1 00:00:00,530 --> 00:00:01,109 All right. 2 00:00:01,110 --> 00:00:05,800 So when I talk about making products users love what I 3 00:00:05,800 --> 00:00:09,500 mean specifically is like how do we make things that 4 00:00:09,500 --> 00:00:12,800 has a passionate user base that our users 5 00:00:12,800 --> 00:00:16,950 are unconditionally wanting it to 6 00:00:16,950 --> 00:00:19,590 be successful both on the products that we build, but 7 00:00:19,590 --> 00:00:20,960 also the companies behind them. 8 00:00:22,400 --> 00:00:25,680 We're gonna go over tons of information. 9 00:00:25,680 --> 00:00:27,259 Try not to take too many notes, 10 00:00:27,260 --> 00:00:28,290 mostly just try to listen. 11 00:00:28,290 --> 00:00:30,750 I'll post the link to the slides on my Twitter account 12 00:00:30,750 --> 00:00:33,470 and on that link, there will be a way for you to 13 00:00:33,470 --> 00:00:35,570 annotate the slide and you can ask me questions. 14 00:00:35,570 --> 00:00:36,840 And so if we don't get to them, 15 00:00:36,840 --> 00:00:38,400 I'll answer them after the talk. 16 00:00:39,120 --> 00:00:40,839 So, you guys have been. 17 00:00:40,840 --> 00:00:41,480 Listening to and 18 00:00:41,480 --> 00:00:45,290 hearing a lot about growth over the last several weeks, 19 00:00:45,300 --> 00:00:48,500 and to me I feel like growth is usually fairly simple. 20 00:00:48,500 --> 00:00:50,410 It's the interaction between two sort of concepts or 21 00:00:50,410 --> 00:00:51,449 variables. 22 00:00:51,450 --> 00:00:54,430 Conversion rate and churn. 23 00:00:54,430 --> 00:00:55,100 Right? 24 00:00:55,100 --> 00:00:57,950 And the gap between those two things pretty 25 00:00:57,950 --> 00:01:00,920 much indicate how fast you're gonna grow. 26 00:01:00,920 --> 00:01:04,190 And most people, especially business type people, 27 00:01:04,190 --> 00:01:06,289 tend to look at this interaction in terms of 28 00:01:06,290 --> 00:01:08,980 a very calculated in a mathematical sort of way. 29 00:01:08,980 --> 00:01:10,700 And today, I sort of want to 30 00:01:10,700 --> 00:01:12,179 talk about these things at a more human scale. 31 00:01:12,180 --> 00:01:14,720 All right, cuz at start up when you're interacting with 32 00:01:14,720 --> 00:01:16,700 your users, you have a fairly intimate 33 00:01:16,700 --> 00:01:19,100 interaction that you have in the early stages. 34 00:01:19,100 --> 00:01:21,200 And so, I think there's a different way of looking at 35 00:01:21,200 --> 00:01:23,200 this stuff in terms of how we build our products. 36 00:01:23,200 --> 00:01:25,800 And we'll look at a lot of different examples of that. 37 00:01:25,800 --> 00:01:26,190 And how it's executed well. 38 00:01:27,500 --> 00:01:31,300 My philosophy behind a lot of things that I teach 39 00:01:31,300 --> 00:01:33,430 startups is the best way to sort of get to a billion 40 00:01:33,430 --> 00:01:36,000 dollars is to focus on the values that help you 41 00:01:36,000 --> 00:01:38,870 get that first dollar, to acquire that first user. 42 00:01:38,870 --> 00:01:41,800 If you sort of get that right everything else will 43 00:01:41,800 --> 00:01:43,270 sort of take care of itself. 44 00:01:43,270 --> 00:01:45,190 It's a sort of faith thing. 45 00:01:45,190 --> 00:01:49,770 So I came to be a partner at YC by way of being alumni. 46 00:01:49,770 --> 00:01:51,800 I went to the program in winter of 2006, so 47 00:01:51,800 --> 00:01:53,890 it's the second ever program and 48 00:01:53,890 --> 00:01:55,420 I built a product called Wufoo. 49 00:01:55,420 --> 00:01:57,500 Wufoo's an online forum builder. 50 00:01:57,500 --> 00:01:58,830 It helps you create contact forums and 51 00:01:58,830 --> 00:02:01,170 online surveys and simple payment forms. 52 00:02:01,170 --> 00:02:03,350 It's basically a database that looks like it's 53 00:02:03,350 --> 00:02:05,890 designed by Fisher Price. 54 00:02:05,900 --> 00:02:05,810 What's. 55 00:02:05,810 --> 00:02:09,449 Interesting, though, is that because it was fairly easy 56 00:02:09,449 --> 00:02:13,290 to use, we're had customers from every industry, market, 57 00:02:13,290 --> 00:02:15,709 and vertical you can think of, including a, 58 00:02:15,710 --> 00:02:18,270 a majority of the Fortune 500 companies out there. 59 00:02:18,270 --> 00:02:21,430 Ran the company for five years, and 60 00:02:21,430 --> 00:02:24,700 then we were acquired by SurveyMonkey in 2013. 61 00:02:24,700 --> 00:02:26,420 And at the time, 62 00:02:26,420 --> 00:02:28,500 we were a very interesting acquisition. 63 00:02:28,500 --> 00:02:30,460 We were only a team of ten people at the time. 64 00:02:30,460 --> 00:02:32,680 And, while we acquired funding out here in 65 00:02:32,680 --> 00:02:35,820 Silicon Valley through Y Combinator we actually ran 66 00:02:35,820 --> 00:02:37,420 the company from Florida. 67 00:02:37,420 --> 00:02:39,329 We had no office. 68 00:02:39,330 --> 00:02:40,470 Everyone worked from home. 69 00:02:41,670 --> 00:02:44,160 And we're an interesting outlier. 70 00:02:44,160 --> 00:02:47,940 So, each dot here represents a start up that was, 71 00:02:47,940 --> 00:02:50,400 that exited through IPO or acquisition. 72 00:02:50,400 --> 00:02:51,760 And we're this outlier to the left. 73 00:02:51,760 --> 00:02:53,690 The bottom is the funding amount they took. 74 00:02:53,690 --> 00:02:55,530 And the vertical axis is, 75 00:02:55,530 --> 00:02:57,440 the valuation of the company at the time. 76 00:02:57,440 --> 00:02:58,170 To sum it up, 77 00:02:58,170 --> 00:03:01,684 the average startup rates as about $25 million. 78 00:03:01,685 --> 00:03:04,930 And they return to their investors about 676%. 79 00:03:04,930 --> 00:03:09,440 Wufoo raised about $118 thousand total. 80 00:03:09,440 --> 00:03:13,830 And our return to our investors is about 29,000%. 81 00:03:13,830 --> 00:03:14,810 >> So a lot of people were very interested. 82 00:03:14,810 --> 00:03:15,800 And are sort of like. 83 00:03:15,800 --> 00:03:16,395 What makes. 84 00:03:16,395 --> 00:03:17,510 Wufoo a little bit different or 85 00:03:17,510 --> 00:03:19,590 how do we run the company very differently, and 86 00:03:19,590 --> 00:03:20,910 a lot of it was focused on product. 87 00:03:22,540 --> 00:03:24,648 We weren't interested in building software that, 88 00:03:24,648 --> 00:03:28,910 I guess that just people wanted to use. 89 00:03:28,910 --> 00:03:30,700 Right. That reminding you that you 90 00:03:30,700 --> 00:03:30,989 worked in a cubicle, 91 00:03:30,990 --> 00:03:33,560 cuz it was database app at its sort of core. 92 00:03:33,560 --> 00:03:36,560 We wanted a product that people wanted to love. 93 00:03:36,560 --> 00:03:38,600 That people wanted to have a relationship with, and we 94 00:03:38,600 --> 00:03:41,920 were actually very fanatical about how we approached. 95 00:03:41,920 --> 00:03:43,570 This idea. 96 00:03:43,570 --> 00:03:44,930 At the point where it's almost sort of 97 00:03:46,920 --> 00:03:48,730 sort of sciency sort of way. 98 00:03:50,170 --> 00:03:52,440 So what we said was like Okay. 99 00:03:52,440 --> 00:03:55,300 What's interesting about start-ups in terms of 100 00:03:55,300 --> 00:03:58,580 us wanting to create things that people love is that, 101 00:03:58,580 --> 00:04:00,930 love and unconditional sort of feelings are things that 102 00:04:00,930 --> 00:04:02,919 are difficult for us to do in sort of real life. 103 00:04:03,990 --> 00:04:06,580 And at start-ups we have to do it sort of at scale. 104 00:04:06,580 --> 00:04:07,590 So we've decided to do is 105 00:04:07,590 --> 00:04:09,510 just start of by just looking at, Okay, 106 00:04:09,510 --> 00:04:12,750 how does real relationships work in the real world, and 107 00:04:12,750 --> 00:04:14,390 how can we apply them to sort of how we 108 00:04:14,390 --> 00:04:17,630 run our business and sort of build our product that way. 109 00:04:17,630 --> 00:04:20,399 So we'll go over basically these two metaphors, 110 00:04:20,399 --> 00:04:22,190 find new users as if we're trying to date them and 111 00:04:22,190 --> 00:04:25,520 existing users as if it were a successful marriage. 112 00:04:25,520 --> 00:04:28,460 So when it comes to dating, lot of the stuff that 113 00:04:28,460 --> 00:04:31,630 we uncovered had to do with first impressions. 114 00:04:33,160 --> 00:04:36,270 all, all of you often talk about your relationships in 115 00:04:36,270 --> 00:04:37,440 terms of the origin story. 116 00:04:37,440 --> 00:04:38,950 If I asked you to tell me about the first kiss, 117 00:04:38,950 --> 00:04:40,760 how you sort of met, how you proposed, 118 00:04:40,760 --> 00:04:42,530 these are the things that we say over and 119 00:04:42,530 --> 00:04:43,250 over and over again. 120 00:04:43,250 --> 00:04:45,580 They're basically the word-of-mouth stories of our 121 00:04:45,580 --> 00:04:47,590 relationships, and they're the same kinds of 122 00:04:47,590 --> 00:04:49,989 things that we do with companies. 123 00:04:49,990 --> 00:04:50,560 Human beings 124 00:04:50,560 --> 00:04:52,260 are relationship manufacturing creatures. 125 00:04:52,260 --> 00:04:56,440 We cannot help, but create and anthropomorphize 126 00:04:56,440 --> 00:04:59,590 the things we interact with over and over again. 127 00:04:59,590 --> 00:05:01,200 So, whether it's the cars we drive, 128 00:05:01,200 --> 00:05:04,219 the clothes we wear, the tools and software's we use, 129 00:05:04,220 --> 00:05:06,470 we eventually prescribe characteristics to it. 130 00:05:06,470 --> 00:05:07,250 A personality. 131 00:05:07,250 --> 00:05:08,570 And we expect it to behave a certain way. 132 00:05:08,570 --> 00:05:11,349 And that's how we sort of interact with it. 133 00:05:11,350 --> 00:05:13,140 Now first impressions are important for 134 00:05:13,140 --> 00:05:14,760 the starting of any relationship, 135 00:05:14,760 --> 00:05:16,420 because it's the one we tell over and over again. 136 00:05:16,420 --> 00:05:16,960 Right? 137 00:05:16,960 --> 00:05:19,710 And there's something special about how we 138 00:05:19,710 --> 00:05:20,909 regard that origin story. 139 00:05:20,910 --> 00:05:22,300 I'll give an example. 140 00:05:22,300 --> 00:05:24,100 If you're on a first date with somebody, 141 00:05:24,100 --> 00:05:26,900 and you're having a nice dinner, and 142 00:05:26,900 --> 00:05:28,510 you catch them picking their nose. 143 00:05:28,510 --> 00:05:31,930 You are probably not gonna have another date with them. 144 00:05:31,930 --> 00:05:33,440 Right, but if you're married to someone for 145 00:05:33,440 --> 00:05:34,130 about 20 to 30 years, 146 00:05:34,130 --> 00:05:37,560 and you catch on the Barca Lounger digging for 147 00:05:37,560 --> 00:05:38,640 gold right? 148 00:05:38,640 --> 00:05:40,380 You don't immediately like call your lawyer, 149 00:05:40,380 --> 00:05:42,659 right, and then say like we have a problem here. 150 00:05:42,660 --> 00:05:45,310 I need to start drawing up papers for divorce. 151 00:05:45,310 --> 00:05:46,850 You shrug your shoulders and 152 00:05:46,850 --> 00:05:48,510 say at least he has a heart of gold. 153 00:05:49,560 --> 00:05:51,470 So solving about. 154 00:05:51,470 --> 00:05:54,740 First time interactions means that the threshold is 155 00:05:54,740 --> 00:05:58,150 so much lower in terms of pass fail. 156 00:05:58,150 --> 00:06:00,590 So, in software, and for 157 00:06:00,590 --> 00:06:03,250 most products in internet software that we use, like, 158 00:06:03,250 --> 00:06:05,290 first impressions are pretty obvious. 159 00:06:05,290 --> 00:06:06,420 And they're the things that you see a lot of 160 00:06:06,420 --> 00:06:07,890 companies sort of pay attention to in terms of 161 00:06:07,890 --> 00:06:10,409 what they send their marketing people to work on. 162 00:06:10,410 --> 00:06:13,830 My argument for people who are very good at product is 163 00:06:13,830 --> 00:06:16,300 they discover so many other first moments, and 164 00:06:16,300 --> 00:06:18,890 they make those something memorable. 165 00:06:18,890 --> 00:06:21,990 Right, the first email you ever get from a piece of 166 00:06:21,100 --> 00:06:21,760 software. 167 00:06:21,760 --> 00:06:24,900 What happens when you first log in? 168 00:06:24,900 --> 00:06:26,479 The links, the advertising, the very first time you 169 00:06:26,480 --> 00:06:27,730 interact with customer support. 170 00:06:27,730 --> 00:06:31,940 All of those are opportunities to seduce. 171 00:06:31,940 --> 00:06:33,890 So how did we think about. 172 00:06:33,890 --> 00:06:35,849 Sort of like making first moments on there, 173 00:06:35,850 --> 00:06:38,610 and we actually took this concept from the Japanese. 174 00:06:38,610 --> 00:06:41,120 They actually have two words for 175 00:06:41,120 --> 00:06:43,200 how to describe things when you're finished with them 176 00:06:43,200 --> 00:06:45,539 in terms of saying like is this a quality item. 177 00:06:45,540 --> 00:06:47,396 And the two words of quality are atarimae hinshitsu and 178 00:06:47,396 --> 00:06:48,290 miryokuteki hinshitsu. 179 00:06:48,290 --> 00:06:51,470 And the first one means, taken for 180 00:06:51,470 --> 00:06:53,200 granted quality, basically functionality. 181 00:06:53,200 --> 00:06:55,120 And the last one sort of means, 182 00:06:55,120 --> 00:06:56,550 enchanting quality, right? 183 00:06:58,100 --> 00:06:59,750 Take for example a pen, right? 184 00:06:59,750 --> 00:07:04,170 Something has miryokuteki, right, if the weight of 185 00:07:04,170 --> 00:07:06,220 the pen, the way the ink flows out of it, the way 186 00:07:06,220 --> 00:07:08,960 it's viewed by the people reading the handwriting from 187 00:07:08,960 --> 00:07:12,200 the pen is pleasurable both to the user of the pen and 188 00:07:12,200 --> 00:07:15,109 the people who experience the byproducts of it, 189 00:07:15,110 --> 00:07:16,840 right, taking it to the sort of next level. 190 00:07:19,150 --> 00:07:20,150 Start with some examples. 191 00:07:20,150 --> 00:07:23,429 So this is Wufoo's Login link, and 192 00:07:23,430 --> 00:07:26,280 it has a dinosaur on it, which I think is awesome. 193 00:07:26,280 --> 00:07:29,800 But if you hover over it, 194 00:07:29,800 --> 00:07:32,600 the spec has the added benefit of having a tool tip 195 00:07:32,600 --> 00:07:33,690 that doesn't explain like how to login or 196 00:07:33,690 --> 00:07:36,260 what it does, but basically, rawr. 197 00:07:36,260 --> 00:07:37,800 And. 198 00:07:37,800 --> 00:07:39,890 What we noticed about this, like in early usability 199 00:07:39,890 --> 00:07:42,280 studies as like this put a smile on people's faces. 200 00:07:42,280 --> 00:07:43,630 Like, hands down. 201 00:07:43,630 --> 00:07:44,170 Right? 202 00:07:44,170 --> 00:07:45,300 Universally. 203 00:07:45,300 --> 00:07:46,640 And I think a lot of times when we 204 00:07:46,640 --> 00:07:48,530 are assessing products, we never think about, 205 00:07:48,530 --> 00:07:50,190 like, hey what is the emotion on 206 00:07:50,190 --> 00:07:52,490 the person's face when they interact with this? 207 00:07:53,580 --> 00:07:55,500 This is Vimeo's log in page this is 208 00:07:55,500 --> 00:07:57,700 actually a couple integr, iterations ago. 209 00:07:57,700 --> 00:07:59,540 It's one I find to be the most beautiful, but. 210 00:08:00,570 --> 00:08:02,200 It lets you know that when you're starting out on 211 00:08:02,200 --> 00:08:03,180 this journey with Vimeo, 212 00:08:03,180 --> 00:08:04,640 that this is gonna be something different. 213 00:08:05,820 --> 00:08:07,170 They do this all over the app. 214 00:08:07,170 --> 00:08:09,810 If you search for the word fart, as you scroll up and 215 00:08:09,810 --> 00:08:12,190 down it makes fart noises as you do this. 216 00:08:12,190 --> 00:08:12,730 Right? 217 00:08:12,730 --> 00:08:13,920 There's something different, 218 00:08:13,920 --> 00:08:15,705 like this site interacts with you. 219 00:08:15,705 --> 00:08:16,289 >> Mm-hm. >> It's a little bit 220 00:08:16,290 --> 00:08:17,530 magical, it's a little bit different. 221 00:08:17,530 --> 00:08:19,510 And it's something that you wanna talk about. 222 00:08:20,940 --> 00:08:23,600 You don't have to always do it with design. 223 00:08:23,600 --> 00:08:24,700 This is a sign up form for 224 00:08:24,700 --> 00:08:26,430 Cork which used to be a social network for 225 00:08:26,430 --> 00:08:28,700 people who loved to drink wine. 226 00:08:28,700 --> 00:08:31,190 On it, it says email address, 227 00:08:31,190 --> 00:08:33,240 it's also your sign-in and has to be legit. 228 00:08:33,240 --> 00:08:35,500 First name, what your mom calls you, last name, 229 00:08:35,500 --> 00:08:37,260 what your Army buddies call you. 230 00:08:37,260 --> 00:08:39,799 Password, something you'll remember, but hard to guess, 231 00:08:39,799 --> 00:08:42,339 password confirmation, think it a, type it again. 232 00:08:42,340 --> 00:08:43,559 Think of it as a test. 233 00:08:43,558 --> 00:08:46,699 It's literally a poem as you fill out the form. 234 00:08:46,700 --> 00:08:47,540 Right? 235 00:08:47,540 --> 00:08:50,689 And this is a kind of like thing where you like, 236 00:08:50,690 --> 00:08:52,110 oh, I like the people behind this. 237 00:08:52,110 --> 00:08:53,850 I, I, I'm gonna enjoy this experience. 238 00:08:53,850 --> 00:08:55,970 Now what does it say when you fill out form like this? 239 00:08:55,970 --> 00:08:56,470 Right. 240 00:08:56,470 --> 00:08:58,960 On Yahoo, about what the personality. 241 00:08:58,960 --> 00:08:59,890 Of the site it's gonna be. 242 00:08:59,890 --> 00:09:02,500 And what's disappointing to me is like, 243 00:09:02,500 --> 00:09:04,570 Yahoo forces every product and 244 00:09:04,570 --> 00:09:07,960 service under them to use this exact same login form. 245 00:09:07,960 --> 00:09:09,450 Flickr, I had thought, 246 00:09:09,450 --> 00:09:11,280 had one of the best sort of call to actions. 247 00:09:11,280 --> 00:09:13,699 It was, get in there. 248 00:09:13,700 --> 00:09:14,200 Right? 249 00:09:16,200 --> 00:09:17,259 This is Heroku's. 250 00:09:19,900 --> 00:09:19,650 Sign up page. 251 00:09:19,650 --> 00:09:21,230 I think this is an older version. 252 00:09:21,230 --> 00:09:24,340 But, what's remarkable about it is that, what you 253 00:09:24,340 --> 00:09:28,180 start getting a feel for is like, oh, scaling up. 254 00:09:28,180 --> 00:09:29,000 My sort of server, and 255 00:09:29,000 --> 00:09:33,190 back end services is as easy as sort of dragging up and 256 00:09:33,190 --> 00:09:36,460 down, different sort of nobs and levers. 257 00:09:36,460 --> 00:09:37,820 It's gonna be beautifully used, and 258 00:09:37,820 --> 00:09:39,570 it looks fairly easy to scale. 259 00:09:41,260 --> 00:09:45,460 Since we're in a room full of computer science people, 260 00:09:45,460 --> 00:09:46,880 I think you'll appreciate this. 261 00:09:46,880 --> 00:09:47,660 This is Chocolat. 262 00:09:47,660 --> 00:09:49,750 This is a code editor, and 263 00:09:49,750 --> 00:09:51,320 they only have one call to action. 264 00:09:51,320 --> 00:09:55,270 When the time limit is up, they say, everything in 265 00:09:55,270 --> 00:09:58,500 terms of all the pieces are exactly the same except we 266 00:09:58,500 --> 00:09:59,599 change the font. 267 00:09:59,600 --> 00:10:00,600 To comic sans. 268 00:10:03,670 --> 00:10:06,540 And what they're basically saying is like hey 269 00:10:06,540 --> 00:10:09,932 we know who our users are, who our real customers are. 270 00:10:09,932 --> 00:10:12,300 They're gonna be the people who care about this. 271 00:10:15,470 --> 00:10:17,480 This is Hurl, this is a website for 272 00:10:17,480 --> 00:10:22,580 checking htp requests, and sometimes the places where 273 00:10:22,580 --> 00:10:24,650 you get errors are opportunities for 274 00:10:24,650 --> 00:10:26,526 first moments, 275 00:10:26,527 --> 00:10:29,390 you had a four of four this is what you get it, 276 00:10:29,390 --> 00:10:32,400 when we need help oftentimes what we do is where we 277 00:10:32,400 --> 00:10:39,770 create like really beautiful mark, marketing materials. 278 00:10:39,770 --> 00:10:42,819 But when you actually need, like, documentation we sort 279 00:10:42,820 --> 00:10:45,850 of like skimp out on sort of design features. 280 00:10:45,850 --> 00:10:48,600 And this is a point that, like, 281 00:10:48,600 --> 00:10:49,829 you see happen over and over again. 282 00:10:49,830 --> 00:10:51,520 A company that gets this right is MailChimp. 283 00:10:51,520 --> 00:10:53,470 And what they did was they redesigned all of their 284 00:10:53,470 --> 00:10:56,350 help guides so that they looked like magazine covers. 285 00:10:56,350 --> 00:10:57,890 And overnight, basically, 286 00:10:57,890 --> 00:10:59,920 readership goes up on all these features. 287 00:10:59,920 --> 00:11:02,790 And customer support for these things that sort of 288 00:11:02,800 --> 00:11:04,220 help people optimize emails goes down. 289 00:11:06,350 --> 00:11:08,400 Speaking of documentation, stripe. 290 00:11:09,810 --> 00:11:13,550 What's interesting about an API company is that there is 291 00:11:13,550 --> 00:11:14,339 no UX. 292 00:11:14,340 --> 00:11:16,810 The UX is actually just documentation, right? 293 00:11:16,810 --> 00:11:19,390 And there's opportunities even in documentation. 294 00:11:19,390 --> 00:11:20,720 Sort of the enchant and amaze. 295 00:11:20,720 --> 00:11:21,660 So one of the things that I 296 00:11:21,660 --> 00:11:24,140 love about them is their examples are wonderful. 297 00:11:24,140 --> 00:11:25,280 But if you're actually, like, 298 00:11:25,280 --> 00:11:26,829 sort of logging into the app, 299 00:11:26,830 --> 00:11:29,340 one of the things that is a super pain for most people, 300 00:11:29,340 --> 00:11:30,770 when you're dealing with most people's APIs, is, 301 00:11:30,770 --> 00:11:33,920 like, grabbing your API credentials and keys. 302 00:11:33,920 --> 00:11:35,610 And what Stripe does is it says oh, 303 00:11:35,610 --> 00:11:36,610 if you're logged into the app. 304 00:11:36,610 --> 00:11:37,390 We automatically put 305 00:11:37,390 --> 00:11:40,110 your API credentials into the examples, so you only 306 00:11:40,110 --> 00:11:43,900 have to copy paste once when trying to learn their API. 307 00:11:47,600 --> 00:11:50,500 When Wufoo wanted to launch the third version of our API 308 00:11:50,500 --> 00:11:51,939 we realized like, Okay, that finally this is good enough 309 00:11:51,940 --> 00:11:54,700 that we want people to sort of build on top of it. 310 00:11:54,700 --> 00:11:55,400 We were trying to figure out, 311 00:11:55,400 --> 00:11:56,740 like, how do we launch this out to the world, 312 00:11:56,740 --> 00:11:58,550 that sort of has our personality behind it. 313 00:11:58,550 --> 00:11:59,430 Because a lot of people, 314 00:11:59,430 --> 00:12:02,000 they usually do things like a programming API contest, 315 00:12:02,000 --> 00:12:03,960 and they give out iPads and iPhones. 316 00:12:03,960 --> 00:12:05,830 And it makes you look like everyone else. 317 00:12:05,830 --> 00:12:09,440 And so, in our company one weird value to 318 00:12:09,440 --> 00:12:10,740 have it's a quirk of us, 319 00:12:10,740 --> 00:12:13,570 is that the co-founders are big medieval nuts. 320 00:12:13,570 --> 00:12:16,510 And we take everyone out to Medieval Times every single 321 00:12:16,510 --> 00:12:19,130 year, on the anniversary of the founding of the company. 322 00:12:19,130 --> 00:12:21,760 And so, we said we have to do something in that flavor. 323 00:12:21,760 --> 00:12:23,650 And so we contacted the guys at armor.com, 324 00:12:23,650 --> 00:12:26,800 we said can you forward us a custom battle-ax. 325 00:12:26,800 --> 00:12:27,630 And what we said was if 326 00:12:27,630 --> 00:12:31,580 you win our programming contest you would win one. 327 00:12:33,390 --> 00:12:35,800 The result is like people wanted to talk about this. 328 00:12:35,800 --> 00:12:37,380 It's something that people wanted to work on cuz 329 00:12:37,380 --> 00:12:39,860 they wanted to be able to say it like I'm programming. 330 00:12:39,860 --> 00:12:40,420 For a weapon. 331 00:12:43,269 --> 00:12:46,850 And what's cool is we had over 25 different 332 00:12:46,850 --> 00:12:50,790 applications created for us of quality and quantity that 333 00:12:50,790 --> 00:12:53,410 we could not have paid for on the budget and 334 00:12:53,410 --> 00:12:54,819 the sort of time that we had for this. 335 00:12:54,820 --> 00:12:56,150 We got things like an iPhone app and 336 00:12:56,150 --> 00:12:58,720 Android and WordPress plugins, right? 337 00:12:58,720 --> 00:13:00,850 And all because what we did was we changed how 338 00:13:00,850 --> 00:13:02,520 people wanna talk about the origin story of 339 00:13:02,520 --> 00:13:04,779 how they're interacting with one of our services. 340 00:13:06,100 --> 00:13:08,990 And go like all day long of going over these examples. 341 00:13:08,990 --> 00:13:11,000 But I'm gonna short cut this by saying, 342 00:13:11,000 --> 00:13:12,750 you should just subscribe a little bit of details. 343 00:13:12,750 --> 00:13:16,480 It's just basically tons of screen shots of software 344 00:13:16,480 --> 00:13:17,370 that's just doing it right. 345 00:13:17,370 --> 00:13:18,160 That shows that they're being 346 00:13:18,160 --> 00:13:21,980 conscientious of the user and the customers. 347 00:13:28,467 --> 00:13:30,799 When it comes to long term relationships or 348 00:13:30,799 --> 00:13:33,843 marriages, the only research that we ended up having to 349 00:13:33,843 --> 00:13:37,390 read was the stuff that was done by John Gottman. 350 00:13:37,390 --> 00:13:39,520 He's been featured in This American Life, 351 00:13:39,520 --> 00:13:41,490 in Malcolm Gladwell's books. 352 00:13:41,490 --> 00:13:43,400 He's a marriage researcher up in Seattle. 353 00:13:43,400 --> 00:13:45,250 And he has an interesting parlor trick that he can do. 354 00:13:45,250 --> 00:13:48,730 He can watch a video tape of a couple fighting about some 355 00:13:48,730 --> 00:13:50,320 issue for 15 minutes. 356 00:13:50,320 --> 00:13:52,900 And predict with an 85% accuracy rate whether that 357 00:13:52,900 --> 00:13:53,930 couple will be together or 358 00:13:53,930 --> 00:13:55,630 not, or divorced in four years. 359 00:13:56,780 --> 00:13:58,770 If he increases that video up to an hour. 360 00:13:58,770 --> 00:14:00,620 And asks them to also talk about their hopes and 361 00:14:00,620 --> 00:14:01,540 dreams. 362 00:14:01,540 --> 00:14:03,166 That prediction rating goes up to 94%. 363 00:14:03,167 --> 00:14:06,150 They showed these same video tapes to marriage 364 00:14:06,150 --> 00:14:07,699 counselors, successfully married couples, 365 00:14:07,700 --> 00:14:11,630 sociologists, psychiatrists, priests, et cetera and 366 00:14:11,630 --> 00:14:14,680 they can't predict with random chance whether people 367 00:14:14,680 --> 00:14:16,250 are gonna be together or not. 368 00:14:16,250 --> 00:14:18,590 So John Gauntman understands something fundamental about. 369 00:14:19,910 --> 00:14:22,110 How relationships work in the long term. 370 00:14:22,110 --> 00:14:25,650 And that basically how we fight, even in a short term 371 00:14:25,650 --> 00:14:29,300 period can indicate sort of the whole system and 372 00:14:29,300 --> 00:14:30,140 what it's gonna look like. 373 00:14:30,140 --> 00:14:31,900 And one of the surprising things he discovered, 374 00:14:31,900 --> 00:14:32,420 it's not that 375 00:14:32,420 --> 00:14:34,430 successfully married people don't fight at all. 376 00:14:34,430 --> 00:14:36,430 It turns out everybody fights. 377 00:14:36,430 --> 00:14:39,890 And we all fight about the exact same things money, 378 00:14:39,900 --> 00:14:41,300 kids, sex, time and others. 379 00:14:41,300 --> 00:14:45,262 And others are things like jealousy and in-laws. 380 00:14:45,263 --> 00:14:47,543 To bring this around, right, you can actually attribute 381 00:14:47,543 --> 00:14:49,543 every single one of these to problems that you see in 382 00:14:49,543 --> 00:14:51,931 customer support when you're building out your products. 383 00:14:53,368 --> 00:14:56,532 Right, so. 384 00:14:56,532 --> 00:14:58,447 This costs too much. 385 00:14:58,447 --> 00:15:00,401 I'm having problems with a credit card. 386 00:15:00,401 --> 00:15:02,260 If you're building a service that helps people deal with 387 00:15:02,260 --> 00:15:03,189 their clients they're very 388 00:15:03,190 --> 00:15:05,480 sensitive about anything happening with that. 389 00:15:05,480 --> 00:15:06,530 Performance. 390 00:15:06,530 --> 00:15:08,750 How long you're up and how fast. 391 00:15:10,780 --> 00:15:14,110 Others are I said jealousy and in-laws. 392 00:15:14,110 --> 00:15:15,960 Right, so that's competition and partnerships. 393 00:15:15,960 --> 00:15:17,310 So anything weird happening there, 394 00:15:17,310 --> 00:15:19,890 people are gonna write to you about. 395 00:15:19,900 --> 00:15:20,520 And the reason I like to think about this in 396 00:15:20,520 --> 00:15:23,430 terms of customer support is that in every ones that 397 00:15:23,430 --> 00:15:25,920 are processing of like a conversion funnel, 398 00:15:25,920 --> 00:15:27,790 customer support is the thing that happens in 399 00:15:27,790 --> 00:15:29,990 between every one of these steps. 400 00:15:29,100 --> 00:15:30,720 It's the reason why people don't make it 401 00:15:30,720 --> 00:15:31,650 further down there. 402 00:15:31,650 --> 00:15:33,629 It's the thing that prevents conversion from happening. 403 00:15:35,460 --> 00:15:38,130 Now, as we were thinking through all these ideas and 404 00:15:38,130 --> 00:15:39,680 as we were building up the company, 405 00:15:39,680 --> 00:15:42,469 we realized that there's a big problem about how 406 00:15:42,470 --> 00:15:43,760 everyone sort of starts their company or 407 00:15:43,760 --> 00:15:46,230 build up their sort of engineering teams. 408 00:15:46,230 --> 00:15:48,760 And that is that there's a broken feedback loop there. 409 00:15:48,760 --> 00:15:50,560 People are divorced from the consequences of 410 00:15:50,560 --> 00:15:51,810 their actions. 411 00:15:51,810 --> 00:15:53,579 And this is the result of actually the natural 412 00:15:53,580 --> 00:15:55,600 evolution of how most companies get founded, 413 00:15:55,600 --> 00:15:57,880 especially by technical co-founders, right. 414 00:15:57,880 --> 00:16:01,630 Before launch it is a time of bliss, nirvana, and 415 00:16:01,630 --> 00:16:03,500 opportunity, right? 416 00:16:03,500 --> 00:16:05,469 Nothing that you do is wrong, right? 417 00:16:06,660 --> 00:16:09,850 By your hand, which you feel is like god, everything that 418 00:16:09,850 --> 00:16:12,750 you write, every line of code feels perfect, right, 419 00:16:12,750 --> 00:16:14,770 and is ingenious to you. 420 00:16:14,770 --> 00:16:17,829 The thing that happens is after launch reality sort of 421 00:16:17,830 --> 00:16:20,300 sets in and then all these other tasks sort of 422 00:16:20,300 --> 00:16:21,579 come into place that we have to deal with. 423 00:16:23,330 --> 00:16:25,720 Now what technical co-founders wanna do is get 424 00:16:25,720 --> 00:16:29,240 back to that initial state and so what we often do, 425 00:16:29,240 --> 00:16:31,380 and what we often see is that companies start 426 00:16:31,380 --> 00:16:34,240 siloing off all these other things that actually is 427 00:16:34,240 --> 00:16:37,410 what makes a start up or a company sort of real, right? 428 00:16:37,410 --> 00:16:39,540 And have other people do them. 429 00:16:39,540 --> 00:16:42,589 To, in our minds, these other tasks are inferior. 430 00:16:42,590 --> 00:16:43,660 Right? And we have other people in 431 00:16:43,660 --> 00:16:44,310 the company do them. 432 00:16:45,822 --> 00:16:47,580 And so for us, what we're trying to figure out 433 00:16:47,580 --> 00:16:49,590 is how do we change software development so that we 434 00:16:49,590 --> 00:16:53,970 inject some values that we don't talk about enough? 435 00:16:53,970 --> 00:16:55,970 Responsibility, accountability, 436 00:16:55,970 --> 00:16:57,780 humility, modesty. 437 00:16:57,780 --> 00:17:01,300 Right, and we called this, like a lot of other people, 438 00:17:01,300 --> 00:17:05,720 we had an acronym, Support Driven Development, and 439 00:17:05,720 --> 00:17:09,900 it's very similar to TDD or other agile practice. 440 00:17:09,900 --> 00:17:12,589 It's a way of creating high quality software, but 441 00:17:12,589 --> 00:17:13,819 it's super simple. 442 00:17:13,819 --> 00:17:15,182 You don't need like, a SCRUM, 443 00:17:15,182 --> 00:17:17,609 you don't need a bunch of post-it notes, all you 444 00:17:17,608 --> 00:17:19,909 have to do is make everyone do customer support. 445 00:17:21,880 --> 00:17:22,800 And what you end up having is 446 00:17:22,800 --> 00:17:23,990 you fix the feedback loop, right? 447 00:17:23,990 --> 00:17:25,450 The people who build the software are the ones 448 00:17:25,450 --> 00:17:26,240 supporting it and 449 00:17:26,240 --> 00:17:28,940 you get all these sort of nice benefits as a result. 450 00:17:30,320 --> 00:17:33,340 So one of them is, support responsible developers and 451 00:17:33,340 --> 00:17:35,120 designers and people who build the stuff, 452 00:17:35,120 --> 00:17:36,600 they give the very best support. 453 00:17:36,600 --> 00:17:38,600 Now we're not the first person to think of this. 454 00:17:38,600 --> 00:17:41,480 Paul English was a big proporter of this at Kayak, 455 00:17:41,480 --> 00:17:44,750 and what he did was install a red customer support phone 456 00:17:44,750 --> 00:17:46,640 line in the middle of the engineering floor. 457 00:17:46,640 --> 00:17:49,950 And they were just regular customer support calls. 458 00:17:49,950 --> 00:17:53,890 And people would ask him often times, 459 00:17:53,890 --> 00:17:55,560 why would you pay engineers $120,000 or 460 00:17:55,560 --> 00:17:59,389 more to do something that you can pay other people 461 00:17:59,390 --> 00:18:02,890 a fraction of to handle in like a call center? 462 00:18:02,890 --> 00:18:04,560 And he says, well, after the second or 463 00:18:04,560 --> 00:18:06,929 third time that phone rings and the engineer gets 464 00:18:06,930 --> 00:18:09,990 the same problem they stop what they're doing, they fix 465 00:18:09,990 --> 00:18:13,940 the bug, and we stop getting phone calls about it. 466 00:18:13,940 --> 00:18:16,670 It, it's a way of having Q-A in a sort of 467 00:18:16,670 --> 00:18:17,840 nice elegant solution. 468 00:18:19,710 --> 00:18:23,820 Now, John Gotman talks about the reason that we often 469 00:18:23,820 --> 00:18:26,600 break up with one another as due to four major causes and 470 00:18:26,600 --> 00:18:27,389 their warning signs. 471 00:18:27,390 --> 00:18:29,300 He calls them the four horsemen, right? 472 00:18:29,300 --> 00:18:32,830 Criticism, contempt, defensiveness, 473 00:18:32,830 --> 00:18:33,679 and stonewalling. 474 00:18:33,680 --> 00:18:38,590 Now, criticism is basically people starting to focus, 475 00:18:38,590 --> 00:18:40,350 not just on the specific issue at hand, but 476 00:18:40,350 --> 00:18:41,447 on the overarching issues. 477 00:18:41,448 --> 00:18:45,780 Like, you never, right, listen to your users or 478 00:18:45,780 --> 00:18:48,480 you never think about us all the time. 479 00:18:48,480 --> 00:18:48,980 Right? 480 00:18:50,390 --> 00:18:53,880 Contempt is when someone is purposely trying to 481 00:18:53,880 --> 00:18:55,170 insult somebody. 482 00:18:55,170 --> 00:18:59,620 Defensiveness is not trying to take accountability 483 00:18:59,620 --> 00:19:01,810 trying to make excuses for the actions. 484 00:19:01,810 --> 00:19:03,532 And stonewalling is basically shutting down. 485 00:19:03,532 --> 00:19:07,120 Stonewalling, to John Gotman, is, is one of 486 00:19:07,120 --> 00:19:09,250 the worst things that we can do in a relationship. 487 00:19:09,260 --> 00:19:10,593 Hold up. 488 00:19:14,588 --> 00:19:17,711 And oftentimes, you know, we don't worry much about this 489 00:19:17,711 --> 00:19:19,970 in customer se, criticism and contempt. 490 00:19:19,970 --> 00:19:20,770 Right? Defensiveness, 491 00:19:20,770 --> 00:19:22,530 you see this all the time, all the times in companies, 492 00:19:22,530 --> 00:19:23,790 especially as they get older. 493 00:19:23,790 --> 00:19:25,320 But stonewalling, this is something I 494 00:19:25,320 --> 00:19:27,230 see happen with start-ups all the time. 495 00:19:28,720 --> 00:19:30,100 You get a bunch of customer support sort of 496 00:19:30,100 --> 00:19:32,379 coming in and you just think I don't need to answer it. 497 00:19:32,380 --> 00:19:33,130 I don't need to respond. 498 00:19:33,130 --> 00:19:34,420 Right? And 499 00:19:34,420 --> 00:19:36,410 that act of just not even getting back to 500 00:19:36,410 --> 00:19:38,290 them is one of the worst things you can do. 501 00:19:38,290 --> 00:19:40,159 And it's probably some of the biggest causes of 502 00:19:40,160 --> 00:19:42,340 churn in the early stages of start-ups. 503 00:19:43,580 --> 00:19:45,429 This is how support worked out with Wufoo. 504 00:19:45,430 --> 00:19:48,250 When we were acquired we had about 500,000 users on 505 00:19:48,250 --> 00:19:51,230 the system, 5 million people used Wufoo forms and 506 00:19:51,230 --> 00:19:52,940 reports, whether they knew it or not, and 507 00:19:52,940 --> 00:19:56,100 all those people got support from the same ten people, 508 00:19:56,100 --> 00:19:56,639 and usually it was 509 00:19:56,640 --> 00:20:00,200 only one person dedicated support a day, or any shift. 510 00:20:00,200 --> 00:20:02,180 Results in about 400 issues a week. 511 00:20:02,180 --> 00:20:04,110 That's about 800 emails. 512 00:20:04,110 --> 00:20:05,928 But a response time from 9 a.m. 513 00:20:05,928 --> 00:20:06,440 to 9 p.m. was 514 00:20:06,440 --> 00:20:07,865 between seven to 12 minutes. 515 00:20:07,865 --> 00:20:08,980 Right? And from 9 p.m. 516 00:20:08,980 --> 00:20:11,130 to midnight it was an hour. 517 00:20:11,130 --> 00:20:13,300 And on the weekend it would be no longer than 24 hours. 518 00:20:13,300 --> 00:20:16,200 And we carried this up all the way up to the scale. 519 00:20:18,300 --> 00:20:19,780 What a lot of people forget about, and 520 00:20:19,780 --> 00:20:23,200 often talk about with Airbnb, is how, like oh, 521 00:20:23,200 --> 00:20:24,870 they did this interesting thing where they had, 522 00:20:24,870 --> 00:20:26,699 went up to New York and offered, like, 523 00:20:26,700 --> 00:20:28,280 professional photographers, and the founders would go 524 00:20:28,280 --> 00:20:30,400 out there and actually take pictures of 525 00:20:30,400 --> 00:20:31,860 the people's apartments to help them sell more. 526 00:20:31,860 --> 00:20:34,389 Focusing on the stories around conversion. 527 00:20:34,390 --> 00:20:36,440 What most people don't realize is a lot of 528 00:20:36,440 --> 00:20:38,783 times when I saw Joe in the early days of Airbnb, 529 00:20:38,783 --> 00:20:41,301 he had a phone, sort of head set stuck to his head all 530 00:20:41,301 --> 00:20:44,720 the time, because he was doing phone support nonstop. 531 00:20:46,286 --> 00:20:48,307 Churn is a story we don't like to talk about often, 532 00:20:48,307 --> 00:20:49,280 all the time. 533 00:20:49,280 --> 00:20:52,540 Airbnb's sort of growth really started picking up 534 00:20:52,540 --> 00:20:53,440 once they figured out 535 00:20:53,440 --> 00:20:56,300 how to match capacity to the demand, or the phone 536 00:20:56,300 --> 00:20:58,730 calls that they were getting into their support system. 537 00:21:00,580 --> 00:21:02,960 At Wufoo we actually constantly did experiments 538 00:21:02,960 --> 00:21:05,550 around support, because we're so obsessed with it. 539 00:21:05,550 --> 00:21:08,540 One experiment we did was, we heard Kathy Sierra 540 00:21:08,540 --> 00:21:10,860 do a talk about there's a disconnect between the 541 00:21:10,860 --> 00:21:14,580 motions that we have when we need help, and sort of. 542 00:21:14,580 --> 00:21:16,620 The content and the reactions we get from 543 00:21:16,620 --> 00:21:18,429 people when we get help from them, especially online. 544 00:21:18,430 --> 00:21:22,780 Because they just don't see all those nonverbal cues. 545 00:21:22,780 --> 00:21:25,440 So she said unless there's face recognition on the web, 546 00:21:25,440 --> 00:21:26,420 we're just always going to be 547 00:21:26,420 --> 00:21:27,980 disconnected from our users. 548 00:21:27,980 --> 00:21:28,760 Our feeling was, 549 00:21:28,760 --> 00:21:31,110 like, well we're not face recognition experts but 550 00:21:31,110 --> 00:21:32,699 I think there's another way of getting empathy. 551 00:21:32,700 --> 00:21:35,710 So, as form builders, we added a drop-down and 552 00:21:35,710 --> 00:21:37,590 what we said was like, hey, what's your emotional state? 553 00:21:39,640 --> 00:21:41,180 And our hypothesis was 554 00:21:41,180 --> 00:21:42,300 that no one was gonna fill this out. 555 00:21:42,300 --> 00:21:45,790 We basically thought, oh okay, you know what the, 556 00:21:45,790 --> 00:21:48,909 this is gonna be pretty a lame experiment, but 557 00:21:48,910 --> 00:21:49,930 we'll see how it sort of goes. 558 00:21:49,930 --> 00:21:51,125 And it turned out the Emotional State 559 00:21:51,125 --> 00:21:53,770 drop-down field was filled out 75.8% of the time. 560 00:21:55,300 --> 00:21:57,374 The browser type drop-down filled just in 561 00:21:57,374 --> 00:22:00,184 comparison was filled out 78.1% of the time. 562 00:22:00,184 --> 00:22:00,816 All right? 563 00:22:00,816 --> 00:22:03,337 So people were basically telling us, for 564 00:22:03,337 --> 00:22:06,506 my technical support issue how I feel about this 565 00:22:06,506 --> 00:22:10,360 problem is just as important as, like, all the technical 566 00:22:10,360 --> 00:22:13,220 details you need to sorta figure out how to debug it. 567 00:22:14,480 --> 00:22:16,580 Now we didn't prioritize things or 568 00:22:16,580 --> 00:22:18,379 triage things by emotion, right, and for 569 00:22:18,380 --> 00:22:19,860 the most part, people didn't game the system. 570 00:22:19,860 --> 00:22:22,820 One of the interesting byproducts of it was that we 571 00:22:22,820 --> 00:22:24,450 noticed that people started being nicer to 572 00:22:24,450 --> 00:22:25,850 us in the customer support. 573 00:22:25,850 --> 00:22:27,270 It was something sort of subconscious. 574 00:22:27,270 --> 00:22:28,590 We just were thinking like, wow, 575 00:22:28,590 --> 00:22:29,870 users are so much better now. 576 00:22:29,870 --> 00:22:31,429 What's going on? 577 00:22:31,430 --> 00:22:33,710 And we went back and looked at the data and 578 00:22:33,710 --> 00:22:36,910 we did some text analysis and we realized is that, oh, 579 00:22:36,910 --> 00:22:39,790 when it comes to only communicating with 580 00:22:39,790 --> 00:22:43,620 people over written words, like email, there's 581 00:22:43,620 --> 00:22:45,750 only three ways that you show strong emotions, right? 582 00:22:45,750 --> 00:22:49,560 Exclamation marks curse words, and all caps. 583 00:22:49,560 --> 00:22:52,139 And sure enough on all three of those metrics, 584 00:22:52,140 --> 00:22:52,840 they've gone down. 585 00:22:52,840 --> 00:22:55,700 In sort of the way people were talking to us in 586 00:22:55,700 --> 00:22:56,340 the customer support. 587 00:22:56,340 --> 00:22:58,780 Once people had a simple outlet for their emotions, 588 00:22:58,780 --> 00:23:00,340 right, people would be a lot more rational, 589 00:23:00,340 --> 00:23:02,996 and it made our jobs a lot more pleasant as a result. 590 00:23:02,997 --> 00:23:08,860 The other byproduct that is awesome is that 591 00:23:08,860 --> 00:23:11,280 you actually build better software when you do this. 592 00:23:11,280 --> 00:23:12,310 Far better software. 593 00:23:12,310 --> 00:23:14,900 This is actually backed up by tons of research. 594 00:23:14,900 --> 00:23:16,189 Jared Spool, a user interface engineer, which is 595 00:23:16,190 --> 00:23:18,745 sort of the big players in this space says like, 596 00:23:18,745 --> 00:23:21,399 there's a direct correlation to how much time we 597 00:23:21,400 --> 00:23:24,810 spend directly exposed to users and 598 00:23:24,810 --> 00:23:27,800 how good our designs sort of get. 599 00:23:27,800 --> 00:23:28,800 He says it has to come in this specific way. 600 00:23:28,800 --> 00:23:30,680 It has to be direct exposure, right? 601 00:23:30,680 --> 00:23:31,710 It can't be something where 602 00:23:31,710 --> 00:23:34,370 someone generates a report or through a graph, you 603 00:23:34,370 --> 00:23:37,700 have to be interacting with them somewhat real time. 604 00:23:37,700 --> 00:23:39,290 It has to be a minimum of every six weeks. 605 00:23:39,290 --> 00:23:41,180 And it has to be for at least two hours. 606 00:23:41,180 --> 00:23:43,720 Otherwise your software will get worse over time. 607 00:23:46,340 --> 00:23:48,860 Our developers, our people who are on, 608 00:23:48,860 --> 00:23:51,200 on Wufoo were getting exposed to our users four to 609 00:23:51,200 --> 00:23:52,820 eight hours every single week. 610 00:23:54,280 --> 00:23:56,570 And what it does is it changes the way you sort of 611 00:23:56,570 --> 00:23:57,437 build software. 612 00:23:57,437 --> 00:24:00,149 Jared Spool has another way of talking about how 613 00:24:00,150 --> 00:24:01,380 we build products. 614 00:24:01,380 --> 00:24:02,650 Right? 615 00:24:02,650 --> 00:24:06,780 Let's imagine that this represents all the knowledge 616 00:24:06,780 --> 00:24:09,899 needed to sort of use your app on a spectrum. 617 00:24:09,900 --> 00:24:11,770 Right? This is like no knowledge. 618 00:24:11,770 --> 00:24:12,340 Right? And this is 619 00:24:12,340 --> 00:24:13,780 all the knowledge needed. 620 00:24:13,780 --> 00:24:14,300 Right? And 621 00:24:14,300 --> 00:24:16,280 these two lines are pretty much your interactions with 622 00:24:16,280 --> 00:24:17,610 users what you're trying to get them to. 623 00:24:17,610 --> 00:24:20,270 This is currently where their knowledge point is and 624 00:24:20,270 --> 00:24:22,460 this is the target knowledge point that you're trying to 625 00:24:22,460 --> 00:24:25,250 get them to, to understand that to use your app. 626 00:24:25,250 --> 00:24:27,102 The gap between those is called the knowledge gap, 627 00:24:27,102 --> 00:24:28,497 Jared Spool called it, Spool calls. 628 00:24:28,497 --> 00:24:31,846 And what's interesting about this is there's only two 629 00:24:31,846 --> 00:24:34,280 ways, right, to sort of fix this. 630 00:24:34,280 --> 00:24:36,129 That gap represents how intuitive your 631 00:24:36,130 --> 00:24:38,100 app is, right? 632 00:24:38,100 --> 00:24:41,210 You either get the user to increase their 633 00:24:41,210 --> 00:24:44,800 knowledge or you decrease the amount of knowledge 634 00:24:44,800 --> 00:24:45,871 that's needed to use your application. 635 00:24:45,872 --> 00:24:48,450 And oftentimes, as engineers and people who build and 636 00:24:48,450 --> 00:24:50,699 work on products, we think let's add new features. 637 00:24:51,800 --> 00:24:52,790 And new features only 638 00:24:52,790 --> 00:24:55,149 means let's increase the knowledge gap. 639 00:24:57,870 --> 00:25:00,700 So for us we actually focused a lot on 640 00:25:00,700 --> 00:25:01,792 the other sort of direction. 641 00:25:01,792 --> 00:25:03,530 And so what that meant we spent a lot of time, 642 00:25:03,530 --> 00:25:06,870 30% of our engineering time was spent on internal tools 643 00:25:06,870 --> 00:25:09,101 to help with our customer support stuff. 644 00:25:09,102 --> 00:25:10,200 But oftentimes it 645 00:25:10,200 --> 00:25:12,750 was spent helping people help themselves. 646 00:25:12,750 --> 00:25:15,870 Things like frequently asked questions, tool tips. 647 00:25:15,870 --> 00:25:17,649 Things like, if you just click the help link, 648 00:25:17,650 --> 00:25:20,510 right, instead of taking you to the generic help sort of 649 00:25:20,510 --> 00:25:21,480 documentation page, 650 00:25:21,480 --> 00:25:24,150 you go to the specific page where you're looking at 651 00:25:24,150 --> 00:25:25,540 is going to be most, sort of, 652 00:25:25,540 --> 00:25:27,960 appropriate for what you're working on. 653 00:25:27,960 --> 00:25:29,700 We redesigned our documentation over and 654 00:25:29,700 --> 00:25:31,870 over again, AB tested it constantly. 655 00:25:31,870 --> 00:25:34,400 One iteration of our documentation reduced 656 00:25:34,400 --> 00:25:36,290 customer support by 30% over night. 657 00:25:37,330 --> 00:25:38,900 It's one of those things where, 658 00:25:38,900 --> 00:25:41,399 like, overnight, all the people that work on 659 00:25:41,400 --> 00:25:43,960 the product immediately had 30% less work to do. 660 00:25:46,800 --> 00:25:48,490 Now what happens if you have everyone work on 661 00:25:48,490 --> 00:25:49,630 customer support constantly, and 662 00:25:49,630 --> 00:25:52,220 thinking about it in terms of a remarkable way? 663 00:25:52,220 --> 00:25:53,430 Well, I talked a lot about, 664 00:25:53,430 --> 00:25:56,290 in the very beginning growth is a function of 665 00:25:56,290 --> 00:25:58,000 conversion and churn. 666 00:25:58,000 --> 00:26:01,140 This is Wufoo's growth curve for the first five years. 667 00:26:01,140 --> 00:26:02,270 Right? 668 00:26:02,270 --> 00:26:03,389 What's interesting is that we 669 00:26:03,390 --> 00:26:08,200 paid no money on advertising, on marketing. 670 00:26:08,200 --> 00:26:10,720 All of it was done by word of mouth growth. 671 00:26:10,720 --> 00:26:11,800 Right? 672 00:26:11,800 --> 00:26:13,480 And the interaction between, like, 673 00:26:13,480 --> 00:26:16,460 new users and downgrades are this. 674 00:26:18,570 --> 00:26:20,939 It's so slight what it takes, that gap, 675 00:26:20,940 --> 00:26:21,800 making that sort of work. 676 00:26:21,800 --> 00:26:24,710 And what a lot of people keep forgetting is that 677 00:26:24,710 --> 00:26:26,970 there's almost no difference between an increase in 678 00:26:26,970 --> 00:26:30,370 conversion rate, 1% increase and a 1% decrease churn. 679 00:26:30,370 --> 00:26:32,840 They do exactly the same thing to your growth. 680 00:26:32,840 --> 00:26:35,709 However, the ladder is actually much easier to do. 681 00:26:35,710 --> 00:26:38,380 It's much cheaper to do, in your apps, and 682 00:26:38,380 --> 00:26:39,940 a lot of the times we neglect this, 683 00:26:39,940 --> 00:26:42,100 neglect this to way far along. 684 00:26:42,100 --> 00:26:42,960 Right? And we usually have our 685 00:26:42,960 --> 00:26:46,900 B team works on these sort of projects and services. 686 00:26:46,900 --> 00:26:47,770 This is actually not the graph that we track most of 687 00:26:47,770 --> 00:26:49,250 the time at Wufoo, it's not even the one I'm proud of. 688 00:26:49,250 --> 00:26:51,855 This is the one I'm proud of. 689 00:26:51,855 --> 00:26:54,980 Cuz even though we have this sort of nice, awesome curve 690 00:26:54,980 --> 00:26:58,400 of growth, this is what, how loud is this scale? 691 00:26:58,400 --> 00:27:00,340 Keep the company small, have an awesome culture. 692 00:27:01,680 --> 00:27:02,920 And that required doing a lot of 693 00:27:02,920 --> 00:27:08,796 these things to help people sort of do what they need. 694 00:27:08,796 --> 00:27:13,709 So John Gotman noticed there was a different type of 695 00:27:13,709 --> 00:27:18,978 behavior for relationships and why people divorced. 696 00:27:18,978 --> 00:27:21,938 Basically there would be some subset of people who 697 00:27:21,938 --> 00:27:24,404 would stay together 10, 15 years and 698 00:27:24,404 --> 00:27:26,870 then all of a sudden they divorce and 699 00:27:26,870 --> 00:27:29,899 there was none of the other indicators which sort of 700 00:27:29,899 --> 00:27:32,320 show that this is what was gonna happen. 701 00:27:32,320 --> 00:27:34,670 And I was looking through the data and 702 00:27:34,670 --> 00:27:35,868 I realized oh, there's no passion, 703 00:27:35,868 --> 00:27:38,194 there's no fire between these people, right? 704 00:27:38,194 --> 00:27:39,678 When it comes to relationships they 705 00:27:39,679 --> 00:27:42,590 kinda follow the second law of thermodynamics, right? 706 00:27:42,590 --> 00:27:44,505 In a closed energy system things tend to run down so 707 00:27:44,506 --> 00:27:46,662 you have to constantly be putting energy and 708 00:27:46,662 --> 00:27:47,720 effort back into it. 709 00:27:47,720 --> 00:27:50,430 Now the way a lot of people sort of think about showing 710 00:27:50,430 --> 00:27:54,400 people that I care about you in products and in companies 711 00:27:54,400 --> 00:27:56,870 is to do things like let's have a blog, right? 712 00:27:56,870 --> 00:27:58,239 Lets' have a newsletter? 713 00:27:59,460 --> 00:28:01,870 The thing is, we look at these rates and basically it 714 00:28:01,870 --> 00:28:03,959 was such a small percentage of our active users that it 715 00:28:03,960 --> 00:28:06,290 was like, most of our users have no idea all 716 00:28:06,290 --> 00:28:08,659 the awesome stuff that we're doing for them. 717 00:28:08,660 --> 00:28:09,980 So we built a new tool. 718 00:28:09,980 --> 00:28:11,540 We called it the Wufoo system, and 719 00:28:11,540 --> 00:28:13,570 what it allowed us to do was just time stamp every new 720 00:28:13,570 --> 00:28:15,990 feature that we're building for users and 721 00:28:15,990 --> 00:28:17,540 then every time they would log in, we would look at 722 00:28:17,540 --> 00:28:20,830 the difference between their log in time or last log in 723 00:28:20,830 --> 00:28:22,580 time and the new features that were implemented. 724 00:28:22,580 --> 00:28:24,120 And when you had this message show up, 725 00:28:24,120 --> 00:28:25,300 hey since you've been gone 726 00:28:25,300 --> 00:28:27,800 here's all the awesome stuff that Wufoo did for you. 727 00:28:27,800 --> 00:28:29,840 Hands down this was the most talked about feature I've 728 00:28:29,840 --> 00:28:32,584 ever had every time I went out to talk to users. 729 00:28:32,585 --> 00:28:34,860 Right they'd say like dude I love that since you've be, 730 00:28:34,860 --> 00:28:37,490 since you've gone thing even though I pay the same amount 731 00:28:37,490 --> 00:28:39,510 every single month you guys are doing something for 732 00:28:39,510 --> 00:28:42,110 me almost every week and it's totally awesome and 733 00:28:42,110 --> 00:28:43,780 makes me feel, I'm getting maximum value. 734 00:28:47,200 --> 00:28:48,200 The other thing that we did in 735 00:28:48,200 --> 00:28:50,300 addition to having everyone support the people that 736 00:28:50,300 --> 00:28:52,500 paid their pay check is have them say thank you. 737 00:28:52,500 --> 00:28:56,790 And this was a large part due to us injecting sort of 738 00:28:56,790 --> 00:28:59,280 humility and modesty in to sort of the equation. 739 00:29:01,200 --> 00:29:03,270 Every single Friday we would get together and 740 00:29:03,270 --> 00:29:05,350 we'd just write simple hand written thank you cards to 741 00:29:05,350 --> 00:29:06,290 our users. 742 00:29:07,340 --> 00:29:10,606 And I know there's tons of people who would not be sort 743 00:29:10,606 --> 00:29:13,799 of excited about doing this, but it was a ritual that 744 00:29:13,799 --> 00:29:16,485 made sort of all the difference in terms of, 745 00:29:16,485 --> 00:29:19,822 like, having a team that was very tightly neat, 746 00:29:19,822 --> 00:29:21,226 tightly knit, also. 747 00:29:21,226 --> 00:29:23,190 And working on stuff that they really cared about. 748 00:29:23,190 --> 00:29:26,130 They always constantly knew what the mission was for, 749 00:29:26,130 --> 00:29:27,960 and why we sort of did what we did. 750 00:29:27,960 --> 00:29:30,100 These aren't fancy thank you cards, right? 751 00:29:30,100 --> 00:29:30,810 They're just simple, 752 00:29:30,810 --> 00:29:32,860 like handwritten stuff on an index card. 753 00:29:32,860 --> 00:29:33,990 We threw in a sticker, and 754 00:29:33,990 --> 00:29:37,140 slapped on a dinosaur on the front of it. 755 00:29:37,140 --> 00:29:41,720 And, what's interesting is we started this practice as 756 00:29:41,720 --> 00:29:45,830 a result of the early days of starting Wufoo. 757 00:29:45,830 --> 00:29:48,390 Chris, Ron, and I were talking, and we're trying to 758 00:29:48,390 --> 00:29:50,103 figure out, what are we gonna do to sort of show 759 00:29:50,103 --> 00:29:52,139 users that we appreciate them around Christmas. 760 00:29:52,139 --> 00:29:55,341 And he, Chris came up with this idea where he said 761 00:29:55,341 --> 00:29:58,824 hey guys, so a couple years ago my mom, like, made me 762 00:29:58,825 --> 00:30:01,831 write thank you notes to all of my relatives. 763 00:30:01,831 --> 00:30:03,554 For my Christmas gifts. 764 00:30:03,555 --> 00:30:05,440 And I didn't really like to do it but 765 00:30:05,440 --> 00:30:07,750 the following year all my presents were super good. 766 00:30:07,750 --> 00:30:12,170 So, I think we should try this for our business and 767 00:30:12,170 --> 00:30:13,770 see how it goes. 768 00:30:13,770 --> 00:30:17,320 So, that first year we wrote handwritten Christmas cards 769 00:30:17,320 --> 00:30:19,750 to all of our users that first year. 770 00:30:19,750 --> 00:30:21,360 Second year rolls around, and 771 00:30:21,360 --> 00:30:22,679 we have too many customers, like, and 772 00:30:22,680 --> 00:30:24,670 it's still just the three founders. 773 00:30:24,670 --> 00:30:26,920 And we were going like we're kinda screwed, 774 00:30:26,920 --> 00:30:28,650 I don't know what we're gonna do. 775 00:30:28,650 --> 00:30:30,940 And we read a book called the ultimate question, and 776 00:30:30,940 --> 00:30:32,540 in it he talks about hey, 777 00:30:32,540 --> 00:30:35,490 just focus on your most profitable users. 778 00:30:35,490 --> 00:30:36,510 And just tend them and 779 00:30:36,510 --> 00:30:38,330 take care of them, then it'll work out. 780 00:30:38,330 --> 00:30:38,909 So we're like, all right, 781 00:30:38,910 --> 00:30:41,000 that, that makes sense, that's scalable. 782 00:30:41,000 --> 00:30:41,770 So that year we 783 00:30:41,770 --> 00:30:43,459 only write to our highest paying customers. 784 00:30:44,480 --> 00:30:47,630 And the January rolls around that second year and 785 00:30:47,630 --> 00:30:50,540 one of our long time loyal users writes us and 786 00:30:50,540 --> 00:30:55,620 he's basically like, hey guys I, I really loved 787 00:30:55,620 --> 00:30:58,290 that Christmas card you sent me the first year and I just 788 00:30:58,290 --> 00:31:01,000 wanted you to know I haven't received my second card yet 789 00:31:01,000 --> 00:31:02,520 and I'm just looking forward to it. 790 00:31:02,520 --> 00:31:04,400 I know you didn't forget about me. 791 00:31:06,500 --> 00:31:07,399 Thanks a lot. 792 00:31:07,400 --> 00:31:11,930 So we're like, fuck, because the best way to sort of 793 00:31:11,930 --> 00:31:13,890 exceed expectations is not to send any to begin with, 794 00:31:13,890 --> 00:31:15,500 so we were like sort of in this conundrum. 795 00:31:15,500 --> 00:31:17,940 And what we decided, after thinking about it for 796 00:31:17,940 --> 00:31:20,270 a while, is that we need to stop doing it, 797 00:31:20,270 --> 00:31:22,400 you know, just one time a year. 798 00:31:22,400 --> 00:31:23,760 It needs to be something that's part of the culture. 799 00:31:23,760 --> 00:31:25,800 Happens every, every, every sort of week, even. 800 00:31:25,800 --> 00:31:27,950 And even though we'll never catch up to all of 801 00:31:27,950 --> 00:31:29,500 our customers. 802 00:31:29,500 --> 00:31:29,710 Just the practice of 803 00:31:29,710 --> 00:31:31,490 doing it will make all the difference. 804 00:31:33,300 --> 00:31:34,310 I talked a lot 805 00:31:34,310 --> 00:31:37,320 about a bunch of like lovey-dovey stuff and 806 00:31:37,320 --> 00:31:39,470 sort of like touchy-feely things that I think a lot of 807 00:31:39,470 --> 00:31:41,450 engineers don't like to think about too often. 808 00:31:41,450 --> 00:31:45,570 And so I'll end on some sort of hard business 809 00:31:45,570 --> 00:31:47,320 data or research. 810 00:31:47,320 --> 00:31:49,159 There's an article that was put out by 811 00:31:49,160 --> 00:31:51,220 the Harvard Business Review several years ago by 812 00:31:51,220 --> 00:31:53,800 Michael Treacy and Fred Wiersema and in it they 813 00:31:53,800 --> 00:31:55,560 talk about the discipline of market leaders. 814 00:31:55,560 --> 00:31:57,360 They say there's only three ways that you 815 00:31:57,360 --> 00:32:01,629 achieve market dominance and depending on how you want to 816 00:32:01,630 --> 00:32:03,300 achieve that market dominance you have 817 00:32:03,300 --> 00:32:05,870 to organize your company in a very specific way. 818 00:32:05,870 --> 00:32:09,350 Best price, best product and best overall solution. 819 00:32:09,350 --> 00:32:09,860 If you want to be 820 00:32:09,860 --> 00:32:12,270 the best price out there you focus on logistics. 821 00:32:12,270 --> 00:32:13,680 A Wal-Mart, an Amazon. 822 00:32:13,680 --> 00:32:15,390 If you want to be the best product out there you 823 00:32:15,390 --> 00:32:16,540 focus on R&D. 824 00:32:16,540 --> 00:32:17,460 Apple's usually 825 00:32:17,460 --> 00:32:19,490 a quintessential example of that. 826 00:32:19,490 --> 00:32:21,400 Best overall solution is about 827 00:32:21,400 --> 00:32:21,930 being customer intimate. 828 00:32:21,930 --> 00:32:23,410 And this is the path that you see followed by 829 00:32:23,410 --> 00:32:25,950 luxury brands and hospitality industry. 830 00:32:27,190 --> 00:32:30,400 What I love about this path towards market dominance is 831 00:32:30,400 --> 00:32:34,750 that the third one is the only one that 832 00:32:34,750 --> 00:32:37,620 everyone can do at any stage of their company. 833 00:32:37,620 --> 00:32:40,459 Requires almost no money to get started with. 834 00:32:40,460 --> 00:32:41,540 Usually just co. 835 00:32:41,540 --> 00:32:44,780 Requires a little bit of humility and some manners. 836 00:32:44,780 --> 00:32:48,590 And as a result you can achieve the success as any 837 00:32:48,590 --> 00:32:50,189 other people in sort of your market. 838 00:32:52,110 --> 00:32:53,679 That's all I got. 839 00:32:53,680 --> 00:32:54,340 Thank you very much. 840 00:33:00,370 --> 00:33:03,229 Yeah, let's take some questions if you guys 841 00:33:03,230 --> 00:33:03,912 have any. 842 00:33:03,912 --> 00:33:04,854 Right in the back there. 843 00:33:04,854 --> 00:33:07,339 >> Building products that users love? 844 00:33:07,339 --> 00:33:10,341 You might have multiple different types of users. 845 00:33:10,342 --> 00:33:14,150 How do you build one product that all users love? 846 00:33:14,150 --> 00:33:16,150 Maybe there is a feature that one really likes but 847 00:33:16,150 --> 00:33:18,110 detracts value from one that. 848 00:33:18,110 --> 00:33:19,149 >> All right. So what do you do 849 00:33:19,150 --> 00:33:20,950 when you have a product with lots of 850 00:33:20,950 --> 00:33:22,250 different type of users, right? 851 00:33:22,250 --> 00:33:23,490 Some users will love one thing and 852 00:33:23,490 --> 00:33:25,250 another will, will another. 853 00:33:25,250 --> 00:33:25,970 And I agree, there's 854 00:33:25,970 --> 00:33:27,290 a interesting fine line for that. 855 00:33:27,290 --> 00:33:29,370 What I always, usually tell people, 856 00:33:29,370 --> 00:33:31,700 is focus on the people who are the most passionate, 857 00:33:31,700 --> 00:33:32,360 especially in the early stages. 858 00:33:32,360 --> 00:33:34,200 Right? Whoever's, whatever niche 859 00:33:34,200 --> 00:33:37,320 it's gonna be, that's who I focus on completely. 860 00:33:37,320 --> 00:33:39,280 Things that a lot of different projects did. 861 00:33:39,280 --> 00:33:40,139 I think Ben Silverman of 862 00:33:40,140 --> 00:33:43,290 Pinterest started off with a designer bloggers, right? 863 00:33:43,290 --> 00:33:44,900 Curtail your thing for 864 00:33:44,900 --> 00:33:46,610 them and eventually you'll figure out sort of universal 865 00:33:46,610 --> 00:33:49,149 values that will appeal to a lot of other people. 866 00:33:49,150 --> 00:33:50,920 So, just start one at a time. 867 00:33:50,920 --> 00:33:51,720 And. 868 00:33:51,720 --> 00:33:53,900 The, a lot of the examples that you see up there, 869 00:33:53,900 --> 00:33:55,280 a lot of people make the mistake is like, oh, 870 00:33:55,280 --> 00:33:56,960 I'll just make my app funny. 871 00:33:56,960 --> 00:34:00,130 But, humor is like really difficult to do. 872 00:34:00,130 --> 00:34:01,200 Right? What you wanna shoot for 873 00:34:01,200 --> 00:34:02,350 is something sort of witty. 874 00:34:02,350 --> 00:34:03,719 And, quite honestly, 875 00:34:03,720 --> 00:34:05,730 you have to get functionality right. 876 00:34:05,730 --> 00:34:07,220 So like the Japanese quality. 877 00:34:07,220 --> 00:34:10,210 If you don't have a on there, right, don't try to 878 00:34:10,210 --> 00:34:13,100 do anything witty, right, cuz it will backfire on you. 879 00:34:13,100 --> 00:34:15,806 So hands down, our number one focus is make it as 880 00:34:15,806 --> 00:34:18,107 easy to use as possible for and 881 00:34:18,108 --> 00:34:20,159 anything else on top was polish. 882 00:34:20,159 --> 00:34:20,596 Right here. 883 00:34:20,596 --> 00:34:28,310 >> So so everybody says that to focus on your product. 884 00:34:28,310 --> 00:34:29,290 I'm also good at that. 885 00:34:29,300 --> 00:34:32,300 I love to do a project and I love to make it the best. 886 00:34:32,300 --> 00:34:35,100 But we are to that certain point that we are focused on 887 00:34:35,100 --> 00:34:37,509 our product but we don't get like constantly right? 888 00:34:37,510 --> 00:34:38,270 Sorry. 889 00:34:38,270 --> 00:34:41,639 So second thing so how much we should focus on product? 890 00:34:41,639 --> 00:34:43,639 But because we should do now marketing. 891 00:34:43,639 --> 00:34:45,489 We should get somebody customers and 892 00:34:45,489 --> 00:34:47,978 like and like start talking to customers but 893 00:34:47,980 --> 00:34:50,179 when you are too focused on your product. 894 00:34:50,179 --> 00:34:52,940 Like users online have them. 895 00:34:52,940 --> 00:34:55,460 Right, so what exactly do you guys mean when you 896 00:34:55,460 --> 00:34:57,856 are saying like focus only on your product and 897 00:34:57,856 --> 00:34:59,130 give the best product? 898 00:34:59,130 --> 00:35:01,548 >> Okay. 899 00:35:01,548 --> 00:35:04,955 So the question sort of is how do we balance this 900 00:35:04,955 --> 00:35:07,889 sort of thing where we wanna be obsessed with 901 00:35:07,889 --> 00:35:09,800 working on product. 902 00:35:09,800 --> 00:35:11,630 Yet. But all the other skills, 903 00:35:11,630 --> 00:35:14,800 and sort of tasks that are needed by a company, 904 00:35:14,800 --> 00:35:15,509 like marketing and branding and 905 00:35:15,510 --> 00:35:17,730 all that stuff, and how, how we sort of balance that. 906 00:35:17,730 --> 00:35:20,100 And the thing is, like, it starts off as you juggling, 907 00:35:20,100 --> 00:35:22,540 like, tons of things constantly in the air. 908 00:35:22,540 --> 00:35:24,440 The thing is, if you're working on products, like, 909 00:35:24,440 --> 00:35:26,500 you should also always have this flip side as 910 00:35:26,500 --> 00:35:28,500 when you're talking to users. 911 00:35:28,500 --> 00:35:28,560 Right? And for 912 00:35:28,560 --> 00:35:31,900 us inside of Wufoo, the way we got people to talk to 913 00:35:31,100 --> 00:35:32,440 users is they just did customer support. 914 00:35:32,440 --> 00:35:34,680 And they got to see firsthand, right away. 915 00:35:34,680 --> 00:35:35,930 Whether that feature sucked or not, 916 00:35:35,930 --> 00:35:37,500 and also impacted everyone else in the company, 917 00:35:37,500 --> 00:35:39,420 because everyone had a customer support shift. 918 00:35:39,420 --> 00:35:42,170 So you have this sort of social incentive to sort of 919 00:35:42,170 --> 00:35:43,360 make everything work. 920 00:35:43,360 --> 00:35:46,150 And so, like I said, there should be no point where 921 00:35:46,150 --> 00:35:47,230 you're only focused on product. 922 00:35:47,230 --> 00:35:49,340 You should always have time where you work on product, 923 00:35:49,340 --> 00:35:51,420 and then you see sort of what users say, say to you. 924 00:35:51,420 --> 00:35:53,270 And you should always have this virtual, like, 925 00:35:53,270 --> 00:35:54,420 feedback loop on there. 926 00:35:54,420 --> 00:35:56,630 So be careful when you don't have that. 927 00:35:56,630 --> 00:35:58,340 Usually what ends up happening, if you're lucky. 928 00:35:59,380 --> 00:36:00,700 In terms of marketing and 929 00:36:00,700 --> 00:36:03,700 sales, like, usually my feeling is like, you have to 930 00:36:03,700 --> 00:36:05,900 spend money on marketing and advertising, all this stuff. 931 00:36:05,900 --> 00:36:06,680 It's usually a tax you pay because you 932 00:36:06,680 --> 00:36:08,279 haven't made your product remarkable. 933 00:36:08,280 --> 00:36:09,510 Right? 934 00:36:09,510 --> 00:36:11,630 Word of mouth growth is the easiest kind of growth, and 935 00:36:11,630 --> 00:36:14,260 it's how a lot of the great companies sort of grow. 936 00:36:14,260 --> 00:36:16,220 So figure out how to wait, how to like, 937 00:36:16,220 --> 00:36:18,120 have a story that people want to tell. 938 00:36:18,120 --> 00:36:18,830 About your product. 939 00:36:18,830 --> 00:36:20,400 Where they're the most interesting person at 940 00:36:20,400 --> 00:36:21,660 the dinner table, right? 941 00:36:21,660 --> 00:36:23,819 And then that person is your sales person, right? 942 00:36:23,820 --> 00:36:25,462 That person is your sales force for you. 943 00:36:25,462 --> 00:36:28,106 Right here. 944 00:36:28,106 --> 00:36:32,112 >> like, where do you find crystal clear customer or 945 00:36:32,112 --> 00:36:36,140 user need and the demand is there is the right solution. 946 00:36:36,140 --> 00:36:38,100 How do you communicate with engineering and 947 00:36:38,100 --> 00:36:41,710 designing team to make sure that because sometimes 948 00:36:41,710 --> 00:36:44,470 people in the team come up with ideas, and but 949 00:36:44,470 --> 00:36:45,620 still at the end of the day, 950 00:36:45,620 --> 00:36:48,339 how can you make a decision with where to go? 951 00:36:48,340 --> 00:36:50,300 >> Oh, so how do you make a decision on product? 952 00:36:51,580 --> 00:36:53,000 And communicate that with your sort of 953 00:36:53,000 --> 00:36:54,350 engineering team when there's like lots of 954 00:36:54,350 --> 00:36:55,410 different directions to go? 955 00:36:57,290 --> 00:36:58,950 My feeling is that. 956 00:37:01,150 --> 00:37:03,180 So for us we just looked at support. 957 00:37:03,180 --> 00:37:04,180 It was really easy cuz 958 00:37:04,180 --> 00:37:06,480 you often just saw what are things that 959 00:37:06,480 --> 00:37:08,410 people are having the most amount of problems with? 960 00:37:08,410 --> 00:37:09,980 Or people asking all the time. 961 00:37:09,980 --> 00:37:12,360 You cannot help but get feature requests from 962 00:37:12,360 --> 00:37:14,280 people no matter like whatever opening or 963 00:37:14,280 --> 00:37:15,940 orifice you have in your product or app. 964 00:37:15,940 --> 00:37:18,430 Like, people will like jam feature requests in there. 965 00:37:18,430 --> 00:37:19,210 So you're easily going to 966 00:37:19,210 --> 00:37:20,450 know sort of what they sort of what. 967 00:37:20,450 --> 00:37:22,270 Your job as a product person and 968 00:37:22,270 --> 00:37:25,500 engineer is to not just do what they say, because that 969 00:37:25,500 --> 00:37:27,680 way, you'll just be a slave, is to figure out, 970 00:37:27,680 --> 00:37:30,930 sort of deeply, what are the reasons why underlying those 971 00:37:30,930 --> 00:37:33,410 things and sort of solve that deep underlying reason. 972 00:37:33,410 --> 00:37:34,606 The thing is that everyone wants to 973 00:37:34,606 --> 00:37:35,990 have a different way of. 974 00:37:35,990 --> 00:37:38,390 To sort of go, then ultimately it comes down to, 975 00:37:38,390 --> 00:37:39,850 like, someone's gonna figure something out. 976 00:37:39,850 --> 00:37:43,180 But I also make the smallest version of each little idea. 977 00:37:43,180 --> 00:37:46,800 No longer than a week or two weeks to build it out there. 978 00:37:46,800 --> 00:37:46,980 And you can try them out and 979 00:37:46,980 --> 00:37:48,890 see sort of what works and don't work. 980 00:37:48,890 --> 00:37:50,600 I think it's dangerous to 981 00:37:50,600 --> 00:37:51,799 have multiple different product directions that 982 00:37:51,800 --> 00:37:53,879 requires lots of time to sort of figure out. 983 00:37:55,430 --> 00:37:56,490 Sam. 984 00:37:56,500 --> 00:37:57,760 >> Related to that can you tell the story of 985 00:37:57,760 --> 00:38:00,910 how the king for a day thing >> Yeah. 986 00:38:00,910 --> 00:38:01,770 Okay. So 987 00:38:03,920 --> 00:38:06,760 so I don't like hackathons. 988 00:38:06,760 --> 00:38:09,660 I think they sort of suck in terms of ones done inside 989 00:38:09,660 --> 00:38:10,210 of companies. 990 00:38:10,210 --> 00:38:11,270 Because. 991 00:38:11,270 --> 00:38:13,937 You spend like 48 hours working on something really 992 00:38:13,937 --> 00:38:16,125 hard that you're sort of passionate about and 993 00:38:16,125 --> 00:38:20,280 99% of them never make it to production right, and 994 00:38:20,280 --> 00:38:22,410 it's sort, sort of real like super sad. 995 00:38:22,410 --> 00:38:24,190 So for us we like flipped it on it's head and 996 00:38:24,190 --> 00:38:25,720 we came up with an idea that we called king for 997 00:38:25,720 --> 00:38:28,600 a day and it actually worked over the weekend. 998 00:38:28,600 --> 00:38:31,920 But how it worked is someone randomly in the company got 999 00:38:31,920 --> 00:38:34,660 drawn and they got to be the king, and the king got to 1000 00:38:34,660 --> 00:38:36,399 tell everyone else what to do on the product. 1001 00:38:36,400 --> 00:38:39,850 So everything that was bothering them about Wufoo. 1002 00:38:39,850 --> 00:38:40,890 About the customer support stuff, or 1003 00:38:40,890 --> 00:38:42,529 some feature they really want to have built. 1004 00:38:42,530 --> 00:38:44,440 They've got the engineering resources, the market 1005 00:38:44,440 --> 00:38:46,650 resources, the advertising resources of everyone inside 1006 00:38:46,650 --> 00:38:48,670 of the company, to make it sort of happen. 1007 00:38:48,670 --> 00:38:49,770 And of course, we'd work with them to 1008 00:38:49,770 --> 00:38:53,200 figure out like what can be actually done in 48 hours. 1009 00:38:53,200 --> 00:38:56,730 But we would do this one to two times a year. 1010 00:38:56,730 --> 00:38:57,820 And it was like a huge hit and 1011 00:38:57,820 --> 00:39:00,310 it was a boost to morale, cuz what people most love. 1012 00:39:00,310 --> 00:39:02,810 It is like working on things where it's like, oh, 1013 00:39:02,810 --> 00:39:04,200 I made a difference to the app. 1014 00:39:05,300 --> 00:39:05,700 Right? 1015 00:39:05,700 --> 00:39:09,290 And so, for us, that's one way that we would like 1016 00:39:09,300 --> 00:39:11,370 sort of divide time for like product direction. 1017 00:39:11,370 --> 00:39:13,900 It's like some times the people that work for 1018 00:39:13,100 --> 00:39:14,730 you, they have a strong opinion about where it, 1019 00:39:14,730 --> 00:39:15,400 where it should go. 1020 00:39:15,400 --> 00:39:17,580 And it's a good way to sort of democratize it 1021 00:39:17,580 --> 00:39:18,939 a little bit, by rotating it around. 1022 00:39:20,760 --> 00:39:21,520 Yes. 1023 00:39:21,520 --> 00:39:23,759 >> You said you guys all work from home. 1024 00:39:23,760 --> 00:39:25,450 Usually seems like a nightmare. 1025 00:39:25,450 --> 00:39:27,509 In that office, how do you make that work? 1026 00:39:27,510 --> 00:39:28,510 >> Okay, so we all work from home. 1027 00:39:28,510 --> 00:39:29,540 So, I will tell you this. 1028 00:39:29,540 --> 00:39:31,610 We all still work within the Tampa Bay area. 1029 00:39:31,610 --> 00:39:32,760 We would allow anybody to 1030 00:39:32,760 --> 00:39:34,510 work from any where but usually. 1031 00:39:34,510 --> 00:39:35,904 As we tried to recruit them,they sort of 1032 00:39:35,904 --> 00:39:37,820 meet our team, and they just decide, 1033 00:39:37,820 --> 00:39:39,500 okay we just want to come and move here anyway. 1034 00:39:41,970 --> 00:39:46,562 Remote working, is especially tricky, 1035 00:39:46,563 --> 00:39:48,350 a lot of people like to romanticize it, 1036 00:39:48,350 --> 00:39:50,430 especially people, who are like employees. 1037 00:39:50,430 --> 00:39:54,109 But the thing is, An office gives you a lot of, sort of, 1038 00:39:54,110 --> 00:39:56,870 benefits, right, and efficiencies that you now 1039 00:39:56,870 --> 00:39:58,880 have to compensate for when you remote working. 1040 00:39:58,880 --> 00:40:00,770 But remote working also has these other sort of 1041 00:40:00,770 --> 00:40:02,590 efficiencies in place, for example, 1042 00:40:02,590 --> 00:40:04,870 I don't have to worry about my employees losing two 1043 00:40:04,870 --> 00:40:07,569 hours of their day to commuting, for instance. 1044 00:40:07,570 --> 00:40:08,980 So the biggest thing that we had to do for 1045 00:40:08,980 --> 00:40:11,860 remote working is to respect people's time. 1046 00:40:11,860 --> 00:40:14,500 And so the way we had it set up is we actually had 1047 00:40:14,500 --> 00:40:16,470 a four and a half day work week at Wufoo. 1048 00:40:16,470 --> 00:40:18,839 Half day on Friday was for all the meetings and stuff. 1049 00:40:18,840 --> 00:40:20,830 We said like no business deal, 1050 00:40:20,830 --> 00:40:22,990 meetings, no talking with other outside parties. 1051 00:40:22,990 --> 00:40:25,569 They all have to be done on Friday, on that half day. 1052 00:40:25,570 --> 00:40:27,320 It couldn't be done in the middle of the week. 1053 00:40:27,320 --> 00:40:29,420 And then also one day of everyone was 1054 00:40:29,420 --> 00:40:30,780 already dedicated customer support. 1055 00:40:30,780 --> 00:40:32,910 So everyone in our company effectively only had 1056 00:40:32,910 --> 00:40:35,170 three days each week to actually build or 1057 00:40:35,170 --> 00:40:36,570 work on whatever they were doing. 1058 00:40:36,570 --> 00:40:38,370 But I actually firmly believe that if you 1059 00:40:38,370 --> 00:40:41,270 have three solid days, right, eight to ten hours, 1060 00:40:41,270 --> 00:40:43,230 where you are only working on what you need to build, 1061 00:40:43,230 --> 00:40:45,350 you can get a ton of shit done. 1062 00:40:45,350 --> 00:40:47,600 And so what we said was, 1063 00:40:47,600 --> 00:40:48,770 you have to respect everyone's time during that 1064 00:40:48,770 --> 00:40:51,130 three day period, if they're in that three day period. 1065 00:40:51,130 --> 00:40:53,270 And what we came up with is a 15 minute rule and the way 1066 00:40:53,270 --> 00:40:56,580 it worked is you can have a discussion like a chat or 1067 00:40:56,580 --> 00:40:59,380 a phone call or whatever with someone but it could go 1068 00:40:59,380 --> 00:41:01,640 no longer than 15 minutes so if you have some complicated 1069 00:41:01,640 --> 00:41:04,490 issue that you couldn't figure out we'd say, 1070 00:41:04,500 --> 00:41:07,320 at 15 minutes you are to immediately table that item. 1071 00:41:07,320 --> 00:41:10,500 Right and have us discuss it on Friday and 1072 00:41:10,500 --> 00:41:12,610 you'd move on to the next item on your list right, 1073 00:41:12,610 --> 00:41:13,690 the enhanced productivity. 1074 00:41:15,500 --> 00:41:18,430 More often than not I, I would say 90% of the time 1075 00:41:18,430 --> 00:41:20,450 the item never got brought up on Friday because usually 1076 00:41:20,450 --> 00:41:22,810 what would happen is people would sleep on it and then 1077 00:41:22,810 --> 00:41:25,730 they'd just magically say like, hey I found a solution 1078 00:41:25,730 --> 00:41:27,740 or like hey that's not a big problem whatsoever. 1079 00:41:29,300 --> 00:41:30,760 Because most problems inside of companies, 1080 00:41:30,760 --> 00:41:34,290 they don't need to be solved in real time, or right away. 1081 00:41:34,290 --> 00:41:36,990 The only things are like when the site is down, or 1082 00:41:36,990 --> 00:41:38,310 payments aren't working. 1083 00:41:38,310 --> 00:41:39,490 Right? Everything outside of 1084 00:41:39,490 --> 00:41:43,479 that is basically kind of luxury, so 1085 00:41:43,480 --> 00:41:45,520 focus on your primaries as much as possible. 1086 00:41:45,520 --> 00:41:46,419 And, as a result, 1087 00:41:46,419 --> 00:41:48,444 our ten person team did far more than many, 1088 00:41:48,445 --> 00:41:51,150 many other companies, as a result. 1089 00:41:51,150 --> 00:41:55,400 But, it takes extra work to make remote working happen. 1090 00:41:55,400 --> 00:41:57,700 We are an extremely disciplined sort of team. 1091 00:41:57,700 --> 00:41:58,370 And I would have to say 1092 00:41:59,790 --> 00:42:02,850 there's almost not many YC companies that actually have 1093 00:42:02,850 --> 00:42:04,299 been able to replicate sort of what we do. 1094 00:42:04,300 --> 00:42:06,250 I think there's only two other companies in YC that 1095 00:42:06,250 --> 00:42:09,600 sort of have the same sort of discipline working style. 1096 00:42:09,600 --> 00:42:11,560 It takes more work in a very different fashion. 1097 00:42:11,560 --> 00:42:12,100 Right? 1098 00:42:12,100 --> 00:42:14,779 An office allows you to be a little bit lazier, right, 1099 00:42:14,780 --> 00:42:16,620 in terms of all these things around productivity. 1100 00:42:20,880 --> 00:42:21,880 Okay. Over here. 1101 00:42:21,880 --> 00:42:23,330 >> Right. 1102 00:42:23,330 --> 00:42:26,440 Just to go off that question as the leader of the team 1103 00:42:26,440 --> 00:42:29,891 how did you manage to instill a company culture 1104 00:42:29,891 --> 00:42:30,948 but and also the count, 1105 00:42:30,948 --> 00:42:34,180 accountability the employees especially cuz. 1106 00:42:34,180 --> 00:42:35,700 Working space. 1107 00:42:35,700 --> 00:42:38,700 >> Okay, how do we, how do we set up accountability for 1108 00:42:38,700 --> 00:42:39,390 employees? 1109 00:42:39,390 --> 00:42:40,230 As, as a manager. 1110 00:42:40,230 --> 00:42:43,530 Alright, so, at Wufoo we were profitable nine months 1111 00:42:43,530 --> 00:42:46,920 after launch, so, we had profit sharing, right? 1112 00:42:46,920 --> 00:42:49,510 And so, it makes pretty simple and clear. 1113 00:42:49,510 --> 00:42:53,110 It, it would be a multiple of whatever bonus pull that 1114 00:42:53,110 --> 00:42:54,420 we sort of had, and 1115 00:42:54,420 --> 00:42:56,720 the performance measures would be based on sort of, 1116 00:42:56,720 --> 00:42:58,350 how they did in customer support. 1117 00:42:58,350 --> 00:42:59,880 All right, on their duties there, and sort of 1118 00:42:59,880 --> 00:43:02,810 what they said they were wanting to accomplish or do. 1119 00:43:02,810 --> 00:43:04,790 I don't like process, and I don't like lots of 1120 00:43:04,790 --> 00:43:07,110 tools to help get people to be productive. 1121 00:43:07,110 --> 00:43:08,547 So the only thing that we had for 1122 00:43:08,547 --> 00:43:10,400 helping people manage, like, 1123 00:43:10,400 --> 00:43:12,270 sort of their projects, is to-do lists. 1124 00:43:12,270 --> 00:43:13,810 And that is, like, simple text files that we 1125 00:43:13,810 --> 00:43:15,490 shared in our Dropbox account. 1126 00:43:15,490 --> 00:43:17,379 Each person had their name on it, and you got to 1127 00:43:17,380 --> 00:43:19,710 see every time somebody updated their to-do list. 1128 00:43:19,710 --> 00:43:21,600 What we said is every single night just 1129 00:43:21,600 --> 00:43:22,650 set everything that you did that day. 1130 00:43:22,650 --> 00:43:23,560 Right? And then on Friday, 1131 00:43:23,560 --> 00:43:24,690 we would just go over. 1132 00:43:24,690 --> 00:43:25,770 Okay, this is what you 1133 00:43:25,770 --> 00:43:26,970 said last week that you're gonna do. 1134 00:43:26,970 --> 00:43:28,240 This is what you actually got done, or 1135 00:43:28,240 --> 00:43:29,350 the sort of the problems at hand. 1136 00:43:29,350 --> 00:43:31,370 And it's super simple, right? 1137 00:43:31,370 --> 00:43:33,650 It creates this like nice written trail for 1138 00:43:33,650 --> 00:43:35,550 how to sort of handle stuff, right? 1139 00:43:35,550 --> 00:43:37,720 And I don't have to worry about managing them, right? 1140 00:43:37,720 --> 00:43:39,299 They sorta set the tone for 1141 00:43:39,300 --> 00:43:41,610 how they want to be sort of assessed. 1142 00:43:41,610 --> 00:43:42,550 And it makes it really simple. 1143 00:43:42,550 --> 00:43:45,370 And for people who are excellent at what they do. 1144 00:43:45,370 --> 00:43:47,490 Right? It works very, very well. 1145 00:43:47,500 --> 00:43:48,900 And then when you actually have problems, 1146 00:43:48,900 --> 00:43:50,570 it's very easy to fire people. 1147 00:43:50,570 --> 00:43:51,760 I was fortunate that I 1148 00:43:51,760 --> 00:43:54,110 never had to fire anyone at Wufoo, right? 1149 00:43:54,110 --> 00:43:55,490 But we were able to correct a lot of 1150 00:43:55,490 --> 00:43:57,529 people's behavior very, very quickly. 1151 00:43:57,530 --> 00:43:58,750 Cuz we just kinda look at this, and 1152 00:43:58,750 --> 00:44:00,160 it's like look, this is your pattern of behavior. 1153 00:44:00,160 --> 00:44:02,509 You finish a fraction of the items on your list. 1154 00:44:02,510 --> 00:44:03,900 You do most of 1155 00:44:03,900 --> 00:44:05,810 the items at the last second right before Friday. 1156 00:44:05,810 --> 00:44:07,240 That's a problem, you've gotta manage your time 1157 00:44:07,240 --> 00:44:10,259 better and this is evidence that you've provided to us. 1158 00:44:10,260 --> 00:44:12,420 All we have to do is sort of describe it back to you and 1159 00:44:12,420 --> 00:44:14,650 because everyone in the company sort of sees it, 1160 00:44:14,650 --> 00:44:16,000 right, there's social pressure that's put 1161 00:44:16,000 --> 00:44:17,870 into place that helps make it all sort of happen. 1162 00:44:19,510 --> 00:44:20,810 Right here. 1163 00:44:20,810 --> 00:44:22,930 >> How did you hire people that, you know, 1164 00:44:22,930 --> 00:44:25,629 you felt would be able to work with in this kind of 1165 00:44:25,630 --> 00:44:28,216 environment that's, that's >> So, 1166 00:44:28,216 --> 00:44:31,720 how do you hire people that can work, remotely? 1167 00:44:31,720 --> 00:44:33,990 And then sort of work in this sort of fashion. 1168 00:44:33,990 --> 00:44:35,879 So, pretty easily, 1169 00:44:35,880 --> 00:44:38,170 you have them work on a side project for you. 1170 00:44:38,170 --> 00:44:39,330 So you contract them out, and 1171 00:44:39,330 --> 00:44:40,920 they have to work remotely. 1172 00:44:40,920 --> 00:44:42,610 As such, usually, the projects I like to 1173 00:44:42,610 --> 00:44:45,160 have them work on is about a month long, right? 1174 00:44:45,160 --> 00:44:46,839 I could do things much faster for a week. 1175 00:44:46,840 --> 00:44:48,100 But usually get a good sense of, 1176 00:44:48,100 --> 00:44:49,870 like, how well people sort of manage themselves, and 1177 00:44:49,870 --> 00:44:51,930 work on things from a project like that. 1178 00:44:51,930 --> 00:44:53,750 So that was always the first assessment. 1179 00:44:53,750 --> 00:44:55,560 Like we never did anything just by interviews. 1180 00:44:56,840 --> 00:44:59,740 The other thing we had to, sort of, screen them for 1181 00:44:59,740 --> 00:45:01,350 is their ability to do customer support. 1182 00:45:01,350 --> 00:45:02,190 Because not any, 1183 00:45:02,190 --> 00:45:05,980 every engineer sort of has the empathy skills to 1184 00:45:05,980 --> 00:45:08,130 handle that stress. 1185 00:45:08,130 --> 00:45:12,100 So sometimes I would have people write breakup letters 1186 00:45:12,100 --> 00:45:14,440 to me, right, in an interview, just like hey, 1187 00:45:14,440 --> 00:45:17,310 pretend you have to break up with me you have 15 minutes 1188 00:45:17,310 --> 00:45:19,320 to write in on there and you can only write it by hand, 1189 00:45:19,320 --> 00:45:21,600 what are you gonna say? 1190 00:45:21,600 --> 00:45:23,740 And so you get a good sense of sort of their writing 1191 00:45:23,740 --> 00:45:25,669 skills, because like 90% of what you do in customer 1192 00:45:25,670 --> 00:45:27,940 support is tell them bad news, like, oh, we don't 1193 00:45:27,940 --> 00:45:30,710 support that feature, sorry that's not gonna work, or. 1194 00:45:30,710 --> 00:45:32,700 It's not gonna be available. 1195 00:45:32,700 --> 00:45:33,985 And so people have to have sort of tacked at that. 1196 00:45:33,985 --> 00:45:36,300 >> How 'bout one more question? 1197 00:45:36,300 --> 00:45:38,650 >> One question, right here the glasses. 1198 00:45:38,650 --> 00:45:40,430 >> So, since Wiki has all these like tricks and 1199 00:45:40,430 --> 00:45:42,529 experiments that have really helped the company. 1200 00:45:42,530 --> 00:45:44,950 Do you INAUDIBLE] of ones that didn't work out? 1201 00:45:44,950 --> 00:45:46,120 >> Have all these tricks and 1202 00:45:46,120 --> 00:45:47,240 experiments to help the company, 1203 00:45:47,240 --> 00:45:49,359 are there any ones that didn't work out? 1204 00:45:50,470 --> 00:45:51,520 All right, I'll talk about one. 1205 00:45:51,520 --> 00:45:54,770 So one of the things that we did early on to try to 1206 00:45:54,770 --> 00:45:56,940 motivate ourselves was try to get, 1207 00:45:58,530 --> 00:46:01,300 like we understood this idea of like crunch mode and 1208 00:46:01,300 --> 00:46:02,820 that it's really bad for people. 1209 00:46:02,820 --> 00:46:04,370 Like if you're doing the subscription business you 1210 00:46:04,370 --> 00:46:06,230 need people to last for the long term, and 1211 00:46:06,230 --> 00:46:08,400 video games a lot of times they like crunch people. 1212 00:46:08,400 --> 00:46:09,660 All, all the time. 1213 00:46:09,660 --> 00:46:10,859 For like a specific deadline or 1214 00:46:10,860 --> 00:46:12,610 have multiple sprints every two weeks and 1215 00:46:12,610 --> 00:46:13,930 you have to shoot up to this deadline and 1216 00:46:13,930 --> 00:46:15,290 it's like exhausting. 1217 00:46:15,290 --> 00:46:17,900 And so, because, when it's happening it's like you 1218 00:46:17,900 --> 00:46:18,560 might get an increase in productivity but 1219 00:46:18,560 --> 00:46:20,299 the recovery period that you need for people is 1220 00:46:20,300 --> 00:46:23,770 always greater, right, than the productivity you gain. 1221 00:46:23,770 --> 00:46:24,630 And at a company where you 1222 00:46:24,630 --> 00:46:25,810 need everyone doing customer support and 1223 00:46:25,810 --> 00:46:27,420 being on their game and constantly put, pushed out 1224 00:46:27,420 --> 00:46:29,910 features you don't have time for recovery. 1225 00:46:29,910 --> 00:46:33,490 So, we were thinking about, okay, we want to build like, 1226 00:46:33,490 --> 00:46:36,740 a company vacation into how Wufoo sort of works to 1227 00:46:36,740 --> 00:46:38,299 reward our users every single year. 1228 00:46:38,300 --> 00:46:38,830 And we said okay, 1229 00:46:38,830 --> 00:46:40,330 if the vacation is sort of built in there for 1230 00:46:40,330 --> 00:46:43,290 the recovery, we could have one crunch period, right, 1231 00:46:43,300 --> 00:46:44,300 before that vacation set up. 1232 00:46:44,300 --> 00:46:46,470 And we'll just only do customer support that will 1233 00:46:46,470 --> 00:46:48,430 just sort of scale with people. 1234 00:46:48,430 --> 00:46:49,259 So. 1235 00:46:49,260 --> 00:46:55,700 The way we did the very first crunch mode 1236 00:46:55,700 --> 00:46:57,380 is that it was just between the three founders. 1237 00:46:57,380 --> 00:47:00,900 And we had each of us draw a ten-item to-do list. 1238 00:47:00,900 --> 00:47:02,140 That would be fairly aggressive. 1239 00:47:02,140 --> 00:47:02,890 And the first person to 1240 00:47:02,890 --> 00:47:05,640 get through seven of their items would win. 1241 00:47:05,640 --> 00:47:07,290 And the the last person to 1242 00:47:07,300 --> 00:47:08,630 get through seven of their items. 1243 00:47:08,630 --> 00:47:10,950 Would become what we called trippage. 1244 00:47:10,950 --> 00:47:13,180 And trippage meant that you carried other 1245 00:47:13,180 --> 00:47:15,629 person's luggage and got people drinks when you're on 1246 00:47:15,630 --> 00:47:16,500 the company vacation. 1247 00:47:17,560 --> 00:47:21,540 So we did that and during that period, 1248 00:47:21,540 --> 00:47:23,140 everyone was like pretty excited about it and 1249 00:47:23,140 --> 00:47:24,850 motivated, and only the winner got to 1250 00:47:24,850 --> 00:47:27,640 choose the next company vacation the following year. 1251 00:47:27,640 --> 00:47:31,240 And then all the sudden Ryan had basically poorly 1252 00:47:31,240 --> 00:47:33,490 estimated the items on his list. 1253 00:47:33,500 --> 00:47:36,400 And he realized very quickly I'm going to fucking lose. 1254 00:47:36,400 --> 00:47:38,950 And so he was just like I give up. 1255 00:47:38,950 --> 00:47:40,250 And he just sort of stopped. 1256 00:47:40,250 --> 00:47:42,410 So crunch mode turned out to be blah mode for 1257 00:47:42,410 --> 00:47:44,339 him because he knew he was gonna lose and 1258 00:47:44,340 --> 00:47:46,300 became pretty demoralized. 1259 00:47:46,300 --> 00:47:47,380 So as a result of doing that, we 1260 00:47:47,380 --> 00:47:50,130 decided not to do it in that similar fashion anymore. 1261 00:47:50,130 --> 00:47:52,610 So, good idea that we like to talk about. 1262 00:47:52,610 --> 00:47:53,980 But is one that we, we never did again. 1263 00:47:55,120 --> 00:47:55,750 All right, guys. 1264 00:47:55,750 --> 00:47:56,930 Thanks a lot. 1265 00:47:56,930 --> 00:47:58,928 You can email me at kevin@