1
00:00:00,080 --> 00:00:03,000
Everybody, welcome to your 
episode of Newcomer. 

2
00:00:03,000 --> 00:00:06,240
Tom Doton here, joined by 
Madeline Renbarger and Eric 

3
00:00:06,240 --> 00:00:09,680
Newcomer of Newcomer. 
We're all in San Francisco right

4
00:00:09,680 --> 00:00:12,640
now, but we are not all in the 
same place, actually. 

5
00:00:12,800 --> 00:00:16,280
Madeline and I are holed up at 
our friend Volley's offices, 

6
00:00:16,280 --> 00:00:20,680
which is as close to newcomer HQ
in San Francisco as you can get.

7
00:00:20,720 --> 00:00:22,920
I would say this is the Newcomer
Satellite office. 

8
00:00:23,080 --> 00:00:24,680
Absolutely. 
You gotta come in. 

9
00:00:24,840 --> 00:00:27,160
We can't all be in the same 
place for security reasons. 

10
00:00:27,160 --> 00:00:29,760
We don't want all newcomer 
employees together. 

11
00:00:29,840 --> 00:00:31,480
It was dangerous enough this 
week. 

12
00:00:31,760 --> 00:00:34,440
Yeah, well, right. 
Good transition to the reason 

13
00:00:34,440 --> 00:00:37,880
we're here, which is that we are
fresh off the Breaking the Bank 

14
00:00:37,880 --> 00:00:40,680
Summit, the Fintech Summit 
annual thing. 

15
00:00:40,680 --> 00:00:42,280
Although what is it the second 
year you've done it? 

16
00:00:42,320 --> 00:00:45,080
Well, the first year we called 
it the Newcomer Banking Summit, 

17
00:00:45,080 --> 00:00:48,320
and we realized fintech was much
more exciting than banking. 

18
00:00:48,400 --> 00:00:51,280
So in some ways this was the 
first year and in some ways it 

19
00:00:51,280 --> 00:00:53,520
was the second. 
You know, that was a little 

20
00:00:53,520 --> 00:00:55,560
awkward on stage. 
It's like people are coming 

21
00:00:55,560 --> 00:00:57,040
back, but this is the first 
time. 

22
00:00:57,240 --> 00:01:00,840
But yeah, this was, yeah, our 
first pure Fintech summit with 

23
00:01:00,840 --> 00:01:03,920
some returning faces. 
Right, there were some returning

24
00:01:04,080 --> 00:01:05,880
returning faces, like Jackie 
Reesus. 

25
00:01:05,880 --> 00:01:08,560
Yeah, Matt Harris from Bain 
Capital Ventures came back for 

26
00:01:08,720 --> 00:01:12,000
presentation round two. 
Big hit of both events I would 

27
00:01:12,000 --> 00:01:13,920
say. 
All in all, it was a lot of 

28
00:01:13,920 --> 00:01:18,000
people across the fintech world.
It was fun kind of getting to 

29
00:01:18,280 --> 00:01:21,920
talk to them in between sessions
at the after party as well, 

30
00:01:21,920 --> 00:01:25,000
which was exclusive. 
We try not to talk about the 

31
00:01:25,000 --> 00:01:27,320
after party, but. 
Don't tell them I. 

32
00:01:27,520 --> 00:01:29,800
You know I ran into like. 
The guy who'd gotten viral 

33
00:01:29,800 --> 00:01:33,360
because he was trying to have 
people take out debt to make 

34
00:01:33,360 --> 00:01:35,120
investment accounts, I don't 
know if you saw that the other 

35
00:01:35,120 --> 00:01:36,160
day. 
I think it was like Basic 

36
00:01:36,160 --> 00:01:38,960
Capital, who I had no idea was 
even going to be there there. 

37
00:01:38,960 --> 00:01:40,880
So that's what I love about 
these events. 

38
00:01:40,880 --> 00:01:42,440
You know, we try to get great 
founders there. 

39
00:01:42,440 --> 00:01:44,760
And then I'm like, Oh yeah, I've
I've heard of your company. 

40
00:01:45,200 --> 00:01:47,760
So, yeah, it was a good crowd. 
And then excited about what we 

41
00:01:47,760 --> 00:01:49,640
have to share about what 
happened on stage. 

42
00:01:50,160 --> 00:01:53,360
Conversations ranged from a ton 
of different topics around 

43
00:01:53,360 --> 00:01:54,960
fintech. 
I mean, the premise was, you 

44
00:01:54,960 --> 00:01:57,920
know, fintech is back, baby. 
And then, you know, we had to go

45
00:01:57,920 --> 00:01:59,520
on stage and really ask, is it 
back? 

46
00:01:59,920 --> 00:02:01,800
But stable coins certainly are 
back. 

47
00:02:02,400 --> 00:02:06,040
The big resonating theme of the 
day was everyone is super pumped

48
00:02:06,040 --> 00:02:07,040
about stable coins. 
Probably. 

49
00:02:07,040 --> 00:02:08,919
Talking about stable coins is 
back. 

50
00:02:09,160 --> 00:02:12,640
Tripe a done, you know a billion
dollar acquisition, billion plus

51
00:02:12,640 --> 00:02:15,680
dollar acquisition of Bridge. 
We had the Bridge CEO as the 

52
00:02:15,680 --> 00:02:20,080
second person on stage. 
Jackie Reesus at Lead Bank is a 

53
00:02:20,080 --> 00:02:23,480
bank that's really backed stable
coin fintech companies. 

54
00:02:23,480 --> 00:02:25,800
So we definitely open the day 
strong with people who are 

55
00:02:26,080 --> 00:02:27,720
excited about. 
It, I mean, later in the day, 

56
00:02:27,720 --> 00:02:30,800
too, we had, you know, Eric 
Gleiman from Ramp who has 

57
00:02:30,800 --> 00:02:34,200
launched with Stripe and Bridge 
for a staple coin card. 

58
00:02:34,200 --> 00:02:34,600
Yeah. 
Yeah. 

59
00:02:35,080 --> 00:02:37,960
I mean, in the background of all
this, you know, Tether is making

60
00:02:37,960 --> 00:02:40,960
more money than per employee 
that almost like any company in 

61
00:02:40,960 --> 00:02:45,320
history off of stable coins. 
So everybody sees that there's 

62
00:02:45,320 --> 00:02:48,720
like a ton of money to be made. 
And of course, Congress was 

63
00:02:48,720 --> 00:02:52,040
passing legislation right as we 
were holding the event to make 

64
00:02:52,040 --> 00:02:55,880
stable coins much more legal to 
do or clearly legal. 

65
00:02:55,880 --> 00:02:58,360
So yeah, that I think that was 
theme A. 

66
00:02:58,360 --> 00:03:02,680
We've decided the selects for 
this episode will be from Theme 

67
00:03:02,680 --> 00:03:04,800
B, which I thought was equally 
interesting. 

68
00:03:05,080 --> 00:03:08,640
Yeah, theme BI mean there was a 
big debate that I feel like is 

69
00:03:08,640 --> 00:03:10,760
still kind of unsettled, which I
find interesting. 

70
00:03:10,760 --> 00:03:12,400
So we'll hear different 
perspectives on it later. 

71
00:03:12,400 --> 00:03:17,040
But around, you know, the idea 
of AI enabling startup founders 

72
00:03:17,040 --> 00:03:20,920
to go after service businesses, 
you know, like accounting, legal

73
00:03:20,920 --> 00:03:23,040
tech services, but you know, 
accounting, especially since 

74
00:03:23,040 --> 00:03:26,800
this was a fintech summit and 
where the merits are around, you

75
00:03:26,800 --> 00:03:30,000
know, fully pitching your 
startup as a service business 

76
00:03:30,000 --> 00:03:32,400
that can be fully enabled with 
AI, or rather, you know, 

77
00:03:32,400 --> 00:03:34,560
sticking to software and you 
know, being the software 

78
00:03:34,560 --> 00:03:37,200
provider for these businesses. 
Matt Harris at Bain Capital 

79
00:03:37,200 --> 00:03:40,840
Ventures gave a presentation 
which I think captured some of 

80
00:03:40,840 --> 00:03:43,680
the venture capitalists 
sentiment well, which is 

81
00:03:43,880 --> 00:03:46,400
financial services businesses 
are enormous. 

82
00:03:46,440 --> 00:03:50,240
You know, you think about 
accounting or broader services 

83
00:03:50,240 --> 00:03:53,800
businesses like law. 
There are many, many, many 

84
00:03:53,800 --> 00:03:56,720
billions of dollars of money to 
be made there. 

85
00:03:56,720 --> 00:03:59,440
But of course, they're human 
intensive businesses, the kind 

86
00:03:59,680 --> 00:04:02,080
that tech companies 
traditionally shy away from. 

87
00:04:02,400 --> 00:04:05,480
But the argument is that now, 
thanks to large language models,

88
00:04:05,960 --> 00:04:08,360
start up founders should go 
after those categories. 

89
00:04:08,360 --> 00:04:11,600
And Harris was arguing 
basically, definitely good for 

90
00:04:11,600 --> 00:04:15,400
start up founders, maybe not as 
good for venture capitalists 

91
00:04:15,400 --> 00:04:18,600
because they're still gonna be, 
you know, you might out operate 

92
00:04:18,600 --> 00:04:20,760
a law firm, but you're not 
necessarily going to build, you 

93
00:04:20,760 --> 00:04:24,160
know, a Facebook. 
So that was Harris's argument. 

94
00:04:24,440 --> 00:04:27,320
And then as you're going to 
listen in these conversations, I

95
00:04:27,320 --> 00:04:31,360
put that question to Digit CEO 
Jeff Siebert and Rogo CEO 

96
00:04:31,360 --> 00:04:34,120
Gabriel Stengel in our first 
conversation. 

97
00:04:34,120 --> 00:04:37,920
And then I talked with Josh 
Reeves, the CEO of Gusso, 

98
00:04:38,160 --> 00:04:40,720
probably one of the most 
experienced founders we had on 

99
00:04:40,720 --> 00:04:44,280
stage and asked them, you know, 
do you really want to run a 

100
00:04:44,280 --> 00:04:48,000
services business? 
Jeff at Digits had really stood 

101
00:04:48,000 --> 00:04:51,760
up an accounting practice to 
build their QuickBooks 

102
00:04:51,760 --> 00:04:55,360
competitor. 
And then Josh Augusto certainly 

103
00:04:55,360 --> 00:04:58,120
has built some of these services
businesses. 

104
00:04:58,120 --> 00:05:01,000
But again, I think to sort of 
see if they can build software. 

105
00:05:01,000 --> 00:05:04,640
So where did you guys net out in
terms of the argument? 

106
00:05:05,200 --> 00:05:06,800
Matt Harris, back to his 
presentation. 

107
00:05:06,800 --> 00:05:09,600
He ended very strong in an 
interesting way by basically 

108
00:05:09,600 --> 00:05:12,520
saying I'm tired of nibbling 
around the edges of fintech. 

109
00:05:12,840 --> 00:05:16,000
You know, this is a $33 trillion
opportunity, which I guess means

110
00:05:16,040 --> 00:05:18,840
all banking around the world. 
Well, it's also including 

111
00:05:18,840 --> 00:05:21,000
accounting, it's including 
accounts receivable, it's 

112
00:05:21,000 --> 00:05:24,280
including, you know, health 
insurance brokerages, as much 

113
00:05:24,280 --> 00:05:27,000
money as possible that you know,
any money industry I. 

114
00:05:27,200 --> 00:05:29,680
Don't know if you guys heard 
this, but it elicited a woo. 

115
00:05:29,680 --> 00:05:32,120
Yeah, from Jackie Reese's. 
I at least heard it. 

116
00:05:32,120 --> 00:05:33,720
She was right in front of me 
while he was she also. 

117
00:05:33,720 --> 00:05:35,800
Sent me an e-mail saying she 
loved that quote too. 

118
00:05:35,840 --> 00:05:37,920
So yeah, she clearly agreed with
that idea. 

119
00:05:38,280 --> 00:05:40,640
Well, who wouldn't want to be 
part of a $33 trillion 

120
00:05:40,640 --> 00:05:42,520
opportunity? 
But but I think that kind of 

121
00:05:42,520 --> 00:05:45,080
speaks to the challenge that 
fintech has had, which is like, 

122
00:05:45,080 --> 00:05:48,760
is this just some sort of, you 
know, skimming of, of businesses

123
00:05:48,760 --> 00:05:51,440
that the big banks and other 
financial institutions aren't 

124
00:05:51,440 --> 00:05:53,640
already taking? 
Or is it like truly disruptive, 

125
00:05:53,680 --> 00:05:57,280
right? 
Is fintech, Venmo transfers, 

126
00:05:57,280 --> 00:06:01,520
weird loans, helping people, you
know, finance their burritos? 

127
00:06:01,520 --> 00:06:05,440
Or is it, you know, the core of 
the American economy that big 

128
00:06:05,440 --> 00:06:07,560
services companies have been 
able to deliver? 

129
00:06:07,560 --> 00:06:10,280
Or a bunch of small ones, 
Honestly, a lot of small 

130
00:06:10,280 --> 00:06:13,480
businesses doing accounting and 
law and all these things that 

131
00:06:13,480 --> 00:06:15,360
they want to tackle. 
Right. 

132
00:06:15,360 --> 00:06:17,880
And like bringing it back to 
like large language models, like

133
00:06:17,880 --> 00:06:21,120
is this the entry point to this 
larger business, to this larger 

134
00:06:21,120 --> 00:06:24,920
opportunity or you know, are we 
still going to be about these 

135
00:06:24,920 --> 00:06:28,760
kind of marginal disruptive 
plays that can build real 

136
00:06:28,760 --> 00:06:32,360
businesses like Klarna or a 
firm, but aren't, you know, I 

137
00:06:32,360 --> 00:06:34,680
don't think are really 
necessarily taking 33 trillion 

138
00:06:34,680 --> 00:06:38,440
as the opportunity there. 
I will say, I mean the success 

139
00:06:38,440 --> 00:06:42,200
of Rogo so far, you know, 
they're basically building an AI

140
00:06:42,200 --> 00:06:45,440
agent that can automate the work
of an investment making analyst,

141
00:06:45,440 --> 00:06:49,200
which is very specific, but has 
taken off, especially in an 

142
00:06:49,200 --> 00:06:52,800
industry where, you know, unlike
accounting, you can get most of 

143
00:06:52,800 --> 00:06:55,640
the way there and the people 
will say, you know, this is good

144
00:06:55,640 --> 00:06:59,040
enough, we got this done. 
And so that is one place where I

145
00:06:59,040 --> 00:07:01,880
can see the service automation 
be very promising. 

146
00:07:01,960 --> 00:07:04,520
But that being said, it is a 
pretty specific service that 

147
00:07:04,520 --> 00:07:06,560
they're automating there. 
So we'll see how that expands 

148
00:07:06,560 --> 00:07:08,760
accounting. 
Of course you know it's tricky 

149
00:07:08,760 --> 00:07:10,440
'cause you don't want to mess 
that up. 

150
00:07:10,920 --> 00:07:15,040
I, I think that's a key point, 
right, Gabe with Rogo is, you 

151
00:07:15,040 --> 00:07:17,320
know, they're trying to replace 
like the Goldman Sachs analyst 

152
00:07:17,320 --> 00:07:20,800
or even just, you know, give you
a sort of cheap analyst that 

153
00:07:20,800 --> 00:07:23,240
that has a rundown on companies 
you care about. 

154
00:07:23,480 --> 00:07:25,520
You know, you imagine the 
bankers already like going on 

155
00:07:25,520 --> 00:07:27,680
ChatGPT and be like quick before
this client meeting. 

156
00:07:27,680 --> 00:07:30,880
Tell me the download on this 
random like oil and gas company 

157
00:07:30,880 --> 00:07:33,520
I've never heard of. 
And so there, it's like Rogo. 

158
00:07:33,520 --> 00:07:36,600
Doesn't need to be 100% crack. 
They just need to be better than

159
00:07:36,600 --> 00:07:39,800
sort of the lazy version of 
this. 

160
00:07:39,800 --> 00:07:42,320
And obviously they have, you 
know, aspirations to be much 

161
00:07:42,320 --> 00:07:45,240
higher quality than that, but 
they don't need to be 100% 

162
00:07:45,240 --> 00:07:47,080
correct. 
They need to help you get pretty

163
00:07:47,080 --> 00:07:50,000
broad coverage and be able to 
move quicker than you can right 

164
00:07:50,000 --> 00:07:52,080
now. 
Whereas accounting, as Jeff was 

165
00:07:52,080 --> 00:07:55,720
saying, you know, you need to 
get it right and you know, a lot

166
00:07:55,720 --> 00:07:59,360
of what they would need language
models for is sort of the boring

167
00:07:59,360 --> 00:08:03,320
business of tagging all your 
expenses, which is somehow is 

168
00:08:03,320 --> 00:08:08,480
still a task that's like cannot 
be fully solved by technology 

169
00:08:08,480 --> 00:08:10,880
right now. 
And so he talks about how they 

170
00:08:10,880 --> 00:08:13,600
use machine learning for a lot 
of the tagging because he finds 

171
00:08:13,600 --> 00:08:17,240
it much more accurate. 
And then when they really can't 

172
00:08:17,240 --> 00:08:19,840
figure it out, they sort of see 
if a language model can sort of 

173
00:08:20,120 --> 00:08:22,560
crack it, but they're not really
relying on it. 

174
00:08:22,560 --> 00:08:26,360
So Madeline, I think exactly 
like you're saying, it really 

175
00:08:26,360 --> 00:08:29,480
depends what business you're in,
how much you trust language 

176
00:08:29,480 --> 00:08:31,320
models that aren't totally 
consistent. 

177
00:08:31,520 --> 00:08:35,039
You know, Jeff made the point 
like, well, ultimately you can't

178
00:08:35,039 --> 00:08:37,720
sue the AI. 
So if you mess up my accounting 

179
00:08:37,720 --> 00:08:42,000
books, like, you'll sue the CPA.
So there's a different stakes. 

180
00:08:42,840 --> 00:08:45,400
You know, the last thing I'll 
point to before we throw it to 

181
00:08:45,400 --> 00:08:48,520
you can listen to them. 
I did all that reporting on 

182
00:08:48,800 --> 00:08:52,240
bench, the accounting firm in 
Canada that blew up where the 

183
00:08:52,240 --> 00:08:55,320
founder basically said, oh, if 
you hadn't pushed me out, this 

184
00:08:55,320 --> 00:08:57,200
would might have turned out 
differently. 

185
00:08:57,600 --> 00:09:00,840
And you know, you can read those
stories and it had a lot to do 

186
00:09:00,840 --> 00:09:03,840
with the particular investors 
and executives. 

187
00:09:04,520 --> 00:09:07,560
There was this real lesson that 
building a startup that hires a 

188
00:09:07,560 --> 00:09:09,760
bunch of accountants is is a big
headache. 

189
00:09:09,800 --> 00:09:12,960
And so building a tech company 
where you're going to try and 

190
00:09:13,520 --> 00:09:17,280
hire all the people and then 
slowly figure out how to do it 

191
00:09:17,280 --> 00:09:20,720
better with software, it's hard 
to guarantee that you can bridge

192
00:09:20,720 --> 00:09:22,360
that gap. 
And then one day you may find 

193
00:09:22,360 --> 00:09:26,560
yourself just operating a 
regular accounting firm and your

194
00:09:26,560 --> 00:09:30,160
investors want you to have, you 
know, the margin profile of a 

195
00:09:30,520 --> 00:09:32,440
hot start up. 
And then you're in big trouble. 

196
00:09:32,720 --> 00:09:35,640
Good recap of the event, guys, 
and I think now we should just 

197
00:09:35,800 --> 00:09:38,720
kick it on over to two of the 
highlights of the event. 

198
00:09:43,960 --> 00:09:46,400
Very excited about this panel. 
We're going to have a great 

199
00:09:46,400 --> 00:09:49,400
discussion about what is 
possible with artificial 

200
00:09:49,400 --> 00:09:54,040
intelligence in fintech. 
We have two of the very well 

201
00:09:54,040 --> 00:09:56,120
financed cutting edge startups 
here. 

202
00:09:56,120 --> 00:09:58,560
Both of you can certainly claim 
that. 

203
00:09:58,880 --> 00:10:01,040
Gabe, you want to start with 
Rogo and just give the quick 

204
00:10:01,040 --> 00:10:03,520
like what you guys are doing? 
Happy to. 

205
00:10:03,520 --> 00:10:05,400
Thanks for having me. 
I'm Gabe. 

206
00:10:05,400 --> 00:10:08,840
I'm the CEO and founder of Rogo.
We're training an AI analyst for

207
00:10:08,840 --> 00:10:11,200
investment banks, private equity
firms, and hedge funds. 

208
00:10:11,960 --> 00:10:15,560
We're a Series B company backed 
by Khosla, Thrive, Eric Schmidt.

209
00:10:15,560 --> 00:10:19,000
A number of other folks deployed
a, you know, a number of public 

210
00:10:19,000 --> 00:10:21,440
investment banks, large 
alternative asset managers, 

211
00:10:21,600 --> 00:10:24,600
large public equity investors. 
And you're, you're replacing the

212
00:10:24,600 --> 00:10:27,080
banking analyst, right? 
Not yet, not yet. 

213
00:10:27,200 --> 00:10:31,000
Banking analyst right now, I 
mean, everything can be broken 

214
00:10:31,000 --> 00:10:34,160
up into augmentation and 
automation and we're much more 

215
00:10:34,160 --> 00:10:36,840
on the augmentation side still, 
even though, you know, I think 

216
00:10:36,960 --> 00:10:38,960
we and a lot of folks see that 
changing quickly. 

217
00:10:39,640 --> 00:10:42,160
And Jeff, the Digit story. 
Yeah. 

218
00:10:42,160 --> 00:10:44,880
So two months ago, we launched 
the first feature complete 

219
00:10:44,880 --> 00:10:47,960
replacement to QuickBooks in 20 
years since 0 came out. 

220
00:10:47,960 --> 00:10:50,400
It was a long build. 
We started the company in 2018 

221
00:10:50,400 --> 00:10:52,840
and we're basically in R&D 
stealth mode ever since. 

222
00:10:53,120 --> 00:10:56,080
And so our goal has been to 
basically automate and make 

223
00:10:56,080 --> 00:10:58,920
small business finance real time
intuitive, actually helpful to 

224
00:10:58,920 --> 00:11:01,200
the business owner, not put them
in a position where they're 

225
00:11:01,200 --> 00:11:03,920
waiting two to three weeks to 
get a black and white PNL, which

226
00:11:03,920 --> 00:11:05,440
is sort of. 
You're coming for a QuickBooks. 

227
00:11:05,680 --> 00:11:08,560
Space 100%, yeah. 
And there there are lots of 

228
00:11:09,480 --> 00:11:12,280
problems with QuickBooks. 
Some of them have nothing to do 

229
00:11:12,280 --> 00:11:14,880
with what a foundation model 
could produce, and some. 

230
00:11:15,160 --> 00:11:18,560
So dude, how do you break down 
Yeah, how much of what's making 

231
00:11:18,560 --> 00:11:21,960
digits different has to do with 
things foundation models can 

232
00:11:21,960 --> 00:11:22,800
produce? 
So. 

233
00:11:22,840 --> 00:11:25,080
It is pretty fundamental. 
I would broaden it to machine 

234
00:11:25,080 --> 00:11:26,800
learning. 
So when I started the company in

235
00:11:26,800 --> 00:11:30,280
2018, our premise was can we 
build the first GLA general 

236
00:11:30,280 --> 00:11:33,240
Ledger for accounting that's ML 
native. 

237
00:11:33,560 --> 00:11:35,760
We honestly got very lucky with 
this whole AI wave. 

238
00:11:35,960 --> 00:11:37,120
Now it's AI native. 
Great. 

239
00:11:37,120 --> 00:11:40,720
OK. 
But we primarily do traditional 

240
00:11:40,720 --> 00:11:42,320
machine learning. 
If you look at accounting as a 

241
00:11:42,320 --> 00:11:44,880
field, it is predictive, It is 
not generative. 

242
00:11:45,040 --> 00:11:47,800
You do not want an LLM 
hallucinating your books. 

243
00:11:48,160 --> 00:11:51,360
And so we custom train and run 
our own models in production. 

244
00:11:51,520 --> 00:11:54,600
We do fall back to LLM 
Foundation models as sort of a 

245
00:11:54,600 --> 00:11:57,480
worst case scenario. 
But our story to the industry is

246
00:11:57,480 --> 00:11:58,960
we are automating the 
bookkeeping. 

247
00:11:59,120 --> 00:12:02,400
We automate 93% of it today. 
And what's unique about the 

248
00:12:02,400 --> 00:12:06,000
accounting industry is they want
that they are as a profession 

249
00:12:06,080 --> 00:12:08,720
trying to up level the 
profession away from bookkeeping

250
00:12:08,880 --> 00:12:12,160
and into advisory and sort of 
client communication work. 

251
00:12:12,440 --> 00:12:14,000
And so that's what they're going
for, I mean. 

252
00:12:14,000 --> 00:12:17,240
Basically, you use machine 
learning tag expenses, get it 

253
00:12:17,240 --> 00:12:20,200
100% right, and then if you're 
tagging system doesn't work, 

254
00:12:20,200 --> 00:12:22,880
you're like, all right, we'll 
try an LLM and see if it's sort 

255
00:12:22,880 --> 00:12:26,040
of more free thinking style can 
tag it itself. 

256
00:12:26,160 --> 00:12:27,360
That's the right way to think 
about it. 

257
00:12:27,360 --> 00:12:30,280
So if a business has seen the 
transaction before, our 

258
00:12:30,280 --> 00:12:32,640
predictive models are 
effectively 100% perfect. 

259
00:12:32,920 --> 00:12:36,200
If the transaction is novel to 
the business, we then fall back 

260
00:12:36,200 --> 00:12:39,160
to different tiers of models 
that ultimately result in an 

261
00:12:39,160 --> 00:12:41,000
agent. 
Like what does your bookkeeper 

262
00:12:41,000 --> 00:12:42,120
do? 
If you have something totally 

263
00:12:42,120 --> 00:12:43,440
new? 
They Google it. 

264
00:12:43,640 --> 00:12:45,360
What does the agent do? 
It Googles it. 

265
00:12:47,520 --> 00:12:49,560
Gabe, that I don't know. 
I don't take that as the most 

266
00:12:49,560 --> 00:12:52,400
optimistic view of what LLMS can
do. 

267
00:12:52,400 --> 00:12:54,400
It's sort of like, oh, we have 
our accurate ones and then we 

268
00:12:54,400 --> 00:12:57,760
have our like guy who Googles 
things and like sometimes he's 

269
00:12:57,760 --> 00:13:00,040
helpful. 
Like you're much more leaned 

270
00:13:00,040 --> 00:13:04,720
into, like LLMS can deliver real
value to bankers. 

271
00:13:04,920 --> 00:13:07,920
What's possible right now? 
How accurate is your software? 

272
00:13:07,960 --> 00:13:09,720
It's. 
Funny, Jeff and I were were just

273
00:13:09,800 --> 00:13:12,240
discussing it backstage because 
we have two very different 

274
00:13:12,360 --> 00:13:13,800
products and very different 
users. 

275
00:13:13,800 --> 00:13:17,040
If you get financials that are 
not 100% accurate and you are a 

276
00:13:17,120 --> 00:13:19,400
small business and you don't 
know how to go in and audit 

277
00:13:19,400 --> 00:13:20,760
those books yourself, that's a 
pain. 

278
00:13:21,040 --> 00:13:23,280
If I give something that's, you 
know, good enough for an 

279
00:13:23,280 --> 00:13:26,160
associate to check, you know, 
maybe as accurate as their first

280
00:13:26,160 --> 00:13:28,840
year analyst who's not 
perfectly, perfectly accurate, 

281
00:13:28,960 --> 00:13:31,200
that's actually very value 
additive. 

282
00:13:31,880 --> 00:13:35,200
Part of the, the benefit we had 
is that when we went in, folks 

283
00:13:35,200 --> 00:13:38,520
were used to using products like
Chacha, BT and a lot of our 

284
00:13:38,600 --> 00:13:41,440
early users realized these tools
can be very valuable without 

285
00:13:41,440 --> 00:13:44,120
being 100% accurate. 
And so I mean, the last thing I 

286
00:13:44,120 --> 00:13:46,800
would ever say when selling is 
that we are in 100% accurate 

287
00:13:46,800 --> 00:13:48,240
tool. 
What I would say is you can get 

288
00:13:48,240 --> 00:13:50,880
a lot more value than you think 
before they're 100% accurate. 

289
00:13:50,880 --> 00:13:52,760
You need to be better than 
someone who's right out of 

290
00:13:52,760 --> 00:13:56,200
college, and you need to be 
better than a lazy banker using 

291
00:13:56,200 --> 00:13:57,200
chachi. 
Exactly. 

292
00:13:57,200 --> 00:14:00,160
And as we learn, right, the way 
your intern learns to be an 

293
00:14:00,160 --> 00:14:02,640
analyst, learns to be an 
associate, VP, director, when 

294
00:14:02,640 --> 00:14:04,720
we're at director level, we're 
not going to get wooed by a 

295
00:14:04,720 --> 00:14:07,240
better paycheck somewhere else 
and all the enterprise value 

296
00:14:07,240 --> 00:14:09,520
isn't going to walk out the door
and go to a competitive bank. 

297
00:14:10,160 --> 00:14:12,440
And what I mean, what's the 
coolest thing you can do today? 

298
00:14:12,440 --> 00:14:15,160
Like where do you think Rogo is 
really excelling? 

299
00:14:15,160 --> 00:14:17,560
Like what is the sort of query 
I? 

300
00:14:17,600 --> 00:14:19,160
Mean it does things why you 
should use. 

301
00:14:19,320 --> 00:14:21,960
I was an investment banking 
analyst at at Lazard. 

302
00:14:22,280 --> 00:14:25,320
I did buy Side M and A coverage 
for healthcare companies and I 

303
00:14:25,320 --> 00:14:28,160
didn't know anything about 
healthcare companies and we 

304
00:14:28,160 --> 00:14:32,000
would help large biopharma 
companies spend $5 billion to 

305
00:14:32,000 --> 00:14:35,520
buy biotechs and it can do 
almost all the work I did in the

306
00:14:35,520 --> 00:14:37,280
analysis of those companies, 
right? 

307
00:14:37,280 --> 00:14:39,800
Like looking at their R&D 
pipeline, thinking about the 

308
00:14:39,800 --> 00:14:42,680
different you sort of oncology 
areas that might be additive for

309
00:14:42,680 --> 00:14:45,000
Johnson and Johnson or or 
someone similar and then 

310
00:14:45,000 --> 00:14:48,640
preparing a presentation on why 
it would be additive to J and JS

311
00:14:48,640 --> 00:14:51,240
overall M and a strategy. 
I mean we can put together the 

312
00:14:51,240 --> 00:14:53,480
materials diligence, the 
company, look through their 

313
00:14:53,480 --> 00:14:57,080
financials, do a do a lot of 
what these folks are doing. 

314
00:14:58,600 --> 00:15:01,640
Jeff, I mean, we saw sort of the
presentation from Matt Harris 

315
00:15:01,720 --> 00:15:04,520
earlier, which is like get into 
the services business. 

316
00:15:04,520 --> 00:15:08,400
And I, I wrote last year a fair 
bit about like bench accounting,

317
00:15:08,400 --> 00:15:11,480
which tried to sort of do 
accounting plus the actual 

318
00:15:11,840 --> 00:15:14,080
accountants. 
I mean, you're going after 

319
00:15:14,080 --> 00:15:15,520
QuickBooks, right? 
You are not building an 

320
00:15:15,520 --> 00:15:17,800
accountant team. 
You've, I think you've used some

321
00:15:17,800 --> 00:15:21,040
accounting firms to sort of 
understand your product, but why

322
00:15:21,040 --> 00:15:24,680
not go I guess whole hog and say
we will be your accountant. 

323
00:15:24,680 --> 00:15:27,360
We know how to use our software 
better than anybody. 

324
00:15:27,360 --> 00:15:29,560
We'll take advantage of it and 
do the whole thing. 

325
00:15:29,800 --> 00:15:31,920
This is such a good question. 
So and just to clarify, we 

326
00:15:31,920 --> 00:15:33,080
actually do have an accounting 
team. 

327
00:15:33,080 --> 00:15:35,560
We do have an accounting firm. 
Some of you may have seen our 

328
00:15:35,560 --> 00:15:37,280
billboards. 
We do full service accounting 

329
00:15:37,280 --> 00:15:38,920
for hundreds of top startups 
now, OK. 

330
00:15:39,320 --> 00:15:42,160
That is not the business that we
are not scaling that OK. 

331
00:15:42,240 --> 00:15:43,160
Yeah, because. 
We are. 

332
00:15:43,160 --> 00:15:45,040
I didn't think I was that wrong.
Yeah, yeah, yeah, we do it, but 

333
00:15:45,040 --> 00:15:46,120
we're not going to do it at 
scale. 

334
00:15:46,120 --> 00:15:48,120
So we have. 
Capped the client, got it, got 

335
00:15:48,120 --> 00:15:49,960
it. 
The reason we did that is we 

336
00:15:49,960 --> 00:15:52,320
need to present a model firm to 
the industry, right? 

337
00:15:52,600 --> 00:15:54,440
And so if you look at the 
accounting industry, they're of 

338
00:15:54,440 --> 00:15:57,200
course traditionally relatively 
risk averse, relatively slow to 

339
00:15:57,200 --> 00:15:59,520
adopt new tools. 
The cloud transition took them 

340
00:15:59,520 --> 00:16:01,000
15 years. 
I'm not joking. 

341
00:16:01,360 --> 00:16:03,320
And so we want to accelerate 
that a bit. 

342
00:16:03,520 --> 00:16:07,080
And so we did build out a model 
firm to show you how you can run

343
00:16:07,080 --> 00:16:09,920
digits as a practice that is not
the core business. 

344
00:16:09,920 --> 00:16:12,040
And so I do think there's been a
trap in the space. 

345
00:16:12,200 --> 00:16:15,080
If you look at some of the well 
known sort of previous offerings

346
00:16:15,080 --> 00:16:18,000
that have tried to build a 
services business with some 

347
00:16:18,000 --> 00:16:20,680
internal tech. 
And I mean you even see Atrium 

348
00:16:20,680 --> 00:16:22,120
that failed at that in the legal
space. 

349
00:16:22,280 --> 00:16:25,120
I think it's incredibly hard to 
bring together a software 

350
00:16:25,120 --> 00:16:28,080
business and a services business
and actually scale it to high 

351
00:16:28,080 --> 00:16:29,920
margin. 
And so we've been really 

352
00:16:29,920 --> 00:16:32,280
disciplined on, we hire software
engineers. 

353
00:16:32,280 --> 00:16:35,120
It is a software product like 
there is no slippery slope of 

354
00:16:35,120 --> 00:16:37,320
humans doing any of the work. 
Why? 

355
00:16:37,320 --> 00:16:40,280
Why has QuickBooks been able to 
hold on so well? 

356
00:16:40,640 --> 00:16:43,920
Yes, it's a marketplace product,
believe it or not. 

357
00:16:44,160 --> 00:16:47,200
And when QuickBooks came onto 
the scene in the early 90s, it 

358
00:16:47,200 --> 00:16:48,880
also had taken them five years 
to build. 

359
00:16:49,440 --> 00:16:52,080
They were sort of building on 
the backs of the personal 

360
00:16:52,080 --> 00:16:53,760
computer revolution. 
For the first time, you could 

361
00:16:53,760 --> 00:16:56,000
really do accounting for your 
own business, like in your own 

362
00:16:56,000 --> 00:16:58,480
office. 
And the problem is you need both

363
00:16:58,480 --> 00:17:02,040
the business owner and the 
accountant to really buy in on. 

364
00:17:02,040 --> 00:17:04,880
I'm going to use this software. 
And now 30 years later, 

365
00:17:05,079 --> 00:17:06,760
accountants view it as a career 
choice. 

366
00:17:06,760 --> 00:17:08,599
It's a religious preference. 
Like I'm a QuickBooks 

367
00:17:08,599 --> 00:17:10,440
accountant, I'm a zero 
accountant, I'm a NetSuite 

368
00:17:10,440 --> 00:17:12,440
accountant. 
And so that's the barrier to 

369
00:17:12,440 --> 00:17:15,160
break down. 
Fortunately, there is a macro 

370
00:17:15,160 --> 00:17:19,640
trend in our favour. 
So 33% of fewer people are 

371
00:17:19,640 --> 00:17:21,200
pursuing Cpas. 
Gen. 

372
00:17:21,200 --> 00:17:23,560
Z does not want to be an 
accountant, which you may or may

373
00:17:23,560 --> 00:17:26,880
not blame them. 
And 75% of Cpas are at 

374
00:17:26,880 --> 00:17:28,880
retirement age. 
And so they're seeing this 

375
00:17:28,880 --> 00:17:32,560
talent crisis as the sort of 
final motivation of, OK, we need

376
00:17:32,560 --> 00:17:34,680
to actually change software, We 
need to automate our role, We 

377
00:17:34,680 --> 00:17:37,960
need to up level the profession.
So I think that trend is very 

378
00:17:37,960 --> 00:17:40,640
favorable. 
Gabe, you know, want to sort of 

379
00:17:40,640 --> 00:17:43,400
give a guide book for other 
companies, you know, you're, 

380
00:17:43,960 --> 00:17:46,520
you're, you're sort of at the 
cutting edge of using language 

381
00:17:46,520 --> 00:17:50,000
models in your business. 
I'm curious how much you think 

382
00:17:50,240 --> 00:17:54,520
what we've seen with ChatGPT is 
an endorsement of text based 

383
00:17:54,520 --> 00:17:58,280
exchange interfaces. 
Like do you think the regular 

384
00:17:58,280 --> 00:18:02,920
person or your banker customer 
wants to type inquiries and get 

385
00:18:02,920 --> 00:18:05,360
great responses or do you think 
that's sort of an entry point 

386
00:18:05,360 --> 00:18:06,640
over time? 
It's more and more like 

387
00:18:06,640 --> 00:18:09,640
software. 
Look, I think we're lucky in 

388
00:18:09,640 --> 00:18:12,600
that we, we have a lot of users 
who use the product we have 

389
00:18:12,600 --> 00:18:14,560
today. 
I don't know if that means we 

390
00:18:14,560 --> 00:18:17,520
have, you know, any predictive 
abilities about what that UX 

391
00:18:17,520 --> 00:18:21,160
should be 5 years from now. 
For me, the abstraction is what 

392
00:18:21,160 --> 00:18:23,560
is the easiest way for me to 
communicate with someone on my 

393
00:18:23,560 --> 00:18:25,640
team. 
It's probably Slack, e-mail, 

394
00:18:25,640 --> 00:18:28,000
text message, phone call. 
You know, if I ask for a 

395
00:18:28,000 --> 00:18:30,320
deliverable, I get sent to PDFI 
review it. 

396
00:18:30,480 --> 00:18:33,240
Occasionally I'll jump into the 
PowerPoint, the Excel backup 

397
00:18:33,240 --> 00:18:35,320
myself. 
I think that's a very human way 

398
00:18:35,320 --> 00:18:37,720
to work and there's a lot of 
throughput in that, you know, 

399
00:18:37,720 --> 00:18:39,280
you can communicate, indicate a 
ton. 

400
00:18:39,520 --> 00:18:41,840
I think at the limit, that's 
what you want these interfaces 

401
00:18:41,840 --> 00:18:44,320
to look like. 
If you're building an assistant,

402
00:18:44,320 --> 00:18:47,200
if you're building, you know, an
analyst replacement, if you're 

403
00:18:47,200 --> 00:18:50,000
building Jeff's business, you 
know, I, I'd rather just have my

404
00:18:50,000 --> 00:18:52,760
books integrate directly into, 
you know, whatever systems those

405
00:18:52,760 --> 00:18:54,840
need to go into. 
I don't know anything about. 

406
00:18:54,960 --> 00:18:58,160
No, this is actually, this is 
great because this is such a a 

407
00:18:58,160 --> 00:18:59,960
product specific difference, 
right? 

408
00:18:59,960 --> 00:19:02,240
Like as a startup founder, 
business owner, I don't want 

409
00:19:02,240 --> 00:19:04,520
another person to manage. 
I don't want another thing to 

410
00:19:04,520 --> 00:19:07,960
talk to, I just want the 
accounting done right and so we 

411
00:19:07,960 --> 00:19:11,560
very explicitly have no chat UI 
in the product the data comes 

412
00:19:11,560 --> 00:19:12,560
in. 
It is booked where? 

413
00:19:12,560 --> 00:19:15,280
The output of Rogue, oh, I mean 
analysts would produce a report,

414
00:19:15,280 --> 00:19:17,200
so it's not that. 
Yeah, I got you, but. 

415
00:19:17,200 --> 00:19:19,320
The way you would iterate might 
be, Hey, why'd you do this? 

416
00:19:19,320 --> 00:19:21,160
Hey, add a page on this. 
Hey, Are you sure? 

417
00:19:21,160 --> 00:19:23,440
You know, that was the post 
money valuation for the company 

418
00:19:23,440 --> 00:19:25,880
and in the comp set. 
But I mean, it's, it's, you 

419
00:19:25,880 --> 00:19:28,280
know, for, for Jeff, it's like, 
I don't want to hire an FPNA 

420
00:19:28,320 --> 00:19:30,440
person, right? 
We're getting to the scale where

421
00:19:30,440 --> 00:19:32,760
it's like, you know, we need 
someone and I really don't want 

422
00:19:32,760 --> 00:19:34,080
to do it. 
I'd rather just have those 

423
00:19:34,080 --> 00:19:36,800
systems work and I don't want to
talked about it with someone 

424
00:19:36,840 --> 00:19:37,560
either. 
Right. 

425
00:19:37,920 --> 00:19:41,480
Are you an evangelist when you 
talk to other fintech founders 

426
00:19:41,480 --> 00:19:44,280
of like they should be applying 
foundation models to their 

427
00:19:44,280 --> 00:19:46,000
business? 
Or are you more on the side of 

428
00:19:46,000 --> 00:19:48,840
like, we picked the right area, 
it makes sense where we're doing

429
00:19:48,840 --> 00:19:50,120
it. 
And like, I don't know if it 

430
00:19:50,120 --> 00:19:51,960
makes sense for your business. 
Yeah. 

431
00:19:52,640 --> 00:19:55,200
I was I yesterday at Khosla 
Ventures is one of our big 

432
00:19:55,200 --> 00:19:58,080
backers and, and they had a big 
summit yesterday and, and 

433
00:19:58,080 --> 00:20:00,280
something that one of the 
speakers kept hammering was it's

434
00:20:00,280 --> 00:20:04,040
very hard for incumbents in a 
space to innovate, to brainstorm

435
00:20:04,040 --> 00:20:06,280
the sort of, you know, not just 
the incremental way to apply 

436
00:20:06,280 --> 00:20:09,360
LLMS to your business, but the 
whole new business model that 

437
00:20:09,360 --> 00:20:11,760
emerges. 
And part partly why digits is so

438
00:20:11,760 --> 00:20:13,960
fascinating is because it's 
innovating on the business model

439
00:20:14,160 --> 00:20:16,560
in addition to, you know, the 
actual technology. 

440
00:20:16,760 --> 00:20:19,640
But I mean, I was, I was pushing
Jeff in our we had lunch 

441
00:20:19,640 --> 00:20:20,960
together to catch up before 
this. 

442
00:20:20,960 --> 00:20:22,480
And I'm saying it sounds like 
you're not using. 

443
00:20:22,640 --> 00:20:26,000
LLMS enough and he had a great 
retort and he obviously knows 

444
00:20:26,000 --> 00:20:28,840
his business better than I do. 
But I mean, I think anyone who's

445
00:20:28,840 --> 00:20:32,520
not trying to use these tools in
in kind of a step change way as 

446
00:20:32,520 --> 00:20:35,520
opposed to an incrementalist way
is is waiting for someone else 

447
00:20:35,520 --> 00:20:37,800
to Jeff. 
You're an incrementalist. 

448
00:20:39,600 --> 00:20:41,200
This is great. 
No, I mean to be clear, we do 

449
00:20:41,200 --> 00:20:43,560
use LLMS, we do use agents. 
We've been running in in 

450
00:20:43,560 --> 00:20:46,880
production for 18 months now. 
I'd say we are extremely 

451
00:20:46,880 --> 00:20:50,000
disciplined on making sure we 
can guarantee the accuracy of 

452
00:20:50,000 --> 00:20:52,200
the output. 
And so for an example like where

453
00:20:52,200 --> 00:20:55,960
we do use LLMS, we actively 
prevent them from doing math 

454
00:20:56,240 --> 00:20:58,080
because yes, they've gotten 
better at math, but they can 

455
00:20:58,080 --> 00:21:01,040
still slip up. 
Our team literally wrote an RPN 

456
00:21:01,040 --> 00:21:04,240
calculator tool that we give to 
the LLMS and we tell them if you

457
00:21:04,240 --> 00:21:06,320
need to do math, do not try use 
this tool. 

458
00:21:06,760 --> 00:21:08,680
And so there are a bunch of 
things you can do to put 

459
00:21:08,800 --> 00:21:10,480
safeguards around it. 
What? 

460
00:21:10,480 --> 00:21:13,800
Do you think about that? 
Well, you know, I, I think it's 

461
00:21:13,800 --> 00:21:16,640
kind of the analogy I would use 
and it's going to be a really 

462
00:21:16,640 --> 00:21:18,320
strained analogy, so bear with 
me. 

463
00:21:18,320 --> 00:21:21,160
But the sort of Jeff Bezos thing
of, you know, your margin is my 

464
00:21:21,160 --> 00:21:24,040
opportunity, your accuracy 
threshold is someone else's 

465
00:21:24,040 --> 00:21:25,920
opportunity because they're 
going to build something that's 

466
00:21:25,920 --> 00:21:27,520
not going to work today or 
tomorrow. 

467
00:21:27,640 --> 00:21:29,800
But all of a sudden the base 
models will be good enough where

468
00:21:29,960 --> 00:21:33,400
you could skip all these layers 
of complexity and structure to, 

469
00:21:33,440 --> 00:21:35,000
you know, make up for that 
accuracy. 

470
00:21:35,160 --> 00:21:37,360
And suddenly, you know, they 
will have some of that more 

471
00:21:37,360 --> 00:21:39,320
human reasoning and be capable 
of so much more. 

472
00:21:39,560 --> 00:21:43,240
I don't know if that's right in.
Jeff on like the continued 

473
00:21:43,240 --> 00:21:47,200
improvement of the models and if
you spend too much time fixing 

474
00:21:47,200 --> 00:21:49,920
the models today, you're missing
out on the, I mean saving the 

475
00:21:49,920 --> 00:21:51,400
problem is. 
How do you thread the needle? 

476
00:21:51,400 --> 00:21:54,240
How do you get enough adoption 
and traction today to, you know,

477
00:21:54,240 --> 00:21:56,880
validate putting more in R&D for
the future? 

478
00:21:56,880 --> 00:21:59,480
And like, depending on how 
accurate you need to be, yeah, 

479
00:21:59,480 --> 00:22:02,040
you might need to do some 
anachronistic things or like, 

480
00:22:02,040 --> 00:22:04,720
you know, not use it, just throw
LLMS at every problem, right? 

481
00:22:04,920 --> 00:22:07,120
It's not, you know, you don't 
just want to use LLMS for 

482
00:22:07,120 --> 00:22:10,920
everything. 
But I do think that like, you 

483
00:22:10,920 --> 00:22:14,400
know, if if you aren't, if you 
aren't willing to, to accept 

484
00:22:14,400 --> 00:22:17,920
some faultiness, some bugginess,
some hallucination in parts of a

485
00:22:17,920 --> 00:22:21,280
product like someone else will. 
And then someone else will 

486
00:22:21,280 --> 00:22:23,920
potentially figure out the new 
product paradigm or UX that's 

487
00:22:23,920 --> 00:22:26,680
going to work when the models 
just get, you know, 10% more 

488
00:22:26,680 --> 00:22:30,120
reliable. 
Jeff, we're in this moment 

489
00:22:30,120 --> 00:22:32,280
where, you know, people are 
like, oh, we don't need software

490
00:22:32,280 --> 00:22:33,600
anymore. 
You're going to use like a 

491
00:22:33,600 --> 00:22:36,080
coding agent to build it. 
Like I'm going to build my own 

492
00:22:36,080 --> 00:22:38,960
custom model. 
Like why isn't the next digits 

493
00:22:38,960 --> 00:22:43,960
just like a home brewed version 
of 0 instead of like your your 

494
00:22:43,960 --> 00:22:46,000
company? 
Right now this, this is a really

495
00:22:46,000 --> 00:22:47,760
good question. 
And Can you imagine the models 

496
00:22:47,760 --> 00:22:50,600
get so good that you don't need 
the accounting software at all, 

497
00:22:50,800 --> 00:22:52,120
right? 
And like, could the model just 

498
00:22:52,120 --> 00:22:53,880
you, you give it your data and 
it does your books? 

499
00:22:54,880 --> 00:22:57,520
I, I don't think that is the 
future you want because 

500
00:22:57,520 --> 00:22:59,360
ultimately this is still a 
workflow. 

501
00:22:59,360 --> 00:23:01,880
It's a real business process 
that you need everyone involved 

502
00:23:01,880 --> 00:23:03,640
with. 
You need user accounts, 

503
00:23:03,640 --> 00:23:06,840
permissioning, notification 
sharing, like you need an actual

504
00:23:06,840 --> 00:23:09,320
experience. 
And so even to the extent we 

505
00:23:09,320 --> 00:23:12,400
invest in the AII, do think it 
comes down to you also need to 

506
00:23:12,400 --> 00:23:14,440
build better accounting 
software, right? 

507
00:23:14,440 --> 00:23:17,680
And it's the same thing I would 
say with Figma, it's like, OK, I

508
00:23:17,680 --> 00:23:19,040
don't think Figma is 
disappearing. 

509
00:23:19,040 --> 00:23:20,520
It can help you. 
It can automate a lot of the 

510
00:23:20,520 --> 00:23:22,720
stuff, but you still ultimately 
want to canvas your team can 

511
00:23:22,720 --> 00:23:25,560
collaborate around. 
And that's how we you finance. 

512
00:23:25,560 --> 00:23:28,200
Like if you could picture a 
Figma for finance, what would it

513
00:23:28,200 --> 00:23:30,360
be? 
So we try to look at it as a 

514
00:23:30,360 --> 00:23:34,040
broader product perspective 
where the AI is a core tool, but

515
00:23:34,040 --> 00:23:36,240
it's one piece of the entire 
vision. 

516
00:23:36,480 --> 00:23:39,360
Just like if you were to build 
like, oh, Redis came out now, do

517
00:23:39,360 --> 00:23:41,600
you not need a whole other like 
class of products? 

518
00:23:42,240 --> 00:23:44,920
You sort of still do like Redis 
is great technology, but you 

519
00:23:44,920 --> 00:23:46,600
still need the actual experience
around it. 

520
00:23:46,720 --> 00:23:48,280
Do. 
You think we're close to you 

521
00:23:48,280 --> 00:23:52,200
having the problem of someone 
gets an accountant who that 

522
00:23:52,200 --> 00:23:54,640
accountant is using language 
models that are trying to 

523
00:23:54,640 --> 00:23:57,040
interact with your software and 
then you're trying to figure out

524
00:23:57,400 --> 00:24:00,560
how much to serve them or have 
you started to see that or are 

525
00:24:00,560 --> 00:24:02,680
we too early? 
We're we're a little too early. 

526
00:24:02,680 --> 00:24:04,840
The accountants aren't that 
quite They're not only. 

527
00:24:04,880 --> 00:24:08,920
Adopters. 
But we we think deeply about the

528
00:24:08,920 --> 00:24:10,880
collaborative flows like the 
accountant is in the picture. 

529
00:24:10,880 --> 00:24:13,920
And this is actually an 
interesting point for folks as 

530
00:24:13,920 --> 00:24:17,960
we talk with business owners. 
They do not want AI accounting, 

531
00:24:18,000 --> 00:24:21,960
IE the the AI does 100%. 
They actually want the AI to do 

532
00:24:21,960 --> 00:24:23,960
most of it. 
And they want their books 

533
00:24:23,960 --> 00:24:27,920
blessed by a real CPA. 
Because you can't sue the AI, 

534
00:24:28,400 --> 00:24:29,840
right? 
You need someone to I. 

535
00:24:29,960 --> 00:24:33,480
Need a human I can blame? 
And so that is actually really, 

536
00:24:33,480 --> 00:24:35,720
really important and that's why 
the industry is not going 

537
00:24:35,720 --> 00:24:37,680
anywhere, right? 
They will up level and be able 

538
00:24:37,680 --> 00:24:40,440
to serve more clients. 
But you still need that human to

539
00:24:40,440 --> 00:24:43,760
have blessed the books. 
Gabe, the rapper question, I 

540
00:24:43,760 --> 00:24:45,880
mean, we sort of in the 
presentation earlier today, sort

541
00:24:45,880 --> 00:24:48,200
of, you know, we've, we've moved
to a point where it's like, OK 

542
00:24:48,200 --> 00:24:49,920
to be a rapper. 
We don't use that term because 

543
00:24:49,920 --> 00:24:53,120
it's pejorative. 
Like where are I, I, you know, I

544
00:24:53,120 --> 00:24:55,200
love to use pejorative words 
positively. 

545
00:24:55,200 --> 00:24:58,480
Like I've been on the case that 
it's like, no, you differentiate

546
00:24:58,480 --> 00:25:01,320
on the product and software, not
necessarily like on the model. 

547
00:25:01,600 --> 00:25:04,200
How would you calibrate that in 
terms of how much you need to 

548
00:25:04,200 --> 00:25:08,080
have at some point, your own 
models, your own thinking, or 

549
00:25:08,240 --> 00:25:11,400
we're going to be the best at 
directing chachi tea and 

550
00:25:11,400 --> 00:25:13,960
anthropic and everybody towards 
our use case. 

551
00:25:14,560 --> 00:25:17,040
Look, it kind of goes back to 
how Jeff answered the question, 

552
00:25:17,040 --> 00:25:20,040
which is even if you have a 
profoundly intelligent model, 

553
00:25:20,240 --> 00:25:22,960
you still need the tools to, you
know, do the accounting workflow

554
00:25:22,960 --> 00:25:24,080
right. 
It's like even if you have a 

555
00:25:24,080 --> 00:25:26,560
robot that can move the way 
human hand, they still need a 

556
00:25:26,560 --> 00:25:30,120
drill to go, you know, install 
whatever they're installing for 

557
00:25:30,120 --> 00:25:31,440
us. 
We're building the tools that 

558
00:25:31,440 --> 00:25:35,000
some eventual agent intelligence
either owned by us or someone 

559
00:25:35,000 --> 00:25:38,600
else might use, Whether that's 
an Excel interface for an agent,

560
00:25:38,600 --> 00:25:40,560
whether that's a PowerPoint 
interface, whether that's 

561
00:25:40,720 --> 00:25:43,200
structured financial data 
querying or unstructured 

562
00:25:43,200 --> 00:25:46,360
financial data querying or 
integrations into the internal 

563
00:25:46,360 --> 00:25:48,840
data systems of the banks and 
private equity firms we work 

564
00:25:48,840 --> 00:25:51,600
with. 
That said, I think it's very 

565
00:25:51,600 --> 00:25:54,880
clear that reinforcement 
learning is working and POST 

566
00:25:54,880 --> 00:25:57,360
training is working on top of 
these Frontier models. 

567
00:25:57,640 --> 00:26:01,800
And with very great specific 
evals you can actually 

568
00:26:01,800 --> 00:26:05,080
outperform the Frontier. 
You with post training alone, 

569
00:26:05,080 --> 00:26:07,840
you can find a lot of 
technological differentiation 

570
00:26:07,960 --> 00:26:09,320
that that makes it more 
valuable. 

571
00:26:09,320 --> 00:26:11,400
Yeah, I I mean, I think. 
I think that's what early 

572
00:26:11,400 --> 00:26:13,320
results are showing. 
I don't know if there's, I mean,

573
00:26:13,320 --> 00:26:16,120
actually cognition had a great 
paper that they, I mean, not 

574
00:26:16,120 --> 00:26:17,600
paper. 
It was a Twitter thread, but it 

575
00:26:17,640 --> 00:26:19,200
but it was a great Twitter 
thread that. 

576
00:26:19,200 --> 00:26:20,880
Sums up something in the current
it was a. 

577
00:26:20,880 --> 00:26:24,080
Great Twitter thread. 
Like maybe a week ago on they 

578
00:26:24,080 --> 00:26:27,600
had, they had post trained a 
model with RFT for, you know, 

579
00:26:27,600 --> 00:26:30,760
writing like CUDA kernels or 
something that that's quite 

580
00:26:30,760 --> 00:26:32,560
difficult for O3. 
And they got a lot of 

581
00:26:32,560 --> 00:26:36,000
outperformance with the types of
technologies and and 

582
00:26:36,000 --> 00:26:37,800
reinforcement techniques that 
are now diffusing through 

583
00:26:37,800 --> 00:26:39,320
industry. 
And so I think we'll start to 

584
00:26:39,320 --> 00:26:41,840
see a lot more of that. 
It reminds me of when folks used

585
00:26:41,840 --> 00:26:45,000
to overclock their CPUs and now 
CPUs are fast enough where no 

586
00:26:45,000 --> 00:26:47,960
one really bothers as much. 
Yeah, No, no, that's, I mean, 

587
00:26:47,960 --> 00:26:50,400
it's totally possible. 
I mean, I think the way the way 

588
00:26:50,400 --> 00:26:53,800
that I think about RFT and post 
training is not necessarily that

589
00:26:53,800 --> 00:26:56,440
it's going to make the model 
smarter, but it's going to sort 

590
00:26:56,440 --> 00:26:59,760
of prune the decision tree and 
make it better at using your 

591
00:26:59,760 --> 00:27:02,040
tools, right. 
So like opening eyes never going

592
00:27:02,040 --> 00:27:04,000
to have your tool structure and 
it's training data unless you 

593
00:27:04,000 --> 00:27:06,200
collaborate with that. 
And so the same way if you get a

594
00:27:06,200 --> 00:27:08,480
super smart human starting at 
your firm on Monday, they're 

595
00:27:08,720 --> 00:27:11,560
going to need to learn how to 
use QuickBooks and zero and I 

596
00:27:11,560 --> 00:27:14,600
hope digits, you know the model 
you want to teach it how to use 

597
00:27:14,600 --> 00:27:17,720
those tools to Yep. 
Jeff, last question, last word. 

598
00:27:18,720 --> 00:27:22,760
Would you tell a founder today 
to build a services business 

599
00:27:22,760 --> 00:27:26,680
powered by you've done the sort 
of flirtation with having the 

600
00:27:26,680 --> 00:27:28,920
accounting firm? 
Do you buy that investor 

601
00:27:28,920 --> 00:27:30,720
narrative or do you think it's 
oversold? 

602
00:27:30,920 --> 00:27:33,760
I think it's oversold and it's 
so tough because it is such a 

603
00:27:33,760 --> 00:27:35,480
tempting narrative. 
It's build your services 

604
00:27:35,480 --> 00:27:37,920
business, look at what's slow, 
automate that, move on, right? 

605
00:27:37,920 --> 00:27:39,320
You can drive a high margin 
business. 

606
00:27:39,600 --> 00:27:43,440
The change management, the 
actual people management is so 

607
00:27:43,440 --> 00:27:46,200
challenging because it depends 
on which profession you're 

608
00:27:46,200 --> 00:27:49,080
trying to automate. 
But there's so much built in 

609
00:27:49,280 --> 00:27:52,160
just like native practices in 
that profession that in order to

610
00:27:52,160 --> 00:27:55,600
really automate everything and 
build the trust, it's a long 

611
00:27:55,600 --> 00:27:57,280
journey to get to that margin 
versus. 

612
00:27:57,280 --> 00:27:59,560
Just turns out people outside 
Silicon Valley, they know their 

613
00:27:59,560 --> 00:28:01,400
businesses too. 
They're hard businesses. 

614
00:28:01,400 --> 00:28:03,440
Exactly right. 
And again with accounting, 

615
00:28:03,440 --> 00:28:06,360
accounting is complicated like. 
There is no way to simplify 

616
00:28:06,360 --> 00:28:06,960
that. 
Great. 

617
00:28:07,040 --> 00:28:13,480
Thank you very much. 
Thank you, Schuber say, about 

618
00:28:13,480 --> 00:28:18,640
this conversation, two things. 
One, Gusto's the rare startup 

619
00:28:18,640 --> 00:28:21,400
where I sort of have a rooting 
interest in that we are a small 

620
00:28:21,400 --> 00:28:25,160
business, and every time Gusto 
solves a problem, I am relieved 

621
00:28:25,160 --> 00:28:26,680
that you get to take it off my 
plate. 

622
00:28:27,240 --> 00:28:29,480
The second thing is it's funny 
to be hanging out with you with 

623
00:28:29,480 --> 00:28:32,720
shoes on, you know, do you guys 
still have you, have you gotten 

624
00:28:32,720 --> 00:28:35,200
rid of your no shoes policy or 
are you still I? 

625
00:28:35,200 --> 00:28:37,080
Feel like I have to give context
to people here. 

626
00:28:37,080 --> 00:28:39,960
So excited to be here. 
This is a fun space for me 

627
00:28:39,960 --> 00:28:43,520
because we had a Gusto holiday 
party here in 2015, but our 

628
00:28:43,640 --> 00:28:47,000
three first offices were only a 
block away and we were raised by

629
00:28:47,000 --> 00:28:48,880
our parents to take our shoes 
off at home. 

630
00:28:49,320 --> 00:28:52,640
So for many years our offices 
were shoes off, walking around 

631
00:28:52,640 --> 00:28:56,320
in slippers, socks. 
We passed them out to folks and 

632
00:28:56,320 --> 00:28:58,360
then when the pandemic started, 
obviously people can do whatever

633
00:28:58,360 --> 00:29:00,960
they want at their home. 
Today, it's more of an optional 

634
00:29:00,960 --> 00:29:02,800
policy. 
But I do describe it as a 

635
00:29:02,800 --> 00:29:05,200
tradition that that felt right 
for gusto at the time. 

636
00:29:05,800 --> 00:29:08,600
You know, yeah, in some ways 
it's like you've become this 

637
00:29:08,640 --> 00:29:12,160
grown up company. 
Can you give, you know, 

638
00:29:12,160 --> 00:29:15,000
obviously payroll is at the 
heart of it, but talk through 

639
00:29:15,000 --> 00:29:18,600
just briefly what the pieces of 
Gusto's businesses are today in 

640
00:29:18,600 --> 00:29:21,920
terms of the focus and where 
you're really strong? 

641
00:29:22,240 --> 00:29:25,400
So something really important to
know about Gusto is we obsess 

642
00:29:25,400 --> 00:29:27,640
over a small business. 
So if we have any customers 

643
00:29:27,640 --> 00:29:29,240
here, we're honored to serve 
you. 

644
00:29:29,680 --> 00:29:33,400
And just know that there are 
more dentist offices in the US 

645
00:29:33,400 --> 00:29:36,680
than tech startups. 
And so we really focus on 

646
00:29:36,680 --> 00:29:40,520
mainstream small business of the
six million ish employers in 

647
00:29:40,520 --> 00:29:43,200
America, 2/3 or less than 5 
employees. 

648
00:29:43,680 --> 00:29:47,120
And so with that customer focus,
payroll is our first product. 

649
00:29:47,880 --> 00:29:50,600
I always like to say if you 
don't pay someone, they quit. 

650
00:29:50,960 --> 00:29:53,560
So it's a pretty non optional 
product. 

651
00:29:54,120 --> 00:29:57,880
So we're really happy we started
there, but kind of broadening 

652
00:29:57,880 --> 00:30:00,000
through the back office would be
the way to describe it. 

653
00:30:00,000 --> 00:30:04,960
We really obsess over customer 
pull versus company push, but 

654
00:30:05,160 --> 00:30:06,920
benefits is a big investment 
area for us. 

655
00:30:06,920 --> 00:30:09,920
There's many, many types of 
benefits, something I know we're

656
00:30:09,920 --> 00:30:12,640
excited to get into more. 
We have a pretty broad product 

657
00:30:12,640 --> 00:30:16,120
group inside Gusto called Gusto 
Money that's focused on things 

658
00:30:16,120 --> 00:30:18,960
related to cash flow management.
For a lot of our customers, 

659
00:30:18,960 --> 00:30:21,840
their biggest expense is 
payroll, paying their team. 

660
00:30:22,520 --> 00:30:26,200
And one of the biggest points of
stress is when APAR don't line 

661
00:30:26,200 --> 00:30:28,280
up and they don't have the money
in their bank account at the 

662
00:30:28,280 --> 00:30:31,680
right moment to fund payroll. 
So we launched a bill payment 

663
00:30:32,520 --> 00:30:34,640
not too long ago. 
Actually next Wednesday, we're 

664
00:30:34,640 --> 00:30:37,600
going live with invoicing. 
So you'll see us do a lot more 

665
00:30:37,600 --> 00:30:40,600
product investments, more in 
that B2B fintech part, but 

666
00:30:40,600 --> 00:30:42,640
again, really, really focused on
small business. 

667
00:30:43,120 --> 00:30:46,720
And overall business sort of 
somewhere between 500 million 

668
00:30:46,720 --> 00:30:48,800
and a billion in revenue. 
Can you? 

669
00:30:48,800 --> 00:30:51,480
Describe it that way. 
We serve over 400,000 companies.

670
00:30:51,480 --> 00:30:56,120
I mean, one fun way we always 
grounded in customer we're you 

671
00:30:56,120 --> 00:30:59,480
know, in the range of, you know,
8-9 percent now of all employers

672
00:30:59,480 --> 00:31:02,400
in America. 
And with the business model that

673
00:31:02,400 --> 00:31:04,880
leads to, you know, good 
revenue, we've been free cash 

674
00:31:04,880 --> 00:31:07,800
flow positive for many years. 
We reinvest that money back into

675
00:31:07,800 --> 00:31:10,600
building new product to solve 
more pain for our customer. 

676
00:31:11,000 --> 00:31:15,400
And I mean, you're an important 
sort of partner to other fintech

677
00:31:15,400 --> 00:31:16,080
companies. 
I know. 

678
00:31:16,080 --> 00:31:18,920
I think the guidelines CEO was 
here at one point. 

679
00:31:19,080 --> 00:31:22,760
What what do you see in terms of
like the partnership strategy 

680
00:31:22,840 --> 00:31:24,520
for Gusso? 
Yeah. 

681
00:31:24,520 --> 00:31:27,040
So a big part of our approach to
we're here to build a multi 

682
00:31:27,040 --> 00:31:30,160
decade company, we're still 
early in the journey even though

683
00:31:30,160 --> 00:31:33,320
it's, you know, over 10 years. 
And that to me is a factual 

684
00:31:33,320 --> 00:31:35,280
statement, right? 
We're only at about 89% of 

685
00:31:35,280 --> 00:31:37,520
employers. 
You still have 30% of companies 

686
00:31:37,520 --> 00:31:40,560
in the US doing even things like
payroll by hand on pen and 

687
00:31:40,560 --> 00:31:42,800
paper. 
So with the obsession on small 

688
00:31:42,800 --> 00:31:45,400
business, a lot of what we're 
doing is replacing manual 

689
00:31:45,400 --> 00:31:48,360
process, but there's just a lot 
of other pain points we want to 

690
00:31:48,360 --> 00:31:50,400
help with. 
And if we're going to get there,

691
00:31:50,400 --> 00:31:52,560
it's not going to be entirely 
through first party products. 

692
00:31:52,560 --> 00:31:55,720
So we've had a playbook, which 
we will continue to use of 

693
00:31:55,720 --> 00:31:58,320
choosing product by product. 
Are we going to build? 

694
00:31:58,680 --> 00:32:01,760
Are we going to partner? 
4O1K is a good example of where 

695
00:32:01,760 --> 00:32:04,720
we've partnered historically. 
And then there's also obviously 

696
00:32:04,720 --> 00:32:06,160
acquisition. 
We've done that a few times. 

697
00:32:06,160 --> 00:32:09,720
We'll do that more going forward
for where we deem something that

698
00:32:09,720 --> 00:32:11,400
was third party becoming first 
party. 

699
00:32:11,800 --> 00:32:14,000
But you know, we're not going to
build it all ourselves, and 

700
00:32:14,000 --> 00:32:15,480
we're excited to partner where 
it makes sense. 

701
00:32:16,480 --> 00:32:20,840
Has artificial intelligence 
changed the pace of product 

702
00:32:20,840 --> 00:32:22,560
development? 
I mean, you're, you're an 

703
00:32:22,560 --> 00:32:26,280
interesting case in that on the 
one end, it's sort of the, you 

704
00:32:26,280 --> 00:32:28,080
know, I don't know, Klarna's the
world where they're like, well, 

705
00:32:28,120 --> 00:32:29,520
we'll build everything 
internally. 

706
00:32:29,520 --> 00:32:32,440
Like, you know, payroll could be
imagined that way. 

707
00:32:32,680 --> 00:32:34,800
On the other hand, it can 
accelerate your growth. 

708
00:32:34,800 --> 00:32:37,920
Like how do you, how are you 
seeing sort of artificial 

709
00:32:37,920 --> 00:32:41,120
intelligence, the ability to 
build faster impacting gusto 

710
00:32:41,120 --> 00:32:43,000
today, I mean? 
I'm surprised it took this long 

711
00:32:43,000 --> 00:32:46,920
to get to AI as a topic. 
I mean there's so many threads. 

712
00:32:46,920 --> 00:32:48,640
I think a lot of folks here are 
technologists. 

713
00:32:48,640 --> 00:32:52,440
I'll have to just choose like on
productivity and how we build. 

714
00:32:52,440 --> 00:32:54,720
We have 1000 person plus R&D 
team. 

715
00:32:54,720 --> 00:32:56,760
I don't think we're that 
different than most software 

716
00:32:56,760 --> 00:33:00,000
companies in that it's obviously
driving pretty meaningful shift 

717
00:33:00,000 --> 00:33:03,160
in how we build. 
You know, the iteration speed. 

718
00:33:03,160 --> 00:33:05,640
I think we're going to talk 
later about, you know, there's 

719
00:33:05,640 --> 00:33:09,320
the service as a software, 
software as a service, everyone 

720
00:33:09,320 --> 00:33:11,880
gets SAS. 
A lot of what we do is taking 

721
00:33:11,880 --> 00:33:16,120
stuff that involves compliance, 
involves manual filings and 

722
00:33:16,120 --> 00:33:19,760
digitizing it into software. 
In the past I said we use 

723
00:33:20,080 --> 00:33:22,720
paperless cloud mobile. 
Now obviously AI is a key 

724
00:33:22,720 --> 00:33:25,760
ingredient to that, but that's 
on the more internal how we 

725
00:33:25,760 --> 00:33:28,480
build side on the. 
Doesn't mean fewer engineers 

726
00:33:28,480 --> 00:33:31,760
just to put a fine line on it. 
Yeah, doesn't mean fewer 

727
00:33:31,760 --> 00:33:34,120
engineers. 
Given the scope of our ambition.

728
00:33:34,120 --> 00:33:38,680
It means like the amount of work
we can do per gusty grows 

729
00:33:38,680 --> 00:33:41,600
dramatically. 
But probably in the grand scheme

730
00:33:41,600 --> 00:33:44,880
of things, it means less hiring 
than if AI didn't exist. 

731
00:33:45,440 --> 00:33:48,040
But we're also pretty ambitious,
so we're doing a lot of hiring 

732
00:33:48,040 --> 00:33:50,120
anyways. 
But it's because we want to go 

733
00:33:50,120 --> 00:33:51,560
solve more pain point for our 
customers. 

734
00:33:51,600 --> 00:33:56,440
And then on this sort of, do you
feel the threat of customers 

735
00:33:56,440 --> 00:33:59,160
building in house versions of 
Gus? 

736
00:33:59,760 --> 00:34:01,880
Honestly, as a small business 
that's sort of incoherent, 

737
00:34:01,880 --> 00:34:03,600
right? 
Because it's like I'm, I'm not 

738
00:34:03,600 --> 00:34:05,960
going to build an AI team as a 
five person startup. 

739
00:34:06,480 --> 00:34:09,639
It would matter more to your 
enterprise level competitors or 

740
00:34:09,639 --> 00:34:11,280
how do you think about that 
challenge I. 

741
00:34:11,280 --> 00:34:14,239
Mean, I don't think too many 
dentists want to go, right, 

742
00:34:14,239 --> 00:34:17,560
build software. 
That said, the reality is most 

743
00:34:17,560 --> 00:34:20,080
of what we're doing for a 
dentist office or other small 

744
00:34:20,080 --> 00:34:23,000
businesses they haven't had 
access to in the past. 

745
00:34:23,000 --> 00:34:24,639
They've been on their own. 
They've kind of had to do it 

746
00:34:25,239 --> 00:34:29,480
manually or just not do it. 
And so for us, AI is enabling 

747
00:34:29,480 --> 00:34:32,760
products to exist in a small 
business category that never 

748
00:34:32,760 --> 00:34:34,480
existed. 
I think in mid market 

749
00:34:34,480 --> 00:34:36,719
enterprise, you're navigating 
more of this complexity of 

750
00:34:36,960 --> 00:34:40,800
displacing jobs. 
But our due N is to like grow 

751
00:34:40,800 --> 00:34:42,600
the small business economy, 
right? 

752
00:34:42,600 --> 00:34:45,840
And so we want to bring all of 
the things big companies have 

753
00:34:45,840 --> 00:34:48,280
had historically to a small 
company. 

754
00:34:48,480 --> 00:34:51,719
The only way we're going to do 
that right with a ratio that 

755
00:34:51,719 --> 00:34:54,320
makes sense and with a 
scalability that makes sense is 

756
00:34:54,320 --> 00:34:55,520
through leveraging technology. 
I. 

757
00:34:55,560 --> 00:34:58,520
Mean, if I had to distill the 
angst of a small business, it's 

758
00:34:58,520 --> 00:35:01,400
like I want to know that I'm in 
good standing and I don't know 

759
00:35:01,400 --> 00:35:04,000
everything I'm supposed to do. 
And you guys are, but in the 

760
00:35:04,000 --> 00:35:06,040
products, sometimes we like, you
know, in New York, you're 

761
00:35:06,040 --> 00:35:08,120
supposed to have sexual 
harassment training. 

762
00:35:08,120 --> 00:35:09,600
OK, I'll pay you some money for 
that. 

763
00:35:09,600 --> 00:35:12,680
Like, do you think AI is going 
to move us further along of just

764
00:35:12,680 --> 00:35:14,680
being able to give me an 
assessment? 

765
00:35:14,680 --> 00:35:16,760
Like, have you done everything 
you're supposed to do? 

766
00:35:17,280 --> 00:35:19,240
My combat fee, am I Good 
question. 

767
00:35:19,680 --> 00:35:24,040
And like our customers, 
including you, do not want to be

768
00:35:24,040 --> 00:35:27,320
stressed or worrying about that.
And it's our job to make sure 

769
00:35:27,320 --> 00:35:30,240
you know fully if you're good or
if there's something you need to

770
00:35:30,240 --> 00:35:33,480
do what that exact thing is. 
And if we can do it for you, 

771
00:35:33,480 --> 00:35:35,920
even better. 
But yeah, compliance maps to 

772
00:35:35,920 --> 00:35:37,840
most of our products, right? 
Payroll involves a lot of 

773
00:35:37,840 --> 00:35:41,280
compliance time tracking, PTO 
involves a lot of compliance, 

774
00:35:41,280 --> 00:35:45,400
local, state, federal rules. 
And so AI fits in there in terms

775
00:35:45,400 --> 00:35:49,000
of like ingesting. 
But at the end of the day, it's 

776
00:35:49,000 --> 00:35:52,000
a compliance engine where 
accuracy needs to be to the 

777
00:35:52,000 --> 00:35:56,560
like, you know, 6 Sigma. 
And that's always been true pre 

778
00:35:56,560 --> 00:35:59,240
AI, post AI, that's been true 
since we started the company. 

779
00:35:59,560 --> 00:36:02,080
How much do you think your 
customers want to talk to the 

780
00:36:02,080 --> 00:36:05,480
product? 
Like, do you see a move to text 

781
00:36:05,920 --> 00:36:09,640
interfaces now that they're more
capable or you're like, we 

782
00:36:09,640 --> 00:36:12,880
really need to flow whatever 
advances we have into actually 

783
00:36:12,880 --> 00:36:14,760
the layout and design of the 
product? 

784
00:36:15,000 --> 00:36:17,960
So we always start with like 
small businesses are busy and 

785
00:36:17,960 --> 00:36:20,120
it's way too hard to run and 
build a small business. 

786
00:36:20,120 --> 00:36:23,280
Still, you can attest to that. 
And so how can we help them? 

787
00:36:23,280 --> 00:36:26,720
We can help them by saving time.
We can help them by taking 5 or 

788
00:36:26,720 --> 00:36:30,280
20 or 30 of the hats on their 
head off their head and doing it

789
00:36:30,280 --> 00:36:33,000
for them. 
Pre AI, I would say we did that 

790
00:36:33,000 --> 00:36:37,760
through really clean, elegant, 
easy to use work flows primarily

791
00:36:37,760 --> 00:36:40,560
in a web app or a mobile app. 
That's gone us pretty far, 

792
00:36:40,560 --> 00:36:41,880
right? 
Like that's what we're known 

793
00:36:41,880 --> 00:36:43,800
for. 
Anyone can use Gusto with no 

794
00:36:43,800 --> 00:36:45,680
training, no background in 
running a business. 

795
00:36:46,040 --> 00:36:48,160
We have high NPS, high customer 
satisfaction. 

796
00:36:48,160 --> 00:36:51,520
Our primary way of growing is 
word of mouth with AI. 

797
00:36:52,080 --> 00:36:54,040
It's not about the technology, 
the interface. 

798
00:36:54,040 --> 00:36:55,280
That's what we get excited 
about. 

799
00:36:55,680 --> 00:37:00,160
A conversational interface for a
lot of use cases that Gusto does

800
00:37:00,600 --> 00:37:04,320
is a more intuitive, more 
accessible way to use Gusto. 

801
00:37:04,600 --> 00:37:05,440
You like it? 
You think it's? 

802
00:37:05,480 --> 00:37:07,440
Promising. 
We think it's very promising. 

803
00:37:07,440 --> 00:37:11,000
Our interface for that is called
Gus, which people should 

804
00:37:11,000 --> 00:37:12,760
hopefully get. 
I think it's somewhat like this.

805
00:37:12,840 --> 00:37:14,520
And then, yeah, well, Penny the 
pig is. 

806
00:37:14,560 --> 00:37:16,160
Oh, that's a. 
Different logo OK, but. 

807
00:37:16,160 --> 00:37:19,160
Then like you're like cracking 
Penny in half, but like Penny's 

808
00:37:19,160 --> 00:37:20,800
being sacrificed for the sake of
your conference. 

809
00:37:21,920 --> 00:37:24,320
But yeah, conversational 
interface we don't think 

810
00:37:24,320 --> 00:37:26,600
replaces web app, replaces 
mobile app. 

811
00:37:26,600 --> 00:37:28,040
It's just a different surface 
area. 

812
00:37:28,360 --> 00:37:32,560
But like we are taking 14 years 
of functionality and giving Gus 

813
00:37:32,760 --> 00:37:35,200
those superpowers. 
So we have customers today, 

814
00:37:35,200 --> 00:37:38,240
thousands of customers using Gus
to go create shift schedules, 

815
00:37:38,520 --> 00:37:40,320
right? 
You want to go change who's 

816
00:37:40,320 --> 00:37:43,120
working with today? 
You know, take Sally off Friday,

817
00:37:43,120 --> 00:37:46,400
add 20 hours to gym. 
We go do that all for you. 

818
00:37:46,880 --> 00:37:49,560
You can just tell us that in a 
conversational interface, or you

819
00:37:49,560 --> 00:37:53,040
can go through and navigate, but
that saves time for a small 

820
00:37:53,040 --> 00:37:54,440
business, which they really 
appreciate. 

821
00:37:54,800 --> 00:37:58,280
I wanted to put one of the big 
themes of today to you. 

822
00:37:58,360 --> 00:38:02,600
I mean, we saw Matt Harris give 
a presentation where he's making

823
00:38:02,600 --> 00:38:05,040
sort of a case that other 
investors have also made that 

824
00:38:05,040 --> 00:38:08,920
it's like there's now, thanks to
language models, this 

825
00:38:08,920 --> 00:38:12,760
opportunity to go after services
businesses and sort of be the 

826
00:38:12,760 --> 00:38:15,640
services business while being a 
tech company. 

827
00:38:15,640 --> 00:38:19,080
Then later I had Jeff, the CEO 
of Digits, who's building 

828
00:38:19,080 --> 00:38:21,400
accounting software. 
He's, you know, hired 

829
00:38:21,400 --> 00:38:23,120
accountants to figure it out and
then software. 

830
00:38:23,120 --> 00:38:26,720
And I think he was much more 
skeptical of that sort of, you 

831
00:38:26,720 --> 00:38:30,160
can be a services tech business.
Where where do you land? 

832
00:38:30,160 --> 00:38:34,240
You've built a big business and 
you touch a lot of services. 

833
00:38:34,240 --> 00:38:36,640
Like are you bullish on that 
argument? 

834
00:38:37,520 --> 00:38:41,240
So I think we've had a kind of 
unique perspective to this 

835
00:38:41,240 --> 00:38:46,120
because the first product we 
launched payroll mostly if you 

836
00:38:46,120 --> 00:38:48,800
put aside like ADP paychecks, 
which are a minority of the 

837
00:38:48,800 --> 00:38:51,840
market, the majority of 
companies in America did payroll

838
00:38:51,840 --> 00:38:55,200
by hand or with a local payroll 
service Bureau, which was 

839
00:38:55,200 --> 00:38:59,440
basically a service manual 
localized business solution. 

840
00:38:59,440 --> 00:39:03,080
So a lot of our journey has been
taking and did it with health 

841
00:39:03,080 --> 00:39:04,640
benefits, right? 
We are a broker. 

842
00:39:04,960 --> 00:39:07,600
There are thousands of brokerage
firms across the country that 

843
00:39:07,600 --> 00:39:10,880
mostly do things by hand. 
So we take these complex 

844
00:39:10,880 --> 00:39:15,120
compliance centric spaces and we
do what we do best, which is we 

845
00:39:15,120 --> 00:39:17,800
digitize the heck out of it 
because if we're going to go do 

846
00:39:17,800 --> 00:39:22,000
it for 400,500 thousand, a 
million businesses, we cannot do

847
00:39:22,000 --> 00:39:25,960
it ourselves manually. 
So I don't think it's a zero a 

848
00:39:25,960 --> 00:39:29,440
hundred, a hundred zero. 
Like we have operations teams at

849
00:39:29,440 --> 00:39:32,120
Gusto, but we have over 400,000 
companies today. 

850
00:39:32,440 --> 00:39:36,720
Our Gusty to employee or ratio 
is over 1 to 1000. 

851
00:39:36,960 --> 00:39:38,640
That's only possible with the 
technology. 

852
00:39:38,840 --> 00:39:40,320
Right. 
So you only want to do it if you

853
00:39:40,320 --> 00:39:44,120
can get to the point where it's 
fairly automated, not sort of 

854
00:39:44,120 --> 00:39:47,440
for a long time, keep the humans
sort of trying to figure it out 

855
00:39:47,440 --> 00:39:50,560
or yeah, how long do you spend 
sort of like we'll do the human 

856
00:39:50,560 --> 00:39:53,200
business to try and get to the 
point where we can automate it 

857
00:39:53,200 --> 00:39:55,760
before saying, oh, I guess this 
is sort of, at least with the 

858
00:39:55,760 --> 00:39:57,880
technology today, a perpetually 
human. 

859
00:39:57,880 --> 00:39:59,640
Business, well, we were joking 
before, right? 

860
00:39:59,640 --> 00:40:03,640
Like if a company is scaling a 
service offering to large, large

861
00:40:03,640 --> 00:40:07,000
volumes and it's staying service
based, that's to me not a tech 

862
00:40:07,000 --> 00:40:08,480
company, that's just a service 
business. 

863
00:40:08,760 --> 00:40:11,480
Different multiple profile, 
different overhead costs, 

864
00:40:11,480 --> 00:40:13,000
different operational 
complexity. 

865
00:40:13,880 --> 00:40:15,840
We work backwards from what's 
best for the customer. 

866
00:40:16,280 --> 00:40:18,800
For us to solve these pain 
points, it has to be technology,

867
00:40:18,800 --> 00:40:21,200
it has to be digital. 
It's just more accurate, right? 

868
00:40:21,200 --> 00:40:24,720
We process several $100 billion 
a year of payroll taxes. 

869
00:40:24,720 --> 00:40:27,800
If we were doing that manually, 
it would create human error 

870
00:40:27,840 --> 00:40:30,240
potential. 
It would be too time consuming. 

871
00:40:30,440 --> 00:40:33,000
So we digitize because it's a 
better experience. 

872
00:40:33,280 --> 00:40:35,960
But when we're sub scale and 
trying out new products early, 

873
00:40:35,960 --> 00:40:39,280
early on, I think it's totally 
fine to start more manual. 

874
00:40:39,280 --> 00:40:41,280
It's about getting that feedback
loop going, getting that 

875
00:40:41,280 --> 00:40:43,400
learning going. 
But that's typically with less 

876
00:40:43,400 --> 00:40:46,400
than 100 customers, right? 
If we're going to scale 

877
00:40:46,400 --> 00:40:49,240
something, it better have good 
unit economics, good CAC, good 

878
00:40:49,240 --> 00:40:52,480
gross margin and we should have 
confidence that it can scale in 

879
00:40:52,520 --> 00:40:54,200
a, you know, technology enabled 
way. 

880
00:40:54,440 --> 00:40:56,640
Otherwise, we're just going to 
build a huge operations team. 

881
00:40:57,640 --> 00:41:02,120
Somehow in payroll, you're in 
like 1 of the wildest categories

882
00:41:02,320 --> 00:41:04,280
in Silicon Valley. 
You have, you know, like 

883
00:41:04,480 --> 00:41:08,880
Rippling and Deal, you know, 
making huge accusations, 

884
00:41:08,880 --> 00:41:11,280
obviously Ripley accusing Deal 
of spying on it. 

885
00:41:11,920 --> 00:41:15,680
Have you done an internal check 
to see whether anyone's spying 

886
00:41:15,680 --> 00:41:17,840
on you? 
And what do you make of sort of 

887
00:41:17,840 --> 00:41:24,400
your high profile, sort of same,
same category businesses having 

888
00:41:24,400 --> 00:41:28,680
so much drama? 
We we just focus on the 

889
00:41:28,680 --> 00:41:30,480
customer. 
Have you have you checked? 

890
00:41:30,480 --> 00:41:35,160
Have you checked? 
We have a very, very like 

891
00:41:35,560 --> 00:41:39,880
intense, like super, super 
expensive and worth IT security 

892
00:41:39,880 --> 00:41:42,840
team. 
Yeah, but that's because we have

893
00:41:43,360 --> 00:41:46,040
like hundreds of billions of 
dollars of our customer money 

894
00:41:46,040 --> 00:41:49,240
sitting in our accounts. 
And we, you know, do tax filings

895
00:41:49,240 --> 00:41:52,480
and tax payments and we hold 
Social Security numbers and a 

896
00:41:52,480 --> 00:41:55,680
whole bunch of very sensitive 
data that we take that job very 

897
00:41:55,680 --> 00:41:57,960
seriously. 
So I can tell you with 

898
00:41:57,960 --> 00:42:02,160
confidence, we do not believe 
anyone is spying on Gusto, but 

899
00:42:02,160 --> 00:42:04,520
that's because that's what our 
customers should demand of us, 

900
00:42:04,520 --> 00:42:06,440
that we're obsessed with 
protecting them in every 

901
00:42:06,440 --> 00:42:10,200
possible way. 
Do you see much competition or I

902
00:42:10,200 --> 00:42:13,880
mean, you're very focused on 
small businesses and I like that

903
00:42:13,880 --> 00:42:17,360
focus and it gives you a great 
mandate and sort of a mission to

904
00:42:17,360 --> 00:42:19,160
motivate the company. 
On the other hand, most 

905
00:42:19,160 --> 00:42:21,680
businesses want to keep their 
customers. 

906
00:42:21,720 --> 00:42:23,920
It's like it's much easier to 
keep the customer you have and 

907
00:42:23,920 --> 00:42:25,720
grow with them than to find new 
ones. 

908
00:42:25,720 --> 00:42:29,000
How do you think about sizing up
with your customers? 

909
00:42:29,600 --> 00:42:33,080
So we're both in a like very 
crowded space and a very 

910
00:42:33,080 --> 00:42:35,920
fragmented space. 
And then yes, I think it's even 

911
00:42:36,400 --> 00:42:39,520
very simplistic to say small 
business, medium sized business,

912
00:42:39,520 --> 00:42:41,840
large business, these are huge 
cuts. 

913
00:42:42,320 --> 00:42:44,680
And even within small business, 
you have different industry 

914
00:42:44,680 --> 00:42:48,880
types, different geographies. 
So, you know, hopefully everyone

915
00:42:48,880 --> 00:42:50,600
gets by now. 
We are obsessed with small 

916
00:42:50,600 --> 00:42:53,120
business. 
We tend to see they're less of 

917
00:42:53,120 --> 00:42:58,720
the kind of frothy Silicon 
Valley tech kind of drama, 

918
00:42:58,720 --> 00:43:00,680
frankly. 
But there's still legitimate 

919
00:43:00,680 --> 00:43:03,360
companies that have lots of good
product in market. 

920
00:43:03,360 --> 00:43:06,480
Intuit is a company we highly 
respect and we compete with 

921
00:43:06,640 --> 00:43:10,440
multiple fronts. 
And you know, there's a bunch of

922
00:43:10,440 --> 00:43:12,760
other companies I used to say 
was only into IT folks that have

923
00:43:12,760 --> 00:43:15,320
built really incredible 
franchises and small business. 

924
00:43:15,800 --> 00:43:18,160
You know, Shopify has a huge 
part of their company that's 

925
00:43:18,160 --> 00:43:21,640
focused there, Square, Toast, 
HubSpot. 

926
00:43:21,640 --> 00:43:25,760
So we feel like we're in good 
company and I always encourage 

927
00:43:25,760 --> 00:43:28,320
more entrepreneurs to tackle the
small business category. 

928
00:43:28,800 --> 00:43:32,000
The playbook still is generally 
for SAS just to move up market 

929
00:43:32,000 --> 00:43:35,400
and move to bigger enterprise. 
And in the past, I just meant 

930
00:43:35,400 --> 00:43:36,880
small business was left. 
Do you? 

931
00:43:36,880 --> 00:43:40,840
Want 100% business on Gusto? 
We have a lot of 100% business 

932
00:43:40,880 --> 00:43:43,760
on Gusto. 
Most of them started on Gusto 

933
00:43:43,760 --> 00:43:46,560
when they were one employee. 
So our focus is on small 

934
00:43:46,560 --> 00:43:48,880
business. 
The vast majority will stay 

935
00:43:48,880 --> 00:43:51,880
small if some grow bigger. 
We're honored and happy to serve

936
00:43:51,880 --> 00:43:54,040
them. 
There's probably a point at some

937
00:43:54,040 --> 00:43:57,520
point where they graduate. 
But I like to remind folks, you 

938
00:43:57,520 --> 00:44:00,200
know, that's very rare, right? 
Like the number of companies 

939
00:44:00,200 --> 00:44:02,920
that start at one or two and 
grow to 100 and our entire 

940
00:44:02,920 --> 00:44:07,440
history is less than .1%. 
Is a, you know, we all at the 

941
00:44:07,440 --> 00:44:10,800
end of the day, live or die by 
the economy in some ways, and 

942
00:44:10,800 --> 00:44:14,320
you have a great sort of vantage
into a part of it. 

943
00:44:14,800 --> 00:44:16,760
Is small business in America 
strong right now? 

944
00:44:16,760 --> 00:44:19,760
What's your read of the mood 
among your customers? 

945
00:44:20,440 --> 00:44:22,280
Yeah. 
So we have economists on team. 

946
00:44:22,560 --> 00:44:25,720
We publish a lot of content 
under Gustanomics and a lot of 

947
00:44:25,720 --> 00:44:26,960
folks are interested in the 
data. 

948
00:44:27,320 --> 00:44:31,600
I'd say high level two things we
track a lot new employer starts 

949
00:44:31,600 --> 00:44:34,200
and then net hiring across our 
customer base. 

950
00:44:34,680 --> 00:44:37,400
I'd say on the second that is 
quite depressed. 

951
00:44:37,560 --> 00:44:39,840
New hiring new. 
Hiring, net hiring. 

952
00:44:39,840 --> 00:44:42,840
So just the propensity of our 
customers to add more to their 

953
00:44:42,840 --> 00:44:46,400
team that's been depressed for a
few years now on the new 

954
00:44:46,400 --> 00:44:49,640
employer starts that really with
the pandemic got dramatically 

955
00:44:49,640 --> 00:44:51,960
elevated and it stayed fairly 
high. 

956
00:44:52,560 --> 00:44:55,640
But yeah, it's more interesting 
to track given our size and 

957
00:44:55,640 --> 00:44:57,040
scale. 
We feel like we're pretty 

958
00:44:57,040 --> 00:45:00,560
diversified. 
It's more just net, net a bet on

959
00:45:00,560 --> 00:45:02,880
small business broadly and it 
tends to be pretty durable 

960
00:45:02,880 --> 00:45:05,760
segment. 
Can you talk about the embedded 

961
00:45:06,240 --> 00:45:09,200
payroll part of your business? 
And sort of, I mean there is 

962
00:45:09,200 --> 00:45:12,280
this other trend of, you know, 
API based businesses right now 

963
00:45:12,280 --> 00:45:15,240
where it's like, OK, you're 
helping other people interact 

964
00:45:15,240 --> 00:45:17,880
with your service, but not 
necessarily building out the 

965
00:45:17,880 --> 00:45:20,560
product yourself. 
How do you think about that as a

966
00:45:20,560 --> 00:45:23,840
piece of Gusto's business? 
Yes, we're always going to be 

967
00:45:23,840 --> 00:45:27,360
driven by customer pull. 
So I'm very excited our direct 

968
00:45:27,360 --> 00:45:30,520
business will keep growing past,
you know half a million, a 

969
00:45:30,520 --> 00:45:32,440
million businesses in the coming
years. 

970
00:45:33,160 --> 00:45:36,200
But a couple years ago we 
noticed and maybe some of you 

971
00:45:36,200 --> 00:45:39,960
are part of this trend, but 
vertical SaaS in particular we 

972
00:45:39,960 --> 00:45:42,400
got really excited about where 
you have companies tackling a 

973
00:45:42,400 --> 00:45:46,200
very specific industry could be 
very esoteric, but if you obsess

974
00:45:46,200 --> 00:45:49,600
over that one category kind of 
build business in a box, you can

975
00:45:49,600 --> 00:45:52,000
actually create a viable tech 
company there. 

976
00:45:52,480 --> 00:45:55,320
And every one of these folks 
kept coming to us and saying, 

977
00:45:55,320 --> 00:45:56,880
you know, we don't want to build
payroll. 

978
00:45:57,520 --> 00:46:00,840
No one really wants to build 
payroll cuz it tends to be quite

979
00:46:00,840 --> 00:46:03,080
difficult. 
But we really want to provide 

980
00:46:03,080 --> 00:46:04,960
payroll to our customer. 
And we don't want to just keep 

981
00:46:04,960 --> 00:46:07,680
routing customers to you. 
We want to have it be native, a 

982
00:46:07,680 --> 00:46:10,400
native product experience. 
And so we have some great 

983
00:46:10,400 --> 00:46:14,880
reference points in embedded in 
Fintech, especially like Stripe 

984
00:46:15,000 --> 00:46:17,480
as a good example. 
And so that was where the 

985
00:46:17,480 --> 00:46:19,200
genesis of embedded payroll came
about. 

986
00:46:19,200 --> 00:46:21,280
That's what we call Gusto 
embedded payroll. 

987
00:46:21,720 --> 00:46:25,200
And we're really excited there. 
We have a number of partners on 

988
00:46:25,200 --> 00:46:27,600
the banking side. 
Chase payroll is powered by 

989
00:46:27,600 --> 00:46:30,000
Gusto. 
We recently announced U.S. bank,

990
00:46:30,000 --> 00:46:32,280
we'll be rolling out payroll 
powered by Gusto. 

991
00:46:32,800 --> 00:46:35,560
You know, 0 is another good 
example on the contact side. 

992
00:46:36,200 --> 00:46:39,560
And so anyone that does want to 
launch or offer a payroll 

993
00:46:39,560 --> 00:46:42,360
natively within your product but
doesn't want to build in from 

994
00:46:42,360 --> 00:46:45,080
scratch, please let us know. 
We're eager and excited to 

995
00:46:45,080 --> 00:46:47,320
partner with you. 
That whole business is more of 

996
00:46:47,320 --> 00:46:49,840
an infrastructure. 
Business, my last question, you 

997
00:46:49,840 --> 00:46:51,520
know, we have a lot of founders 
here. 

998
00:46:51,760 --> 00:46:54,440
You've been at this a while. 
Like what what? 

999
00:46:54,440 --> 00:46:57,480
What's your main piece of advice
to somebody starting a company 

1000
00:46:57,640 --> 00:46:59,400
today? 
Or what would you do differently

1001
00:46:59,400 --> 00:47:01,640
if you were starting Gusto right
now? 

1002
00:47:02,760 --> 00:47:05,760
So I'll answer both. 
My main advice I'm pretty 

1003
00:47:05,760 --> 00:47:07,520
consistent with, but I really 
believe it. 

1004
00:47:08,080 --> 00:47:10,760
It's imagine the 10 thousandth 
time you're describing what 

1005
00:47:10,760 --> 00:47:12,480
you're doing. 
Will you be as excited as the 

1006
00:47:12,480 --> 00:47:14,840
first time? 
Because at that point you can't 

1007
00:47:14,840 --> 00:47:19,320
fake it, It will show. 
And you have to have a deeper 

1008
00:47:19,320 --> 00:47:23,240
interest, passion, borderline 
obsession with the thing you're 

1009
00:47:23,240 --> 00:47:25,480
trying to fix, the problem 
you're trying to make better 

1010
00:47:26,360 --> 00:47:28,760
because that's what gets you 
through all the ups and downs of

1011
00:47:28,760 --> 00:47:31,800
company building. 
And in our case, you know, I was

1012
00:47:31,920 --> 00:47:34,640
with a team earlier today that's
an incubation team launching a 

1013
00:47:34,640 --> 00:47:36,560
new product. 
And, you know, we went around 

1014
00:47:36,560 --> 00:47:39,160
the room and it was an easy 
first question. 

1015
00:47:39,160 --> 00:47:41,280
I didn't even have to ask it. 
Everyone just shared their 

1016
00:47:41,280 --> 00:47:44,320
favorite small business. 
And so in our case, you know, 

1017
00:47:44,320 --> 00:47:46,160
that obsession with small 
business comes through hopefully

1018
00:47:46,160 --> 00:47:47,800
very clearly. 
It's a big part of our hiring 

1019
00:47:47,800 --> 00:47:49,800
filter. 
And as long as they're in pain, 

1020
00:47:49,800 --> 00:47:52,680
we feel like we have purpose and
we have so much to do that we 

1021
00:47:52,680 --> 00:47:54,000
feel like we're really still 
early. 

1022
00:47:54,320 --> 00:47:57,880
There's a lot, lot more ahead. 
And if I reflect on learnings 

1023
00:47:57,880 --> 00:48:00,480
from prior chapters, I had a 
prior start up where that wasn't

1024
00:48:00,480 --> 00:48:03,240
the case. 
And that's where, you know, you 

1025
00:48:03,240 --> 00:48:05,880
can, you can start a company or 
if you join a company, you feel 

1026
00:48:05,880 --> 00:48:08,960
like there's a miss there and 
wonder what's off right. 

1027
00:48:09,440 --> 00:48:12,000
And, and I would argue, you 
know, go back to like, what is 

1028
00:48:12,000 --> 00:48:14,440
the purpose? 
What is the reason why you 

1029
00:48:14,440 --> 00:48:16,320
exist? 
And it should not be about you. 

1030
00:48:16,320 --> 00:48:17,520
It should be about your 
customer. 

1031
00:48:18,000 --> 00:48:19,520
Great. 
Well, thank you very much. 

1032
00:48:19,560 --> 00:48:19,960
Thank you.