Jan. 7, 2026

The Biggest Lending Lessons We Learned in 2025

Show Notes
Transcript

In this special year-end edition of Leaders in Lending, Drew Megrey, Barry Roach, and Lynn Sautter Beal reflect on the most impactful conversations from 2025. As lending leaders navigated a year defined by uncertainty, rapid technological change, and evolving regulatory landscapes, this episode revisits the insights that shaped the industry. From credit risk strategy and AI model governance to new member engagement and the future of servicing, hear the best moments from standout episodes featuring voices across the financial ecosystem. Whether you're planning for 2026 or looking back on lessons learned, this compilation captures the defining themes of the year in lending.

WEBVTT

1
00:00:07.879 --> 00:00:10.199
Barry would love to get your perspective on kind of

2
00:00:10.199 --> 00:00:13.720
what's what's driving the Titan credit in twenty twenty five.

3
00:00:14.279 --> 00:00:19.559
Overall, I think just uncertainty. Right since we've had a

4
00:00:19.559 --> 00:00:23.280
new administration come in in January, there were threats of

5
00:00:23.719 --> 00:00:26.280
reductions in government workforce, which some have come to pass

6
00:00:26.320 --> 00:00:29.239
and some haven't. The regulatory environment has sort of been

7
00:00:29.239 --> 00:00:33.679
turned a little bit upside down. The CFPB doesn't seem

8
00:00:33.719 --> 00:00:38.600
to have as much maybe forward influence as an expectation

9
00:00:38.679 --> 00:00:41.159
than maybe what they had in the past. But then

10
00:00:41.320 --> 00:00:44.200
just in the economy, we've been waiting for a recession

11
00:00:44.320 --> 00:00:46.960
hasn't come, been waiting for rates to reduce, hasn't come,

12
00:00:47.119 --> 00:00:50.960
been waiting for unemployment to climb, hasn't necessarily I mean,

13
00:00:50.960 --> 00:00:52.560
it's come up a little bit, but the labor mark

14
00:00:52.640 --> 00:00:55.759
is still really strong. So all the sort of fundamental

15
00:00:55.880 --> 00:01:02.359
economic conditions that would normally pre predate a credit crunch

16
00:01:03.079 --> 00:01:05.519
haven't necessarily come to pass. So it's been more about

17
00:01:05.959 --> 00:01:10.319
our expectation or worry that's coming. And that's actually a

18
00:01:10.359 --> 00:01:14.840
prudent approach in the face of uncertainty. May not necessarily

19
00:01:14.840 --> 00:01:16.840
be the time to take on a whole lot of risk.

20
00:01:17.239 --> 00:01:19.840
Although I'll argue with myself a little bit on this.

21
00:01:19.920 --> 00:01:23.000
The counterpoint is, if everyone else is pulling back, isn't

22
00:01:23.040 --> 00:01:25.760
that a really good time for you to pull forward

23
00:01:25.079 --> 00:01:30.200
in an aggressive but in somewhat measured fashion. When I

24
00:01:30.239 --> 00:01:34.599
say measured meaning more about the quality of risk, you know,

25
00:01:34.640 --> 00:01:38.120
maybe that's something that where some credit unions and banks

26
00:01:38.120 --> 00:01:38.760
could win on.

27
00:01:43.719 --> 00:01:46.000
You said, Q four is the crystal ball. We're going

28
00:01:46.000 --> 00:01:48.280
into twenty six and we want to do X y

29
00:01:48.439 --> 00:01:51.159
Z from a lending perspective, and then things shifting Q

30
00:01:51.239 --> 00:01:53.799
one Are they shifting? Q two? Being able to have

31
00:01:53.840 --> 00:01:57.239
these tools at your fingertips allows you to reallocate your portfolio.

32
00:01:57.319 --> 00:01:59.640
Let's say you were heavily weighted in he lock, maybe

33
00:01:59.680 --> 00:02:02.680
second waited in the auto, and then personal loans, credit cards,

34
00:02:02.719 --> 00:02:06.120
on and so forth. And as things changed in the macro,

35
00:02:06.359 --> 00:02:10.199
as delinquencies rise, if there's margin compression, and you know

36
00:02:10.240 --> 00:02:13.599
the asset backed type of loans, you can reallocate your

37
00:02:13.639 --> 00:02:15.639
your strategy because you kind of know what's going on

38
00:02:15.719 --> 00:02:18.039
in real time or maybe with a three month lag.

39
00:02:18.080 --> 00:02:20.840
So I think it definitely helps in doing that because

40
00:02:20.879 --> 00:02:22.800
I remember back in my day, we would set budget

41
00:02:22.840 --> 00:02:24.800
and we would manage to budget for that full year,

42
00:02:24.919 --> 00:02:26.759
and we really didn't know what was going to change

43
00:02:28.400 --> 00:02:30.520
right away. So you got to the end of the

44
00:02:30.599 --> 00:02:32.159
year and you're like, up, we either hit budget or

45
00:02:32.199 --> 00:02:33.919
we were a little bit off, and the board's wondering

46
00:02:34.199 --> 00:02:36.039
why were we off or why were we over? So

47
00:02:36.120 --> 00:02:38.159
I think having those tools definitely helps kind of the

48
00:02:38.199 --> 00:02:41.280
real time decisions that the executives were making cross the

49
00:02:41.319 --> 00:02:42.159
lending ecosystem.

50
00:02:42.280 --> 00:02:44.280
Yeah, like think back Q one of twenty three when

51
00:02:44.319 --> 00:02:48.800
there was a liquidity crunch. You know, think of your

52
00:02:49.080 --> 00:02:51.759
lending partners, the credit unison banks you talked to across

53
00:02:51.759 --> 00:02:56.000
the country, those that had the ability to sort of

54
00:02:56.000 --> 00:02:59.240
course correct quickly or had already had the foresight to

55
00:02:59.360 --> 00:03:03.319
think of the sort of uh uh uh uh scenarios

56
00:03:03.360 --> 00:03:05.560
where you may need to course correct. I mean, they

57
00:03:05.560 --> 00:03:07.400
were able to move pretty fast, and they were able

58
00:03:07.479 --> 00:03:11.159
to get deposit products out there and and rates out

59
00:03:11.159 --> 00:03:13.520
there that that would quickly allow them to get a

60
00:03:13.560 --> 00:03:16.360
little more liquidy. And I did talk unfortunately some lending

61
00:03:16.400 --> 00:03:19.080
partners who were sort of stuck with well, now, what

62
00:03:20.080 --> 00:03:22.360
you know, where do we go from here? And and

63
00:03:22.520 --> 00:03:26.319
almost some inertia because of that, maybe because they they

64
00:03:26.759 --> 00:03:28.599
didn't take the time to sort of run those alternate

65
00:03:28.639 --> 00:03:30.800
scenarios where they didn't have the tools at their disposal

66
00:03:30.960 --> 00:03:33.599
or didn't know how to use those tools. So those

67
00:03:34.159 --> 00:03:36.919
executives on the sharper edge of the curve, with those

68
00:03:36.919 --> 00:03:39.280
tools available and the ability to use them, I think

69
00:03:39.319 --> 00:03:42.479
we're able to sort of uh to manipulate their their

70
00:03:42.520 --> 00:03:45.599
balance sheet and and their offerings out to their their

71
00:03:45.599 --> 00:03:48.960
consumers faster and be able to sort of mitigate any

72
00:03:49.319 --> 00:03:53.159
any sort of risks of that liquidated crunch that others

73
00:03:53.159 --> 00:03:58.840
had to experience, you know.

74
00:03:58.840 --> 00:04:02.400
As they think about like technology innovation AI, whether it's

75
00:04:02.520 --> 00:04:04.240
you know, the type of AI we use to do

76
00:04:04.560 --> 00:04:08.680
our underwriting and risk assessments or things like generative AI,

77
00:04:09.960 --> 00:04:13.080
the role of a fair lending officer or a fair

78
00:04:13.120 --> 00:04:16.279
responsible banking officer is getting more complex and you really,

79
00:04:16.600 --> 00:04:18.560
you know, it was already a very complex area, but

80
00:04:18.680 --> 00:04:21.759
now taking those rules and kind of applying them to

81
00:04:22.319 --> 00:04:25.199
rapidly changing new technology, Like how do you think about

82
00:04:25.680 --> 00:04:30.519
staying on top of technology developments and taking kind of

83
00:04:30.560 --> 00:04:35.120
the crux of the regulation which hasn't changed, and then

84
00:04:35.240 --> 00:04:39.360
measuring with new technologies like AI. That's a big light,

85
00:04:39.519 --> 00:04:40.160
lightweight question.

86
00:04:40.319 --> 00:04:43.040
I mean it really is, and especially coming from an

87
00:04:43.040 --> 00:04:45.759
online BAK where we do everything digitally to start with you,

88
00:04:46.160 --> 00:04:48.800
really we are becoming more and more involved in the

89
00:04:48.839 --> 00:04:53.279
process of AI machine learning in our institutions than ever before.

90
00:04:53.639 --> 00:04:56.480
And it's really based on the expectation of the regulators.

91
00:04:56.720 --> 00:04:58.399
They want us to make sure we're aware of what

92
00:04:58.480 --> 00:05:01.319
are in our models, What are the variables in there,

93
00:05:01.480 --> 00:05:04.920
Is there any potential risk of discrimination and if so,

94
00:05:05.279 --> 00:05:07.600
how are you controlling for it? Are you changing the

95
00:05:07.680 --> 00:05:10.759
variables that are in there? Are you requiring different standards

96
00:05:10.800 --> 00:05:16.560
around it. So we're keeping in touch with our models

97
00:05:16.600 --> 00:05:20.959
across our across our organization as well as across the industry.

98
00:05:21.240 --> 00:05:24.639
And that's we're again having a platform like CBA where

99
00:05:24.639 --> 00:05:27.879
we can come together and talk about my regulators looking

100
00:05:27.920 --> 00:05:32.040
for this within examinations or how we're looking for oversight

101
00:05:32.560 --> 00:05:35.920
versus a different crudential regulator that might have a different focus.

102
00:05:36.480 --> 00:05:39.519
We know last year there was the CFEB sent out

103
00:05:39.839 --> 00:05:44.199
a questionnaire to multiple organizations to gather more information on

104
00:05:44.680 --> 00:05:48.000
what are they doing for models and oversight internally. So

105
00:05:48.120 --> 00:05:50.720
as we're trying to figure it out, the regulators are

106
00:05:50.720 --> 00:05:57.360
trying to figure it out. As well, how do we.

107
00:05:57.319 --> 00:06:03.439
Get regulators comfortable with partnering with FinTechs and those FinTechs

108
00:06:03.480 --> 00:06:07.480
specifically that we be leveraging AI based under models thinking

109
00:06:07.519 --> 00:06:09.839
of the regulatory environment that we're dealing with today.

110
00:06:09.959 --> 00:06:13.079
Yeah, it's a good question, and I think it's a

111
00:06:13.160 --> 00:06:16.879
shifting regulatory environment on a day to day basis. I

112
00:06:16.879 --> 00:06:19.040
think all of us have certainly watched the changes with

113
00:06:20.160 --> 00:06:25.360
the various financial regulators this year and then now some

114
00:06:25.600 --> 00:06:27.759
court challenges to the leadership and who's on the board

115
00:06:27.759 --> 00:06:29.560
and who's not on the board, and what the rules are.

116
00:06:29.680 --> 00:06:33.519
So I would recommend for a credit union or really

117
00:06:33.560 --> 00:06:37.279
any lending institution, like just focus on the fundamentals. You

118
00:06:37.360 --> 00:06:39.959
may not know exactly what the rules are, You may

119
00:06:39.959 --> 00:06:42.800
not know exactly what the rules will be, but you

120
00:06:42.959 --> 00:06:47.720
do know that you should understand how models work. You

121
00:06:47.720 --> 00:06:51.680
should understand who owns which part of the decision. You

122
00:06:51.720 --> 00:06:54.560
should do your due diligence on any vendors that you

123
00:06:54.639 --> 00:06:57.959
partner with in a traditional way where you're looking at

124
00:06:58.000 --> 00:07:05.399
compliance risk, financial health, reputational risk, like even though those

125
00:07:05.439 --> 00:07:07.639
things are may not be explicitly called out anymore. Like

126
00:07:07.680 --> 00:07:11.480
reputational risk, as an example, that doesn't mean that you

127
00:07:11.519 --> 00:07:13.439
don't want to do that for your business. So think

128
00:07:13.480 --> 00:07:17.480
about from a core business standpoint, what do you care about?

129
00:07:17.560 --> 00:07:19.680
What do you think the key areas are that you

130
00:07:19.759 --> 00:07:23.560
should be able to both understand and document uh And

131
00:07:23.720 --> 00:07:26.399
then how can you up level your organization to really

132
00:07:26.439 --> 00:07:30.680
be confident in understanding and monitoring AI models? And I

133
00:07:30.680 --> 00:07:34.319
think that's a very big challenge. And I think similar

134
00:07:34.319 --> 00:07:37.920
to a lot of conversations we've had about different sized organizations,

135
00:07:37.920 --> 00:07:41.279
Like there's choices of partnering. There are many vendors, there

136
00:07:41.279 --> 00:07:43.959
are many consultants who help in this space today. So

137
00:07:44.439 --> 00:07:48.120
you may not be able to hire deep AI experts

138
00:07:48.120 --> 00:07:50.800
internally you don't need to, but can you find trusted

139
00:07:50.800 --> 00:07:53.319
partners that can help you do that evaluation and help

140
00:07:53.680 --> 00:07:56.959
and opportunities for your team to become to kind of

141
00:07:56.959 --> 00:07:59.959
deepen their subject matter expertise as it relates to AI

142
00:08:00.759 --> 00:08:05.639
and new and more innovative ways of evaluating risk and lending.

143
00:08:05.879 --> 00:08:08.199
Yeah, and that's a good like having a good general

144
00:08:08.319 --> 00:08:11.160
high level knowledge base of all things AI. I mean,

145
00:08:11.199 --> 00:08:15.000
it's the topic at hand most often anymore, right, but

146
00:08:15.199 --> 00:08:17.360
digging even deeper. So if you are partnered with some

147
00:08:17.399 --> 00:08:21.000
type of institution that is driving AI and some metric.

148
00:08:21.480 --> 00:08:24.399
How can you get more knowledge base into your internal

149
00:08:24.439 --> 00:08:28.040
teams on this company, does X, Y and Z. We

150
00:08:28.160 --> 00:08:30.199
kind of understand the full rails of it. Maybe we

151
00:08:30.240 --> 00:08:33.679
don't understand all of it, but we have confidence in

152
00:08:33.720 --> 00:08:36.919
going into a regulatory exam or speaking to regulators of

153
00:08:36.960 --> 00:08:40.360
why we're comfortable with this approach and having that for

154
00:08:40.399 --> 00:08:43.480
every single vendor that you have that leverages AI, so

155
00:08:43.519 --> 00:08:45.759
you have the high level knowledge base and then you

156
00:08:45.840 --> 00:08:48.120
have all the double clicks into everything else you're using

157
00:08:48.159 --> 00:08:55.279
it for. In the auto space right now, there's more

158
00:08:55.600 --> 00:08:57.919
new vehicles that are being purchased at dealerships. It's no

159
00:08:57.960 --> 00:09:02.000
longer kind of this used versus new type of mentality.

160
00:09:02.080 --> 00:09:04.919
If you think about though, people that got into a

161
00:09:04.960 --> 00:09:08.000
car such as there they bought their daughter a vehicle

162
00:09:08.120 --> 00:09:11.480
or so, the rates that they were offered back then

163
00:09:11.679 --> 00:09:15.120
were decent in a sense, probably somewhere between the high

164
00:09:15.120 --> 00:09:18.159
sixes mid eights, depending upon credit quality, things of that nature.

165
00:09:18.679 --> 00:09:22.240
But the newer vehicles that they're now selling are much

166
00:09:22.320 --> 00:09:25.960
higher trim packages, higher payments at aprs right now that

167
00:09:26.039 --> 00:09:29.039
are still fairly high. What do you think moves the

168
00:09:29.080 --> 00:09:31.759
needle for people to go back to purchasing new vehicles

169
00:09:31.879 --> 00:09:35.240
or refinance their existing and how many basis points does

170
00:09:35.279 --> 00:09:37.679
the FED reducing Does it make sense for them to

171
00:09:37.720 --> 00:09:40.080
go into a comfortable payment if they were to buy

172
00:09:40.080 --> 00:09:41.200
a useder versus new.

173
00:09:41.120 --> 00:09:45.159
Yeah, I mean in a refin deal if there's no

174
00:09:45.240 --> 00:09:48.120
other sort of fees associated with it, and typically there aren't.

175
00:09:48.720 --> 00:09:50.720
If I can reduce my payment by twenty five dollars,

176
00:09:50.759 --> 00:09:53.480
why wouldn't I do that? You know, if I'm able

177
00:09:53.519 --> 00:09:56.200
to even stretch out the term a little bit. You know,

178
00:09:56.200 --> 00:09:59.480
cars are lasting a lot longer now too, right, So

179
00:10:00.159 --> 00:10:02.639
what's the average age of vehicles on the road right now?

180
00:10:02.720 --> 00:10:06.039
It's over ten years, just because quality, I think is

181
00:10:06.039 --> 00:10:08.320
way better than it was a generation to go. So

182
00:10:08.480 --> 00:10:11.519
maybe I'm willing to extend that term out a little

183
00:10:11.559 --> 00:10:13.919
bit in order to get a little bit of a

184
00:10:13.960 --> 00:10:18.240
lower payment, even if my rate isn't necessarily all that different.

185
00:10:19.480 --> 00:10:21.200
Well, I certainly know. I know on one of our

186
00:10:21.200 --> 00:10:25.039
prior episodes, we had some of the folks from Experience

187
00:10:25.600 --> 00:10:30.879
on who specifically lead their auto market in the credit

188
00:10:30.960 --> 00:10:34.360
report and credit scoring area, and they're seeing a lot

189
00:10:34.399 --> 00:10:36.559
more of that, like those longer terms that people are

190
00:10:36.559 --> 00:10:38.879
taking a longer term now, whether or not they hold

191
00:10:38.879 --> 00:10:41.759
that car for very long, they're willing to take a

192
00:10:41.799 --> 00:10:45.320
longer term in exchange for a lower payment, and I

193
00:10:45.320 --> 00:10:48.320
think that is a big focus at payment versus rate

194
00:10:48.519 --> 00:10:51.559
is still for most people the biggest number that they

195
00:10:51.600 --> 00:10:55.039
look at, even though the rate long term can mean more,

196
00:10:55.080 --> 00:10:57.000
but it depends how long they're going to keep that car,

197
00:10:57.159 --> 00:10:58.440
or if they're going to trade it in in three

198
00:10:58.480 --> 00:11:01.159
years or they plan to actually.

199
00:11:00.919 --> 00:11:01.360
Pay it off.

200
00:11:01.440 --> 00:11:04.039
Our credit uns even did ninety six month loans for

201
00:11:04.240 --> 00:11:07.480
new vehicles, and we could even charge a little bit

202
00:11:07.480 --> 00:11:09.639
of a premium for that ninety six month loan and

203
00:11:09.919 --> 00:11:12.720
make a little more yield off it. So and what's

204
00:11:12.720 --> 00:11:14.519
stopping you from doing that? In the refine market, you

205
00:11:14.559 --> 00:11:18.519
can do the same sort of pricing approach and take

206
00:11:18.559 --> 00:11:21.679
a few more basis points to make that to improve

207
00:11:21.720 --> 00:11:22.879
the profitability.

208
00:11:22.360 --> 00:11:22.759
Of the plant.

209
00:11:27.080 --> 00:11:29.240
All of our listeners, I'm sure in both of you

210
00:11:30.600 --> 00:11:34.960
that there's this idea and true in a lot of ways,

211
00:11:35.000 --> 00:11:37.080
I think founded by a lot of reality that indirect

212
00:11:37.120 --> 00:11:40.559
lending does not lead to long term members and that

213
00:11:40.559 --> 00:11:43.159
that's a kind of a transaction. And then it's hard

214
00:11:43.279 --> 00:11:46.919
if not impossible to build that. And according to a

215
00:11:46.919 --> 00:11:50.559
Pinwhill study earlier this year, seventy percent of newly open

216
00:11:50.600 --> 00:11:54.039
accounts go inactive in the first ninety days, so seventy percent.

217
00:11:54.120 --> 00:11:56.399
So you're spending the time and money and you know

218
00:11:56.519 --> 00:11:59.039
KYC to open up that account and then it goes

219
00:11:59.240 --> 00:12:02.240
inactive with kind of an anchor option. And I think

220
00:12:02.240 --> 00:12:04.320
part of that is the difficulty of doing things like

221
00:12:04.360 --> 00:12:08.240
switching direct deposit, switching bill pay and that just inertia

222
00:12:08.399 --> 00:12:10.200
that people have a lot to do and that's kind

223
00:12:10.240 --> 00:12:13.679
of the last thing on their list. So how do

224
00:12:13.720 --> 00:12:15.120
you think about that? How do you think about this

225
00:12:15.200 --> 00:12:17.559
kind of indirect lending. Get this new consumer, get this

226
00:12:17.600 --> 00:12:21.320
new member, and drive it to become even if not

227
00:12:21.399 --> 00:12:24.240
maybe like the full wallet share like part of that

228
00:12:24.360 --> 00:12:26.840
and getting a meaningful part that makes that a more

229
00:12:27.399 --> 00:12:29.240
economically attractive new member.

230
00:12:29.559 --> 00:12:32.679
I've heard that very I'm sure you've heard that same

231
00:12:33.039 --> 00:12:36.480
concept of getting an indirect member to become a core

232
00:12:36.559 --> 00:12:40.039
member or a primary financial institution is something that just

233
00:12:40.080 --> 00:12:42.519
can't happen. And I'm going to disagree with that, and

234
00:12:42.639 --> 00:12:46.639
reason because is you have multiple credit unions that play

235
00:12:46.639 --> 00:12:51.240
in the indirect space, they're very good at driving automobile

236
00:12:51.320 --> 00:12:53.720
loans to their balance sheet, right and they're getting a member.

237
00:12:53.759 --> 00:12:56.519
But the consumer at the time of being at a

238
00:12:56.519 --> 00:12:59.519
dealership cares about one thing, and it's driving the vehicle

239
00:12:59.559 --> 00:13:03.559
off the line that day, and at most times probably

240
00:13:03.799 --> 00:13:06.919
not favorable conditions as it relates to APR for probably

241
00:13:06.919 --> 00:13:09.399
the payments a little bit out of their general sense

242
00:13:09.399 --> 00:13:12.320
of what they wanted to do there. But what happens

243
00:13:12.360 --> 00:13:15.480
on the back end, I think is what separates this

244
00:13:15.679 --> 00:13:18.799
actually coming to fruition without naming names. We have a

245
00:13:18.879 --> 00:13:21.399
lending partner that we partner with here at Upstart that

246
00:13:21.480 --> 00:13:23.960
is seeing about a ten to fifteen percent conversion rate

247
00:13:24.279 --> 00:13:27.120
on their indirect side because of the amount of touches

248
00:13:27.159 --> 00:13:29.919
they're doing after the auto loan comes onto their balance sheet.

249
00:13:30.200 --> 00:13:32.480
And I don't know if that's a secret sauce of

250
00:13:33.039 --> 00:13:35.960
on the day, three, day, seven, day fifteen. And then

251
00:13:36.000 --> 00:13:39.039
another thing that is attractive is you have to incent

252
00:13:39.159 --> 00:13:43.440
them just because the institution provides an apy of X

253
00:13:43.960 --> 00:13:47.159
or you're able to use your debit card at wherever

254
00:13:47.200 --> 00:13:50.559
you want for free. If you have some form of

255
00:13:50.600 --> 00:13:54.360
an incentive, then bringing in those members to become core

256
00:13:54.440 --> 00:13:56.559
members and not having them close their account out within

257
00:13:56.679 --> 00:13:59.200
ninety ninety days that you alluded to, you're going to

258
00:13:59.200 --> 00:14:00.000
see higher conversion.

259
00:14:00.399 --> 00:14:03.559
Yeah, I think incentive incentives are really important, and I

260
00:14:03.840 --> 00:14:05.759
like if you would look at my personal wallet, I

261
00:14:05.919 --> 00:14:08.200
have more probably a higher number of credit cards than

262
00:14:08.200 --> 00:14:10.399
you would think I would because I love incentives. I've

263
00:14:10.440 --> 00:14:13.679
got an Amazon Prime one, I use my American Express

264
00:14:13.679 --> 00:14:15.960
for travel. I have very fit for purpose cards that

265
00:14:15.960 --> 00:14:18.559
I use for very specific things. Or maybe I get

266
00:14:18.600 --> 00:14:20.840
a six month no interest on this and then I'll

267
00:14:20.879 --> 00:14:23.399
pay it off before that time. So like those things

268
00:14:23.399 --> 00:14:26.440
are for people who want to use those rewards, those

269
00:14:26.480 --> 00:14:27.480
can be super beatingful.

270
00:14:27.759 --> 00:14:31.720
Sorry you mentioned about onboarding. So multiple touch points through

271
00:14:31.759 --> 00:14:36.519
probably multiple different methods, right, Some might be an email,

272
00:14:36.559 --> 00:14:39.039
some might be a text, some might be you know,

273
00:14:39.080 --> 00:14:41.000
an actual letter that shows up in someone's mail. It

274
00:14:41.039 --> 00:14:43.799
might be a phone call, depending on.

275
00:14:43.759 --> 00:14:45.639
All the above. And then to take it even a

276
00:14:45.639 --> 00:14:50.200
step further is just because Linn walks into a dealership

277
00:14:50.360 --> 00:14:52.919
and buys a specific type of vehicle and Barry walks

278
00:14:52.919 --> 00:14:56.759
into a different dealership, there's different personas everyone's at a

279
00:14:56.759 --> 00:14:58.440
different stage in their life. You have to take it

280
00:14:58.480 --> 00:15:02.360
a deep step further and have different campaigns, different touch points,

281
00:15:02.480 --> 00:15:05.480
different touch points, But what is your incentive offering to

282
00:15:05.519 --> 00:15:07.919
the person at that point in time in their life.

283
00:15:08.039 --> 00:15:10.440
And there's so much data that's at your fingertips that

284
00:15:10.480 --> 00:15:14.480
you're able to create the messaging around that point in

285
00:15:14.480 --> 00:15:18.240
their life, whether it's through looking through their credit reporting agency,

286
00:15:18.320 --> 00:15:20.679
you're able to see the income kind of based on

287
00:15:21.039 --> 00:15:24.399
everything that's come through from underwriting perspective to tailor your

288
00:15:24.440 --> 00:15:27.360
messaging to become or to get people to become more

289
00:15:27.360 --> 00:15:27.879
core members.

290
00:15:27.919 --> 00:15:30.120
I'm going to take your seventy percent stat and I'll

291
00:15:30.200 --> 00:15:32.840
turn it around, say, how about thirty percent actually don't

292
00:15:32.919 --> 00:15:35.480
close their account or aren't inactive in the first ninety

293
00:15:35.559 --> 00:15:38.639
days anyone go fishing before? I mean, if you got

294
00:15:38.679 --> 00:15:41.279
thirty percent of your cast had some sort of a bite,

295
00:15:41.399 --> 00:15:43.080
you'd be out there every day. Right. I quit my

296
00:15:43.159 --> 00:15:45.519
job just to be a full time fishing So you know,

297
00:15:45.559 --> 00:15:48.240
there's an opportunity there that I think is somewhat lost.

298
00:15:48.360 --> 00:15:51.039
We tend to focus on the negatives of it. Of all,

299
00:15:51.480 --> 00:15:53.919
ninety percent are never going to be well, what about

300
00:15:53.919 --> 00:15:55.919
that ten or twenty or thirty, you know, ten to

301
00:15:55.960 --> 00:15:59.120
fifteen percent? It is a lot. It really is and

302
00:15:59.559 --> 00:16:02.200
can be meaningful. So part of that is the cost

303
00:16:02.200 --> 00:16:04.120
of doing business. You know, you're going to cast out

304
00:16:04.159 --> 00:16:05.799
and you're not going to get any bites at all.

305
00:16:10.360 --> 00:16:12.000
One of the things you said before, and I think

306
00:16:12.039 --> 00:16:14.080
you may have shared this at the client summit with

307
00:16:14.120 --> 00:16:19.279
many of our clients and potential clients, is that servicing

308
00:16:19.320 --> 00:16:23.240
is really just underwriting with an operational twist. What did

309
00:16:23.279 --> 00:16:23.840
you mean by that?

310
00:16:24.799 --> 00:16:28.120
Yeah, you know, as you as you mentioned, I have

311
00:16:28.320 --> 00:16:34.240
been at Upstart for I guess seven years last Saturday,

312
00:16:34.840 --> 00:16:37.919
so just just past the seven year mark. And of

313
00:16:37.919 --> 00:16:41.240
course I had spent the first five years working, you know,

314
00:16:41.360 --> 00:16:45.720
very deeply in upstarts core business that is predicting who

315
00:16:45.879 --> 00:16:50.039
will default and then pricing them correctly, our our underwriting business.

316
00:16:50.639 --> 00:16:52.960
And so when I joined servicing a little over two

317
00:16:53.039 --> 00:16:55.639
years ago, I really tried to think quite a lot

318
00:16:55.679 --> 00:16:58.080
about what it meant for us to take our our

319
00:16:58.120 --> 00:17:02.559
core competency and to apply those skills to this relatively

320
00:17:02.600 --> 00:17:04.960
new space. And I guess the way that I would

321
00:17:04.960 --> 00:17:07.759
say is that in underwriting, your job is to take

322
00:17:07.880 --> 00:17:12.279
maybe all of these applicants. You're taking applicants, you're sort

323
00:17:12.279 --> 00:17:14.720
of lining them up in riskiness. You're spacing them not correctly,

324
00:17:14.759 --> 00:17:20.200
and then you're determining generally what risk based price is

325
00:17:20.200 --> 00:17:22.480
associated with the level of risk that you see for

326
00:17:22.559 --> 00:17:26.039
each applicant. And I think in many ways you really

327
00:17:26.079 --> 00:17:28.839
want to do the same kind of thing with servicing.

328
00:17:28.920 --> 00:17:31.240
You want to maybe take all of these loans, you

329
00:17:31.279 --> 00:17:34.799
want to line them up from least risky to most risky.

330
00:17:35.559 --> 00:17:39.319
But in underwriting you have this sort of unfair advantage

331
00:17:39.759 --> 00:17:44.160
that is that you can maybe take any person whose

332
00:17:44.400 --> 00:17:48.440
risk is too high, and if you don't like how

333
00:17:48.519 --> 00:17:50.640
risky they are, well you just don't offer them alone.

334
00:17:51.160 --> 00:17:53.960
And in servicing we actually have a little bit more

335
00:17:54.000 --> 00:17:58.240
operational complexity. We are forced to maintain our relationships with

336
00:17:58.279 --> 00:18:01.200
these customers, and so instead of being able to maybe

337
00:18:01.240 --> 00:18:03.640
just say, oh, this person is too risky and we're

338
00:18:03.799 --> 00:18:06.119
not going to work with them at all, we instead

339
00:18:06.160 --> 00:18:09.519
have to maybe look only at the range of operational

340
00:18:09.599 --> 00:18:12.400
levers that we have to change their riskiness today to

341
00:18:12.440 --> 00:18:15.400
move from how risky they are today to how risky

342
00:18:15.440 --> 00:18:18.200
they might be tomorrow. And I think of this as

343
00:18:18.920 --> 00:18:24.279
a really really powerful part of our overall opportunity at Upstart,

344
00:18:24.319 --> 00:18:27.720
because it allows us to move from just predicting who

345
00:18:27.720 --> 00:18:30.440
will default to actually shaping the whole default.

346
00:18:34.640 --> 00:18:37.559
You shared how Great Lakes has you know, helped members

347
00:18:37.599 --> 00:18:41.839
with thinner credit files start building credit and rebuilding their

348
00:18:41.880 --> 00:18:45.279
you know, their financial confidence. I'd love to hear about

349
00:18:45.400 --> 00:18:47.279
how you went about doing that and how has it

350
00:18:47.319 --> 00:18:50.160
served as a real alternative to UH to pay day lending.

351
00:18:50.279 --> 00:18:54.000
Yes, great question, Drew. Thanks. So if we look back

352
00:18:54.039 --> 00:18:57.440
at GLCU, we're a low income designated credit union, a

353
00:18:57.480 --> 00:19:01.319
community development credit union, so we really start those of

354
00:19:01.440 --> 00:19:05.599
modest means, So over fifty one percent of our membership

355
00:19:05.759 --> 00:19:09.599
is low income. We serve a very diverse population just

356
00:19:09.599 --> 00:19:12.079
because of where we sit in the Chicago suburbs. Have

357
00:19:12.160 --> 00:19:15.839
branches in the city and in suburbia. So when we

358
00:19:15.880 --> 00:19:19.799
look at serving those members, especially those that are more

359
00:19:19.960 --> 00:19:23.039
challenged from a credit perspective, we need to make sure

360
00:19:23.079 --> 00:19:25.640
that we have products and services that are going to

361
00:19:26.039 --> 00:19:29.240
benefit them and add that value. And one thing that

362
00:19:29.279 --> 00:19:33.240
we realized is we just can't have a standard suite

363
00:19:33.359 --> 00:19:38.119
of lending products that we say really fit all demographics,

364
00:19:38.119 --> 00:19:40.920
because as you and I know, they don't. So we've

365
00:19:40.920 --> 00:19:45.119
been very diligent in our efforts to have products and

366
00:19:45.200 --> 00:19:49.839
services that are going to help our members improve their lifestyle. Right,

367
00:19:50.400 --> 00:19:53.720
some of it is changing their behavior, some is just

368
00:19:53.799 --> 00:19:56.680
giving them an opportunity to get back into what we

369
00:19:56.720 --> 00:20:01.240
would consider more traditional banking. So couple examples that we've

370
00:20:01.319 --> 00:20:04.119
utilized at Great Links Credit Union. One is our credit

371
00:20:04.160 --> 00:20:08.920
builder loan. We're very proud of this loan. It's simply

372
00:20:09.039 --> 00:20:13.480
one thousand dollars. We don't charge an application fee, there's

373
00:20:13.640 --> 00:20:17.720
zero percent, and it's set up on a repayment plan.

374
00:20:18.200 --> 00:20:20.920
And the idea is that we're reporting that back to

375
00:20:21.000 --> 00:20:23.880
the credit bureau because they are making payments on it,

376
00:20:24.319 --> 00:20:27.640
we hold that thousand dollars, so in essence, there's never

377
00:20:27.640 --> 00:20:31.759
any delinquency right or a charge off, and it rebuilds

378
00:20:31.799 --> 00:20:34.119
their file and it gives them that ability then to

379
00:20:34.319 --> 00:20:37.039
use that in their credit file to look at other

380
00:20:37.200 --> 00:20:42.480
opportunities for lending in the future. And so I would

381
00:20:42.480 --> 00:20:46.799
say it's been somewhat popular. We have about forty loans

382
00:20:47.400 --> 00:20:51.960
on file right now, and those that have utilized it

383
00:20:52.640 --> 00:20:56.759
have graduated then into other lending products such as getting

384
00:20:56.960 --> 00:21:01.440
a credit card or getting a small personal And so

385
00:21:01.720 --> 00:21:05.200