is a tricky
41:28 one because it really goes to like what
41:31 your current situation is aka do you
41:33 need money right look how big of a
41:36 startup you're looking to build if
41:37 you're looking at with a lifestyle
41:38 business then like I would definitely
41:42 and kind of what your vision of what the
41:43 product should be and so like this one's
41:45 hard to answer just straight away if
41:47 you're thinking oh I don't need this to
41:49 be a venture scale business I want it to
41:51 be something I can do with all my time
41:53 so it needs to make money so I can quit
41:55 my job then charge great if you're
41:57 thinking like oh I'm looking to build a
41:59 free product for teaching people to
42:01 learn the software development because I
42:03 want to monetize it in this other way
42:06 then I'm not so sure and then in terms
42:09 of building new features like that's a
42:11 question for your users like what
42:13 what do your users want what are they
42:15 how are they frustrated with your
42:16 product today I mean you're also looking
42:18 at engagement like our people coming
42:20 back and using your service on a regular
42:23 basis we often get questions like this
42:26 that basically asked us for product
42:28 insights when we're not the customer and
42:30 like be very careful about asking any
42:32 startup person who's not your customer
42:34 for for customer insight because they
42:37 won't have them it's a great point yeah
42:40 and but they'll offer them they will yes
42:43 yeah a bunch of recommendations that are
42:46 based on ok now we're actually we're
42:52 gonna get ready to the YouTube comments
42:53 oh here's a good one how do you deal
42:55 with co-founders conflict with regards
42:57 to company direction so that's never
43:00 yeah I think that step one is to have
43:06 the conversation in a respectful way I
43:09 think oftentimes what happens is that
43:11 people avoid this conversation until it
43:13 becomes so problematic that it causes a
43:17 founder breakup and in the early days
43:20 just people being heard and conflict
43:22 being heard actually helps to spell it
43:25 so that would be my first step I think
43:27 step two is we talked about kind of time
43:29 bounding and dealing with the sceptics
43:31 let's say that your co-founder becomes a
43:33 skeptic ask him ok what's the test that
43:35 would prove that the direction we're
43:37 going in right now is a good one and
43:39 what is the time period that we can all
43:40 agree with to go after that and then BAM
43:42 we're running an experiment now if we
43:44 don't make it we'll switch if we do well
43:48 you'll change your mind much is your way
43:51 to what once you're having a
43:52 conversation and kind of structure
43:53 that way as opposed to like you have to
43:56 listen to me because I'm the product guy
43:58 or I'm the CEO that that really works
44:00 well you give this piece of advice at
44:02 the beginning of YC that I really like
44:04 which is to get all the co-founders
44:06 together in a room and plan the week or
44:08 two to plan your two-week sprint and
44:11 then to ask everyone to put their ideas
44:12 let's on the white board all ideas go up
44:15 there like you don't you don't start
44:17 like deleting ideas at this point and
44:18 then you like once the ideas are up then
44:20 you prioritize I think too often
44:22 especially like developers we're already
44:25 internalizing like how difficult things
44:27 are gonna be to build yeah when we might
44:28 not have understood like the potential
44:30 benefit from building this and it's
44:32 always a trade-off so like go broad
44:34 think think of what think of like
44:36 entertain the idea of what you can build
44:38 and then filter it then and kind of like
44:41 prioritize the list rather than just
44:43 like preemptively shooting down ideas
44:45 yeah brainstorms turn into shooting down
44:47 ideas so fast yeah oh god
45:00 that's more pricing questions hard tech
45:04 cuz at the bottom of the list no I'm
45:05 just looking through to see which one
45:06 okay they're not I was like okay here we
45:11 Elias asks how should I talk to my first
45:14 b2b customers to let them know that I
45:16 want them to sell but also that they are
45:18 part of like a beta test so this is a
45:21 I always think and a lot of times
45:24 founders ask these questions where it's
45:25 basically like the answer is just to be
45:28 honest right like or the answer is you
45:31 can just ask right and this comes up
45:33 with investment car investor
45:34 conversations and customer conversations
45:35 right you're looking for early customer
45:39 who is excited by the opportunity to
45:42 work really closely with the founder be
45:44 able to give feedback quickly to be able
45:45 to get changes fit quickly and it's not
45:47 intimidated by something that's actually
45:48 a feature yeah but a lot of customers
45:50 that's like they would pay more for that
45:52 yeah and so you're looking for that
45:54 customer so be upfront about it and it's
45:56 same thing with an investor really like
45:58 a lot of times we get asks a question
45:59 like what did the investor mean by this
46:01 or what stage of the process in my--in
46:04 and the answer is like just ask the
46:06 investor I think like it actually shows
46:09 a level of sophistication to ask the
46:12 question when you don't know as opposed
46:14 to just kind of like not as if you do
46:16 know and not and then leave without any
46:18 insights and the failure mode here is if
46:20 you don't bring this up now and you just
46:22 like present yourself as this big
46:24 enterprise software company like they're
46:26 gonna be disappointed you're gonna be
46:27 disappointed like everyone's gonna have
46:29 a not fun time so spin it in a positive
46:32 yeah like it's it's again you might even
46:34 close the deal because you bring this up
46:36 maybe don't call it a beta test just say
46:39 like I'm gonna work extremely closely
46:40 with you to make this a success like I'm
46:43 gonna I'll be awake 24/7 making this
46:45 work because you will be awake 24/7 all
46:49 right what else here we go
46:52 Oscar says we're blue a platform where
46:55 housewives resell products from
46:57 suppliers using whatsapp it's like me
46:59 show for let him get that should we give
47:02 incentives to our housewives to trust us
47:04 like uber did at the beginning
47:08 so um interestingly enough this is a
47:13 common thing that happens where a
47:15 start-up thinks they know what happened
47:17 at the beginning of a company and
47:19 sometimes they don't so in the extreme
47:22 beginning of uber what we're so
47:24 interesting was that it launched in San
47:26 Francisco and it launched as uber black
47:29 which is really expensive product
47:30 actually it wasn't well subsidized and
47:33 the real value is that it was impossible
47:36 get a taxi in San Francisco absolutely
47:38 impossible so they didn't have to
47:40 incentivize people to use the first
47:41 version very much because like the
47:44 alternative was like you couldn't move
47:46 around the city at all now as they
47:49 expand to new markets and they figured
47:51 out how marketing worked and they
47:52 figured out that like they could
47:53 accelerate penetrating into a new city
47:56 if they gave away rewards that was all
47:58 stuff they did later much like yes we
48:00 don't copy what you think someone did in
48:02 the beginning when actually that wasn't
48:04 what they did in the beginning in the
48:05 beginning they solved like a really
48:07 hilarious pain point and if uber had
48:09 launched in New York City it wouldn't
48:10 have worked because like yeah did not
48:13 exist in New York City the pain was so
48:15 great great in San Francisco that
48:16 everyone was willing to use uber black
48:18 like they turned a lot of it such a
48:21 money so they did the exact thing that
48:22 in charge way too much and they
48:25 found like a very specific pain point
48:27 that like certain very wealthy people
48:29 had I remember the first time I got an
48:32 overt that was yeah it was good I felt
48:34 like for like a king they had a limo in
48:36 their first initial fleet and every once
48:38 in a while you recall uber and like a
48:39 stretch limo would come and pick you up
48:41 yeah like two people it was crazy they
48:44 charged you for charge extra for the
48:47 excellent idea but it was expensive okay
48:50 okay this is this is a really important
48:52 escort the next question Ashly asks if
48:54 we're gonna be if we're if we're
48:58 starting a technology service that has
48:59 never been done before is it important
49:01 to get it patented before we release the
49:03 MVP and I have I have some ideas on this
49:05 but I want to hear only you your I
49:07 usually abstain far away from my PMS
49:09 it's a buyout company but you have more
49:11 experience than idea so this is such
49:13 like like I I started my company
49:17 and you know a small town in Ontario
49:18 just outside of Toronto like I was not
49:20 in the thick of startup world and I'm
49:22 trying to think back to like how many
49:24 people told me to patent that yeah this
49:26 is like one of those things that people
49:28 who are not in the startup world think
49:30 is a critical piece of starting a
49:32 business because they may have heard of
49:34 like someone else who got screwed out of
49:35 their idea or something like that but if
49:37 you're starting a startup which is kind
49:39 of defined as a technology company
49:40 that's designed to grow rapidly then the
49:44 the number one thing that you need to do
49:46 is find something that people want and
49:48 sell it to them and then iterate and
49:51 grow patents are not a critical step
49:56 necessarily in this path in fact a lot
49:58 of times like I had a conversation with
50:00 a founder NYC and they were talking
50:02 about how they needed to get their stuff
50:04 patented before they launched as a YC
50:05 company and we just I didn't dismiss it
50:08 outright because there's always edge
50:10 cases where it could be could be
50:11 critical but we talk through like what
50:14 are the benefits that you're like then
50:16 you expect to come from this patent and
50:18 most of them are for like circumstances
50:21 that are after you've already hit
50:22 product market fit they're like
50:23 projecting problems that they will have
50:26 in the future there are a multitude of
50:28 problems that you may have you need to
50:30 keep focused on the problems like you
50:32 actually have which is you have no users
50:34 you have no product and you have no
50:35 revenue you solve those ones you will
50:38 have the pleasure of like running into
50:40 patent stuff later in the future should
50:44 we do one more yeah we get a little more
50:46 last one where you talk about design
50:54 there's one of our platform risk where
50:57 is that one this girl daddy was the
50:59 YouTube platform risk alarm there it is
51:02 okay Rahmi asks we have a project that
51:05 solves a very critical and wide pain
51:07 point but it totally depends on youtube
51:09 so if YouTube decides to block it well
51:11 but we we will be out of business is it
51:14 a good risk it doesn't this risk we
51:17 render the business worthless Wow
51:19 I think this is something that's changed
51:21 that feels like it's changed over the
51:26 last 15 years I feel like when I came
51:29 into the startup world there were these
51:32 basically there were these systems that
51:35 were open and so Google felt like a very
51:38 open system for finding websites that
51:41 Google didn't have a whole bunch of
51:42 other products that were competing with
51:44 messaging felt like a very open system
51:47 it was SMS email for like a very open
51:49 system most emails that were sent were
51:52 received even promotional ones
51:54 the tricky thing now is that almost all
51:57 of these systems are now closed and
52:00 almost all of these companies understand
52:02 that they exploited an open system to
52:03 get big and that if their system is open
52:06 if YouTube system isn't open if Facebook
52:08 system is open they will be exploited by
52:11 the next person and so the problem is is
52:13 that you're not only competing with
52:14 other players in your market you're
52:16 competing against the platform you're on
52:18 if this is a significant revenue
52:21 opportunity for the platform you're on
52:22 they have every incentive to either by
52:25 you or kill it and they're not gonna buy
52:28 you for a lot of money because they have
52:29 a lot of leverage over you so the
52:32 question would ask you is like is your
52:33 dependence on youtube today
52:34 or is it dependence on YouTube forever
52:37 if it's that forever question I'd be
52:39 very very cautious about building this
52:41 into a business I might consider making
52:44 it a lifestyle business where you can
52:45 monetize it quickly so that if it does
52:47 go away at least you've made money and
52:49 it brings in this like subscription
52:51 revenue that like youtubers must like
52:53 hardcore youtubers must have this and
52:55 therefore they're willing to pay fifty
52:56 or hundred bucks a month or something
52:57 yeah the only other way you can get away
53:00 with this is kind of like the eBay
53:03 PayPal case we're like your product
53:06 becomes so mainstream on
53:08 you tube so quickly that if you to tried
53:10 to shut it down like there be a mass
53:12 revolt yeah which is hard but
53:15 interesting enough like do you remember
53:16 vine like Twitter shut down the vine
53:19 replace or not fine periscope yep
53:22 and that was popular like that was
53:24 popular with all the cool kids in San
53:26 Francisco and they still got like me
53:27 captain so yeah alright cool everyone
53:31 thank you very much for asking all these
53:33 excellent questions we'll be back next
53:35 Wednesday at 11 a.m. Pacific time to
53:39 answer questions about next week's
53:40 lectures so thanks see you later
![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)
