Meeting Title: Bonde.AI Demo Date: 2025-11-07 Meeting participants: Uttam, Clint Dunn, Samuel Roberts, Demilade Agboola, Awaish Kumar
WEBVTT
1 00:01:34.400 ⇒ 00:01:39.140 Uttam: Hey, dude. Give me a sec. Let me just try to get everybody in.
2 00:01:42.430 ⇒ 00:01:43.840 Clint Dunn: No worries.
3 00:02:37.210 ⇒ 00:02:37.800 Uttam: Sam.
4 00:03:01.550 ⇒ 00:03:03.690 Clint Dunn: Well, I’ll be back in, like, get my symptoms. Okay.
5 00:03:04.850 ⇒ 00:03:05.389 Samuel Roberts: Yeah, here we go.
6 00:03:05.390 ⇒ 00:03:06.120 Clint Dunn: Yo.
7 00:03:06.700 ⇒ 00:03:08.170 Clint Dunn: I can hear you, yeah.
8 00:03:08.170 ⇒ 00:03:08.780 Samuel Roberts: What’s quiet.
9 00:03:10.020 ⇒ 00:03:11.310 Samuel Roberts: Headphone issues.
10 00:03:13.440 ⇒ 00:03:16.440 Clint Dunn: I’m in Cincinnati with my girlfriend for our wedding.
11 00:03:16.690 ⇒ 00:03:28.670 Clint Dunn: we’re, like, split… we have, like, two rooms, because this is, like, an extended stay kind of hotel, but we’re just… like, one of us… I’m gonna have to go stand in the bedroom here in a minute. Yeah.
12 00:03:29.020 ⇒ 00:03:29.990 Uttam: That’s funny.
13 00:03:30.330 ⇒ 00:03:32.030 Uttam: Sam’s in Ohio, too.
14 00:03:32.180 ⇒ 00:03:33.340 Samuel Roberts: Yeah, I’m up to Cleveland.
15 00:03:34.120 ⇒ 00:03:35.090 Clint Dunn: Oh, really?
16 00:03:35.090 ⇒ 00:03:35.890 Samuel Roberts: Yeah.
17 00:03:36.840 ⇒ 00:03:40.329 Clint Dunn: Dude, it was… did you get hit with that storm a minute ago?
18 00:03:40.550 ⇒ 00:03:46.010 Samuel Roberts: There’s… it’s wet outside, I don’t know when it started, but yeah, probably the same thing coming through.
19 00:03:46.220 ⇒ 00:03:51.579 Clint Dunn: Dude, it was… blowin’ here. It was, raining really hard, and .
20 00:03:51.580 ⇒ 00:03:53.259 Samuel Roberts: really hard, and .
21 00:03:53.510 ⇒ 00:03:56.180 Clint Dunn: And a bunch of thunder and stuff, but…
22 00:03:57.050 ⇒ 00:04:11.170 Samuel Roberts: Yeah, I don’t think we quite had that yet, but I’m sure it’s the tail end of that, but yeah, my wife’s mom is down in Cincinnati, and so I hear about the weather down there, and it’s funny, because my family’s in Boston, and so we don’t always have the same weather in Cincinnati, but Boston always is, like, a day behind whatever we have in Cleveland.
23 00:04:11.510 ⇒ 00:04:12.680 Clint Dunn: Yup, yup, yup.
24 00:04:13.150 ⇒ 00:04:15.610 Samuel Roberts: But sometimes Cincinnati’s just different enough, yeah.
25 00:04:16.089 ⇒ 00:04:22.579 Clint Dunn: Like, Scentsy, I’ve learned, is, like, all P&G people. Is, like, Cleveland similar, like, big consumer products?
26 00:04:22.820 ⇒ 00:04:29.459 Samuel Roberts: No, no, cleveland… I mean, the biggest employer here, I think, is the Cleveland Clinic.
27 00:04:29.800 ⇒ 00:04:31.530 Samuel Roberts: They’re a huge hospital system.
28 00:04:31.530 ⇒ 00:04:32.150 Clint Dunn: Hmm.
29 00:04:32.350 ⇒ 00:04:33.520 Samuel Roberts: International and stuff.
30 00:04:33.520 ⇒ 00:04:34.020 Uttam: Okay.
31 00:04:34.020 ⇒ 00:04:35.820 Samuel Roberts: But yeah, it’s not really that. I mean, it’s, it’s like…
32 00:04:35.820 ⇒ 00:04:36.490 Clint Dunn: Oh, yeah.
33 00:04:36.490 ⇒ 00:04:38.040 Samuel Roberts: Cleveland’s one of those, like, Rust Belt…
34 00:04:38.650 ⇒ 00:04:48.200 Samuel Roberts: Kinda lots of old manufacturing that was here. Still some of it here, but, you know, steel plants and stuff that aren’t anymore, so… But a lot of biotech now because of the clinic.
35 00:04:50.390 ⇒ 00:04:55.480 Clint Dunn: Yeah, that pharma just seems, like, so big over here, which I hadn’t really…
36 00:05:04.450 ⇒ 00:05:05.890 Samuel Roberts: What else are we waiting on?
37 00:05:11.120 ⇒ 00:05:14.529 Uttam: I pinged, I pinged Henry and Awash,
38 00:05:15.680 ⇒ 00:05:18.490 Uttam: We can probably get started.
39 00:05:18.490 ⇒ 00:05:18.880 Clint Dunn: Cool.
40 00:05:18.880 ⇒ 00:05:19.650 Uttam: I guess…
41 00:05:19.750 ⇒ 00:05:34.190 Uttam: I kind of give a little bit of an intro, but I guess short intro is Clint and the Bond team are all friends of Brainforge. I think they have an interesting product that we could consider for a few of our clients, but also
42 00:05:34.350 ⇒ 00:05:38.200 Uttam: I think as we started doing a lot more…
43 00:05:38.820 ⇒ 00:05:45.359 Uttam: We’re just starting to do a lot more work that’s very parallel to this, so it could be another offering for us to promote.
44 00:05:45.580 ⇒ 00:05:48.509 Uttam: So yeah, I’ll kind of maybe give it to you, Clint, and…
45 00:05:48.750 ⇒ 00:05:53.709 Uttam: Kind of take it from here. And then, yeah, this recording, too, I’ll circulate internally as well, so…
46 00:06:00.460 ⇒ 00:06:07.490 Clint Dunn: Yeah, cool. I think, can you guys hear me okay? I think my computer’s freezing a little bit.
47 00:06:09.020 ⇒ 00:06:15.000 Uttam: You’re good? Yeah. I feel like you don’t have to go on video if it’s ending up taking too much bandwidth.
48 00:06:15.300 ⇒ 00:06:16.080 Samuel Roberts: Yeah.
49 00:06:19.040 ⇒ 00:06:26.700 Clint Dunn: I think that’s my problem. Alright, going off video, which is… to your guys’ detriment, you don’t get to see me, sorry.
50 00:06:26.710 ⇒ 00:06:40.359 Clint Dunn: But, the… Yeah, Utam and I have known each other for a long time. I guess, like, we’ve been friends for 4 or 5 years now. Utam actually introduced me to my co-founder.
51 00:06:40.390 ⇒ 00:06:43.540 Clint Dunn: For which I am forever indebted, and I’m…
52 00:06:43.600 ⇒ 00:06:46.800 Clint Dunn: Utemo, am I wrong? Did I introduce you to Robert, too?
53 00:06:46.800 ⇒ 00:06:48.230 Uttam: Yes, yes.
54 00:06:48.230 ⇒ 00:06:48.910 Samuel Roberts: Wow.
55 00:06:50.040 ⇒ 00:06:51.010 Clint Dunn: Yeah, so…
56 00:06:51.010 ⇒ 00:06:52.189 Uttam: Very fruitful friendship.
57 00:06:52.190 ⇒ 00:06:52.690 Clint Dunn: book training.
58 00:06:53.150 ⇒ 00:06:53.919 Clint Dunn: Yeah, yeah.
59 00:06:55.960 ⇒ 00:07:16.109 Clint Dunn: And so, my co-founder and Utam used to work together, and so I think, you know, just, like, the way in which we even, like, approach problems is pretty similar to you guys. So, anyway, yeah, this is, like, a pretty open-ended conversation. I think, you know, I’ve been learning a little bit more about what you guys are working on, and how you guys view the world, and so there’s probably some ways we can,
60 00:07:16.110 ⇒ 00:07:23.480 Clint Dunn: we can work together, but, you know, I think for us, we want to see what you guys think, and what excites you, and what you think we should be working on.
61 00:07:24.030 ⇒ 00:07:39.810 Clint Dunn: Super high level, you know, my background, I’m a data guy, I was the marketing data science lead at Afterpay, I built and led the data team in a couple e-com brands, and so, we’re very, very ecom focused, as a company. We…
62 00:07:40.650 ⇒ 00:08:00.229 Clint Dunn: I think the thing that we’ve seen is that data teams are really good at getting out, like, you know, kind of standard monthly reporting, board-level reporting, guiding those big KPIs for the brand, but it’s just really hard to be nimble enough to answer a bunch of ad hoc questions and be part of that decision-making process, so…
63 00:08:00.260 ⇒ 00:08:04.130 Clint Dunn: the AI that we’ve… the product that we’ve been building.
64 00:08:04.250 ⇒ 00:08:11.009 Clint Dunn: Effectively sits on top of a bunch of first- and third-party data, and pulls out
65 00:08:11.390 ⇒ 00:08:22.680 Clint Dunn: It basically sits on top of first- and third-party data, and folks within a brand can ask questions freeform in a chat interface, and get any of those,
66 00:08:22.680 ⇒ 00:08:35.060 Clint Dunn: those answers back really quickly. So we mostly work with, like, you know, multi-hundred-million dollar, e-com brands, where they have a couple dozen employees and are trying to, like, push context and decisions as quickly as possible.
67 00:08:36.549 ⇒ 00:08:45.129 Clint Dunn: That is, like, the super high-level, no-deck version. I’m happy to go deeper on any of the conceptual stuff if anyone wants to.
68 00:08:48.510 ⇒ 00:08:55.469 Uttam: Yeah, I feel like there’s a lot of opportunity, away from MLI for this type of product on Eden, probably.
69 00:08:55.600 ⇒ 00:09:00.070 Uttam: And probably a couple other clients, so yeah, kind of pumped to see it.
70 00:09:00.400 ⇒ 00:09:01.860 Clint Dunn: Yeah, cool.
71 00:09:02.360 ⇒ 00:09:07.439 Clint Dunn: So… Sorry, I’m changing locations here, but
72 00:09:07.950 ⇒ 00:09:13.989 Clint Dunn: you know, I think the conceptual stuff’s a little bit less fun than just showing you what we have going on.
73 00:09:15.560 ⇒ 00:09:17.559 Clint Dunn: So, everyone see this okay?
74 00:09:19.410 ⇒ 00:09:20.020 Awaish Kumar: Yes.
75 00:09:20.930 ⇒ 00:09:38.640 Clint Dunn: So, pretty simple, brands, when they sign up, this is our demo account, the burn-ins, if anyone’s a barbecue guy. This is our barbecue-focused demo account. But brands can come in here, and they connect all of their first-party information.
76 00:09:38.640 ⇒ 00:09:43.730 Clint Dunn: So, you know, fairing is, like, survey data that a lot of our brands use.
77 00:09:43.730 ⇒ 00:09:52.680 Clint Dunn: Pretty dead simple. Authenticate through here, provide us the API key, and as soon as you do that, the agent has access to all of that data.
78 00:09:55.410 ⇒ 00:10:06.410 Clint Dunn: We have a bunch of integrations here, I… we’re, like, that’s a big, big push for us for the rest of the year, is getting as many integrations as possible. And then Q1 next year, we’re gonna be doing, like.
79 00:10:06.540 ⇒ 00:10:11.179 Clint Dunn: A version of this where we can build directly on top of a data warehouse.
80 00:10:12.700 ⇒ 00:10:22.369 Clint Dunn: Which obviously is gonna lend itself a little bit to what you guys are doing, and some more, like, flexibility in using different data sources.
81 00:10:25.060 ⇒ 00:10:28.790 Awaish Kumar: Okay, so right now, you connect directly to the…
82 00:10:29.100 ⇒ 00:10:35.640 Awaish Kumar: Like, for example, these platforms and share the… Audience insights?
83 00:10:35.880 ⇒ 00:10:36.690 Awaish Kumar: Sure.
84 00:10:37.150 ⇒ 00:10:44.459 Clint Dunn: Yeah, yeah, I’ll give you a quick demo of what the, like, agent looks like and what some typical questions are, but the big thing right now is we’re…
85 00:10:44.730 ⇒ 00:10:52.910 Clint Dunn: we’re connecting directly to a lot of these sources. In terms of the agent itself,
86 00:10:53.200 ⇒ 00:10:57.470 Clint Dunn: really any question can be asked, as long as we have the data. So we have…
87 00:10:57.540 ⇒ 00:11:17.290 Clint Dunn: heads of customer service using this, we have merchandisers, we have planners, we have heads of retention, heads of, paid ads, all using it. And so there’s quite a bit of flexibility. So, you know, like, really simply, like, you know, which, products
88 00:11:17.470 ⇒ 00:11:26.840 Clint Dunn: Do consumers buy first, and then have… highest LTVs, right? Which is…
89 00:11:27.000 ⇒ 00:11:29.980 Clint Dunn: This is typically a question that requires…
90 00:11:30.260 ⇒ 00:11:41.399 Clint Dunn: quite a bit of SQL work, and probably some, like, data structure and analytics engineering work, but we’re able to push this out pretty quickly.
91 00:11:42.510 ⇒ 00:11:52.059 Clint Dunn: So, what it’s doing in the background, we obviously have our own data warehouse that we’re managing with all of these data sources, where we’ve cleaned stuff up and kind of
92 00:11:52.170 ⇒ 00:11:55.170 Clint Dunn: made it workable for the AI.
93 00:11:55.490 ⇒ 00:12:00.820 Clint Dunn: And then the AI is going in, it’s executing its own queries, it’s checking its work.
94 00:12:02.300 ⇒ 00:12:05.779 Clint Dunn: And then it spits out a nice, clean little table like this.
95 00:12:12.950 ⇒ 00:12:13.470 Clint Dunn: I keep…
96 00:12:13.470 ⇒ 00:12:14.419 Uttam: What did you guys end up…
97 00:12:14.420 ⇒ 00:12:15.540 Clint Dunn: Sorry.
98 00:12:15.540 ⇒ 00:12:17.210 Uttam: Would you… yeah.
99 00:12:17.380 ⇒ 00:12:19.779 Uttam: What’d you guys end up using for the,
100 00:12:20.030 ⇒ 00:12:23.000 Uttam: Like, ETL, did you end up going with Polyatomic?
101 00:12:23.870 ⇒ 00:12:39.470 Clint Dunn: We didn’t, just because they don’t have… so we’ve been DIYing everything, so far, so everything you’re seeing right now is, like, our own. I don’t think we’ll end up going with Polytomic, just because they don’t have the, kind of, like, coverage that we need.
102 00:12:39.800 ⇒ 00:12:40.629 Uttam: Yeah, okay, fair enough.
103 00:12:41.060 ⇒ 00:12:49.570 Clint Dunn: Yeah, yeah, but, they were super nice, like, I liked them, it just, like, you know, we basically needed to, like, request every…
104 00:12:50.100 ⇒ 00:12:52.870 Clint Dunn: Integration that we were gonna end up using.
105 00:12:53.150 ⇒ 00:13:10.049 Uttam: Totally. Yeah, we… we ended up just doing it because it was… it was so much more reliable and cost-effective than… than Fivetran, and they… they built… they built stuff for us in, like, a week turnaround. But yeah, like, you guys are a product, like, it has to kind of work, but…
106 00:13:10.170 ⇒ 00:13:14.959 Uttam: I don’t know, good guys, like, you should consider them longer term. In terms of the ETL world, like…
107 00:13:15.690 ⇒ 00:13:18.190 Uttam: Yeah, kind of the only people that are, like, worth it.
108 00:13:18.530 ⇒ 00:13:25.090 Clint Dunn: Yeah, I mean, you know, frankly, like, the Powered By product is, like, not… we’re not, like, satisfied with it. Yeah.
109 00:13:25.450 ⇒ 00:13:28.580 Clint Dunn: obviously, there’s some, like, lock-in on the ETL.
110 00:13:28.580 ⇒ 00:13:29.350 Uttam: For sure, for sure.
111 00:13:29.870 ⇒ 00:13:37.240 Clint Dunn: But yeah, like, I… you know, I wouldn’t say, like, our NPS on that, like, powered by Fivetran product is, like, super high.
112 00:13:37.530 ⇒ 00:13:42.329 Uttam: No, that… I mean, it sucks, yeah. It’s… it’s horrible, yeah. I don’t… I mean, like…
113 00:13:42.570 ⇒ 00:13:46.369 Uttam: Yeah, it’s just a buy versus build sort of thought, but yeah, yeah.
114 00:13:46.370 ⇒ 00:13:48.210 Clint Dunn: Okay, cool. Yeah.
115 00:13:48.280 ⇒ 00:14:08.930 Clint Dunn: And so, like, you know, from here, we’ve got this clean, nice table with all of our, kind of, barbecue-related products, what the average LTV is after purchase. It even proactively gave us the number of first-time customers, because it knows sample size is a big deal to us. And, like, pretty dead simple if we want to copy this.
116 00:14:08.940 ⇒ 00:14:16.230 Clint Dunn: Table, we can even download it into a CSV or mark down and share that really easily with other folks on our team.
117 00:14:17.180 ⇒ 00:14:23.560 Awaish Kumar: Is it directly working on the raw data, or, like, you maintain the transformations?
118 00:14:23.700 ⇒ 00:14:25.080 Awaish Kumar: On top of.
119 00:14:25.340 ⇒ 00:14:32.709 Clint Dunn: Yeah, it’s somewhere in between. So we… it’s not… it’s not raw data, like, we are cleaning it, and we normalize it, so…
120 00:14:33.080 ⇒ 00:14:52.400 Clint Dunn: Like, all of the paid ads data is normalized into the same format so that you can compare, you know, Google performance and Facebook performance apples to apples. One of the big things, obviously, is there has to be some customization that goes in.
121 00:14:52.580 ⇒ 00:14:54.749 Clint Dunn: To each brand, so, you know.
122 00:14:55.000 ⇒ 00:15:01.540 Clint Dunn: really simple example, like, some brands calculate revenue differently. Some want to calculate gross revenue.
123 00:15:01.540 ⇒ 00:15:17.330 Clint Dunn: And others want to calculate, you know, revenue after discounts, like net revenue. And so we actually handle that in the, like, AI layer, and let the AI handle that, rather than pushing a lot of complexity into the data modeling layer.
124 00:15:19.670 ⇒ 00:15:20.480 Awaish Kumar: Okay.
125 00:15:20.940 ⇒ 00:15:29.999 Awaish Kumar: And, like, does… The, like, the users of the product can do it, like… From the UI?
126 00:15:30.000 ⇒ 00:15:46.320 Clint Dunn: Yeah, yeah. Next week, we’re rolling out a new page that’ll let you define your own KPIs, define your own, like, goals, and kind of, like, North Star metrics, and then also define kind of, like, your ICP jargon.
127 00:15:46.320 ⇒ 00:16:00.920 Clint Dunn: you know, lore around your brand. So, right now, we’re kind of just, like, handling it in the back end, but next week, that’ll be, a new feature released for us, where you’ll get to actually manage that on your own as a user in the product.
128 00:16:02.660 ⇒ 00:16:09.400 Awaish Kumar: Okay, so, yeah, my… Question is, like, so once I redefine that.
129 00:16:09.560 ⇒ 00:16:14.170 Awaish Kumar: KPIs and metrics for it, which really matter to us, and then…
130 00:16:14.580 ⇒ 00:16:23.279 Awaish Kumar: like, basically, can you… you are connecting directly with the data, which is coming from, for example, Amazon? Yeah.
131 00:16:24.040 ⇒ 00:16:26.060 Awaish Kumar: answers it, like, there’s no…
132 00:16:26.880 ⇒ 00:16:31.129 Awaish Kumar: like, hard toss of wishes. Like, I know, like, there are some cleanups, but, like.
133 00:16:31.480 ⇒ 00:16:33.440 Awaish Kumar: I would understand if…
134 00:16:33.570 ⇒ 00:16:42.070 Awaish Kumar: you’re doing, like, DVD modeling and all of it in between, or it’s just pure, input to the AI agent?
135 00:16:42.810 ⇒ 00:16:45.050 Clint Dunn: Are we doing, like, dbt modeling?
136 00:16:45.370 ⇒ 00:16:46.220 Awaish Kumar: Yeah.
137 00:16:46.220 ⇒ 00:16:58.570 Clint Dunn: Yeah, yeah, we do a lot. You know, we, like, like I said, like, we let the AI handle some of the more opinionated stuff, or, like, brand-specific stuff, but we have our own…
138 00:16:58.780 ⇒ 00:17:15.080 Clint Dunn: kind of… you have to build a data warehouse in a little bit different of a way than you would if you were building it for, like, a human consumption. But we do build, you know, a warehouse, we have data transformation, we have data normalization…
139 00:17:15.310 ⇒ 00:17:22.240 Clint Dunn: We do kind of the basics that you would do internally. And then, you know, on the flip side of that, like.
140 00:17:22.369 ⇒ 00:17:35.870 Clint Dunn: we also, in next year, we’ll start building on top of people’s warehouses directly. And so if you have, like, a strong POV or, you know, like, really robust warehouse, we can just sit on top of that and pull information from there.
141 00:17:39.080 ⇒ 00:17:39.550 Demilade Agboola: Oh.
142 00:17:39.550 ⇒ 00:17:40.260 Awaish Kumar: Okay.
143 00:17:41.330 ⇒ 00:17:50.989 Demilade Agboola: I have a question. I’m wondering how vague you can ask questions to it. Like, if someone just came and said, hey, tell me something interesting about my data.
144 00:17:50.990 ⇒ 00:17:51.730 Clint Dunn: Yeah.
145 00:17:51.730 ⇒ 00:17:55.630 Demilade Agboola: Will they be able to do that, or do you kind of have to be specific?
146 00:18:03.380 ⇒ 00:18:07.540 Clint Dunn: Let’s see. I think proof is in the pudding on this kind of stuff.
147 00:18:13.320 ⇒ 00:18:25.529 Clint Dunn: Yeah, so, I mean, like, your definition of interesting is probably going to be different, depending on who you are. But I’ve found, like, especially when we do survey data, and you say, like, give me an overview
148 00:18:31.810 ⇒ 00:18:38.550 Clint Dunn: And you just say, like, give me an overview, it usually will give you, like, you know, some pretty key facts that it’ll pull out on its own.
149 00:18:45.570 ⇒ 00:18:49.819 Clint Dunn: One of the big things that we’re pushing,
150 00:18:51.220 ⇒ 00:18:57.999 Clint Dunn: in the next couple weeks is, like, more autonomous, agentic work. And so we’re kind of, like.
151 00:18:58.010 ⇒ 00:19:11.380 Clint Dunn: Ramping up for these more open-ended questions and more multi-part questions that’ll allow you to answer really big, big, nubby problems, rather than the kind of, like, pointed questions that it does really well at right now.
152 00:19:22.750 ⇒ 00:19:29.490 Clint Dunn: I can’t see anyone’s face, so I’m just… I’m taking, like, irrationally long pauses in between points.
153 00:19:31.590 ⇒ 00:19:35.569 Demilade Agboola: No, that’s fine. I think, yeah, over time, having, like.
154 00:19:35.950 ⇒ 00:19:41.429 Demilade Agboola: specific… like, special questions are great, but I think sometimes, you know.
155 00:19:41.640 ⇒ 00:20:01.230 Demilade Agboola: People want someone who kind of, like, goes through the data, and just extracts, like, just, like, cool stuff, like, oh, if you did this, potentially you could, you know, make money on this, or, you know, you could do something off of that, or even if it’s honestly making that assumption, or making that conclusion.
156 00:20:01.340 ⇒ 00:20:03.950 Demilade Agboola: But just giving enough…
157 00:20:04.880 ⇒ 00:20:10.419 Demilade Agboola: Enough of, like, interesting data points to then just go in and try and, like.
158 00:20:10.560 ⇒ 00:20:14.029 Demilade Agboola: finalize whatever hypothesis you could come up with, I guess.
159 00:20:14.400 ⇒ 00:20:14.920 Clint Dunn: Yeah.
160 00:20:15.100 ⇒ 00:20:18.479 Clint Dunn: I don’t know what you think about this, but one of our ideas…
161 00:20:18.650 ⇒ 00:20:21.580 Demilade Agboola: That we’ve been building towards is, like.
162 00:20:22.340 ⇒ 00:20:41.739 Clint Dunn: Having the agent summarize all the questions that you’ve been asking over the past week, and let’s say, you know, you as a user were asking a bunch of questions about, like, promotions and discounts, the agent could recognize a bunch of questions and angles that you were taking about promotions and discounts, and then run its own autonomous.
163 00:20:41.740 ⇒ 00:20:44.399 Clint Dunn: Analysis to try and, like.
164 00:20:44.470 ⇒ 00:20:50.750 Clint Dunn: Kind of extend your questions and get you something novel that you hadn’t thought about or hadn’t asked about.
165 00:20:53.280 ⇒ 00:21:04.780 Demilade Agboola: Yeah, I think that would be pretty cool, because a lot of people just, you know, want to get, like, whatever insights, and, like, at the end of the day, the name of the game is insights, and if you can…
166 00:21:05.180 ⇒ 00:21:09.590 Demilade Agboola: Trish turned the data to Insights, and so, like, just being able to see how
167 00:21:10.090 ⇒ 00:21:19.349 Demilade Agboola: yeah, it can get you closer to it. Maybe not necessarily right at the door, but, like, you know, gets you so close that you’re able to, you know, take the final step by yourself.
168 00:21:19.620 ⇒ 00:21:21.830 Clint Dunn: Yeah, yeah. Do you think…
169 00:21:22.550 ⇒ 00:21:26.979 Clint Dunn: Like, where do you stand on, like, the, like, recipe books, or, like, kind of…
170 00:21:28.230 ⇒ 00:21:30.399 Clint Dunn: Like, recipes to ask questions.
171 00:21:34.640 ⇒ 00:21:43.650 Demilade Agboola: I would say, like, a lot of… a lot of places that people care about is, for instance, what campaigns are doing well,
172 00:21:44.090 ⇒ 00:21:46.490 Demilade Agboola: What are the things people are buying a lot of?
173 00:21:46.640 ⇒ 00:21:53.430 Demilade Agboola: What do we make the most margin on? So, like, COGS, like, revenue minus COGS.
174 00:21:53.430 ⇒ 00:21:53.980 Clint Dunn: Yeah.
175 00:21:53.980 ⇒ 00:22:05.879 Demilade Agboola: So, like, are we doing… so, like, that would then lead to things like, are we selling what we need to be selling in terms of margin, versus, like, are we selling things that don’t have as much margin?
176 00:22:06.640 ⇒ 00:22:13.609 Demilade Agboola: Things like, you know, Potential, like, underexplored
177 00:22:14.850 ⇒ 00:22:22.199 Demilade Agboola: campaigns. So, for instance, the campaign might convert higher, but have had less puts into it.
178 00:22:22.200 ⇒ 00:22:22.880 Clint Dunn: Yeah.
179 00:22:23.150 ⇒ 00:22:37.000 Demilade Agboola: Things like that. Just, like, areas in which people will just, like, want to see how can we improve our margins or reduce, increase profit or reduce losses, with this data that we have right now.
180 00:22:37.760 ⇒ 00:22:39.590 Clint Dunn: Yeah, 100%. Yeah, I think…
181 00:22:40.420 ⇒ 00:22:45.299 Clint Dunn: I, you know, we’ve kicked around the idea of, like, having recipes that kind of have these, like, pre…
182 00:22:45.420 ⇒ 00:22:56.360 Clint Dunn: determined questions that you ask, and then the agent, you know… basically, if people don’t have a question they know they need to ask, we give them some ideas. But I do think the kind of, like.
183 00:22:57.200 ⇒ 00:23:06.560 Clint Dunn: More exploratory, like, anonymous angle, or autonomous angle that you were just getting at is, like, probably the most interesting to me.
184 00:23:08.600 ⇒ 00:23:09.410 Demilade Agboola: I agree.
185 00:23:14.100 ⇒ 00:23:17.949 Clint Dunn: Are you guys doing, like, ERP integrations right now?
186 00:23:18.290 ⇒ 00:23:22.320 Uttam: Yeah, we do work with… NetSuite a bunch.
187 00:23:22.320 ⇒ 00:23:24.480 Clint Dunn: Okay. It’s the most common we’re seeing.
188 00:23:25.860 ⇒ 00:23:33.100 Clint Dunn: I haven’t gone down that route on integrations. We’re hearing kind of, like, whispers of it as a longer-term integration.
189 00:23:33.100 ⇒ 00:23:36.030 Uttam: You’re gonna hear NetSuite, like, a shitload, I feel like.
190 00:23:36.030 ⇒ 00:23:38.000 Clint Dunn: Yeah. For the most part.
191 00:23:38.000 ⇒ 00:23:44.510 Uttam: What’s your guys’, like, POV on that? Do you… what do you… what kind of information are you pulling from there, and do you find it useful?
192 00:23:44.980 ⇒ 00:23:47.360 Uttam: Yeah, Demolade is expert.
193 00:23:48.120 ⇒ 00:23:55.259 Demilade Agboola: So for Nets, with the people who use it, to be fair, I’m not sure it’s necessarily the perfect combination.
194 00:23:55.520 ⇒ 00:23:58.800 Demilade Agboola: But the people who use it are,
195 00:23:59.270 ⇒ 00:24:14.399 Demilade Agboola: into, like, they’re flor… like, florists, basically. And so, as a result, they have, like, a lot of back and forth with, like, inventory management. Yeah. I do know that sometimes NetSuite can be a bit of a hassle.
196 00:24:14.490 ⇒ 00:24:16.569 Clint Dunn: Yup. In terms of…
197 00:24:16.620 ⇒ 00:24:23.950 Demilade Agboola: when data hits the warehouse, like, sometimes things have been… they run out of stock. NetSuite does not send a zero…
198 00:24:24.110 ⇒ 00:24:33.360 Demilade Agboola: zero, count row. It just kind of meets it and doesn’t, like, like that, like, let the warehouse know. So you kind of have to, like.
199 00:24:33.380 ⇒ 00:24:49.000 Demilade Agboola: have catches for that, but generally speaking, yeah, NetSuite can do a decent amount, and we do use it for, like, oh, just keeping track of inventory, what was sent to where, because they… can they have, like, a different, like, they have,
200 00:24:49.470 ⇒ 00:25:04.189 Demilade Agboola: a model where, for instance, they have fulfillment centers, so that’s the final delivery point that sends out to the people who purchase flowers, but they also have, like, hubs that get those flowers and then send to the fulfillment center.
201 00:25:04.920 ⇒ 00:25:13.349 Demilade Agboola: need to keep track of where each flower or each batch of flowers are at any point in time. Are they at the hub? Are they at the… they call them spokes, which is fulfillment centers.
202 00:25:13.350 ⇒ 00:25:13.800 Clint Dunn: Yep.
203 00:25:13.800 ⇒ 00:25:26.500 Demilade Agboola: books, or, like, how much has… how much have we… have been ordered, or how much has been ordered for a batch of flowers? How much do we have in stock?
204 00:25:26.730 ⇒ 00:25:27.660 Clint Dunn: Yup. So…
205 00:25:27.760 ⇒ 00:25:42.119 Demilade Agboola: how much do we need to fulfill that? And you also have to factor in things like buffers, because they also have to… because, you know, flowers die, things can go bad, you need to have buffers bad, so things like that. That’s what they basically use the NetSuite for.
206 00:25:43.240 ⇒ 00:25:45.889 Clint Dunn: Interesting. The… like, and then…
207 00:25:46.260 ⇒ 00:25:53.120 Clint Dunn: Like, how… what proportion of questions do you guys feel like are kind of ad hoc?
208 00:25:53.310 ⇒ 00:26:01.220 Clint Dunn: Off-the-wall experimental questions versus the more just, like, straightforward reporting kind of concerns.
209 00:26:02.660 ⇒ 00:26:14.379 Demilade Agboola: So right now, I would say about 90% of what they… Ask are more, like… Straightforward questions, because…
210 00:26:15.220 ⇒ 00:26:16.820 Demilade Agboola: I mean…
211 00:26:17.130 ⇒ 00:26:27.979 Demilade Agboola: They had their inventory data set up, they migrated systems, things had not necessarily been aligned since their migration, and so we came in and kind of helped set that up.
212 00:26:27.980 ⇒ 00:26:42.070 Demilade Agboola: And so right now, they’re just kind of getting back to the point where they feel comfortable having a good idea of what’s happening, in the general scheme of things. Especially, like, ops, like, operations, they are sure that, like, oh, okay, we’re not running out of
213 00:26:42.180 ⇒ 00:26:46.430 Demilade Agboola: stock too often, we’re able to ensure that everything is… so, like, basically.
214 00:26:46.660 ⇒ 00:27:03.599 Demilade Agboola: people need data for, again, pure operations, but right now, they are not yet at that point where, like, they are thinking creatively about it. That’s part of what I was talking to Utam about, about them in particular, that I think we could push them a bit more creatively. Yeah.
215 00:27:04.060 ⇒ 00:27:07.949 Uttam: Yeah, I mean, we are, like,
216 00:27:08.410 ⇒ 00:27:18.069 Uttam: We kind of come in, like, Navy SEALs for the most part, so most of our engagements for the first 3 to 6 months are, like, fixing everything and establishing
217 00:27:18.620 ⇒ 00:27:20.159 Uttam: An analytics group.
218 00:27:20.320 ⇒ 00:27:24.769 Uttam: So, for a lot of our clients, we’re just now getting to the point where
219 00:27:24.950 ⇒ 00:27:36.070 Uttam: We’ve done that, and our time shifts to do what is fun about data, which is actually, like, proactive analysis-heavy stuff, and the modeling ends up being pretty mature.
220 00:27:36.490 ⇒ 00:27:43.610 Uttam: Even for us, and we’re pretty quick, takes, like, 6 months. What we’re getting better at, though, is…
221 00:27:43.820 ⇒ 00:27:52.420 Uttam: we will ship analysis in order to get buy-in to make better infrastructure decisions. Because most of our clients
222 00:27:52.710 ⇒ 00:27:56.810 Uttam: do not care about a warehouse, an ETL tool, like.
223 00:27:56.810 ⇒ 00:27:57.210 Clint Dunn: Correct.
224 00:27:57.210 ⇒ 00:28:03.730 Uttam: API, they don’t care about any of that. And if I’ve just pitched them on that, they’re like, why are you adding all these tools?
225 00:28:04.200 ⇒ 00:28:08.670 Uttam: like, how does this… they don’t get the through line, so for us, we have to sell that.
226 00:28:09.020 ⇒ 00:28:09.680 Clint Dunn: Yup.
227 00:28:09.680 ⇒ 00:28:11.190 Uttam: solving analysis.
228 00:28:11.290 ⇒ 00:28:31.240 Uttam: Pushing out one or two really, like, sophisticated analyses a week, and then using that to say, like, hey, we can develop more of these when they can be more robust and more accurate, and, like, naturally, they’ll say, oh, can we ask this follow-up question? We’re like, well, our… we have… our infrastructure limits us, and then we get to buy in that way. But it takes three to six months.
229 00:28:32.750 ⇒ 00:28:35.279 Uttam: Yeah, like, I’m…
230 00:28:35.540 ⇒ 00:28:41.650 Uttam: I think we’ll get faster at it. Yeah. But, we have some clients that, like, will become
231 00:28:41.750 ⇒ 00:28:57.740 Uttam: very, like, it will become a bigger job, but we’re just starting, and so when we start, we just kind of have to, like, get wins, or at least, and then we do… we challenge our customers and show them that, hey, a great infrastructure, here’s how it enables
232 00:28:58.290 ⇒ 00:29:00.250 Uttam: Faster, better analyses.
233 00:29:00.380 ⇒ 00:29:04.790 Uttam: Not the other way around, which is what I feel like most data places do, and, like.
234 00:29:04.790 ⇒ 00:29:05.340 Clint Dunn: Right.
235 00:29:05.470 ⇒ 00:29:07.910 Uttam: you’re gonna get fired.
236 00:29:08.070 ⇒ 00:29:09.620 Clint Dunn: Yeah. You know, so…
237 00:29:09.620 ⇒ 00:29:10.429 Uttam: Yeah, I was…
238 00:29:10.430 ⇒ 00:29:23.190 Clint Dunn: I… I was wondering if there’s, like, an angle there as well, for us, like, working together, because, you know, if you guys are able to stand up a couple tables in kind of a simple data model, and we sat on top of that, like.
239 00:29:23.380 ⇒ 00:29:28.870 Clint Dunn: This gives them a very easy way for them to explore the limits of that on their own.
240 00:29:29.000 ⇒ 00:29:29.420 Uttam: Yeah.
241 00:29:29.420 ⇒ 00:29:30.439 Clint Dunn: Well, that’s exactly.
242 00:29:30.440 ⇒ 00:29:34.799 Uttam: what I’m… what I’m thinking, too, is, like, to give you an example, we have many clients For whom…
243 00:29:35.100 ⇒ 00:29:38.479 Uttam: Chat with data is a much better medium than a dashboard.
244 00:29:38.650 ⇒ 00:29:39.090 Clint Dunn: Yeah.
245 00:29:39.090 ⇒ 00:29:40.120 Uttam: And,
246 00:29:41.120 ⇒ 00:29:51.699 Uttam: We are actually starting to advertise, like, a one-month sprint, and we’re pretty sophisticated at standing up, like, a core, pretty…
247 00:29:51.870 ⇒ 00:30:08.600 Uttam: like, impactful data infrastructure in, like, less than a month. So, warehouse, ETL, laying model, and, like, kind of arrive at, like, some simple marts, and then arrive at analysis. Like, we can bang that out pretty quickly, and, like, we’re having clients sign us up… sign up with us just for that, because…
248 00:30:08.880 ⇒ 00:30:26.009 Uttam: it’s… it’s huge. It’s a lot of work, but I don’t… I was all… I’ve always been fundamentally kind of, like, beefing with dashboards. I don’t think it’s a great last mile representation of a work, and for the most part, in my career, no one’s gotten better at using dashboards.
249 00:30:26.010 ⇒ 00:30:32.870 Uttam: So, this is the first time where I think there’s a new mode to accessing data, and so my question for you would be.
250 00:30:33.840 ⇒ 00:30:41.829 Uttam: You know, in that… in that sort of offering for us, when we come in and promise that month-long thing, one deliverable could we… could be we leave you with
251 00:30:42.130 ⇒ 00:30:46.129 Uttam: Like, basically an ability for you to chat with data.
252 00:30:46.130 ⇒ 00:30:46.640 Clint Dunn: Yeah.
253 00:30:46.640 ⇒ 00:30:55.780 Uttam: And, again, like, we don’t… we make a… we make almost 6 to 7 vendor decisions in the sprint, by the way. So, I mean, for most of our clients.
254 00:30:55.940 ⇒ 00:30:58.049 Uttam: We… we are,
255 00:30:58.330 ⇒ 00:31:16.190 Uttam: in many cases, we are the single decision maker on vendors, and in other situations, we are a key decision… a key part in that decision, and most likely the most informed. So, there’s a high chance we can implement it. I guess, for me, it would be, I would love to hear, like.
256 00:31:16.480 ⇒ 00:31:20.079 Uttam: How you’re thinking about pricing, how do you feel about, like.
257 00:31:20.280 ⇒ 00:31:29.439 Uttam: trials or proof of concepts, like… Yeah. Because I think this is totally… and then maybe… sorry, this is just a little bit of a rainbow, but I talked to Ril and Omni.
258 00:31:29.660 ⇒ 00:31:33.209 Uttam: Both of which are releasing Chat with AI features.
259 00:31:33.210 ⇒ 00:31:35.020 Clint Dunn: But both of which, I said.
260 00:31:35.720 ⇒ 00:31:44.380 Uttam: Hey, I don’t actually really… I almost don’t really care about the dashboarding piece, like, all too much. I actually would like to just use your engine.
261 00:31:44.500 ⇒ 00:31:58.029 Uttam: to chat with… to chat over data, and they both want, like, annual contracts or everything, and, like, obviously expensive, so… Yeah. I just don’t want to build this, and so I want… I need to… we need to go with a vendor, for sure.
262 00:31:58.030 ⇒ 00:31:58.620 Clint Dunn: Yeah.
263 00:31:58.800 ⇒ 00:32:00.990 Clint Dunn: Yeah, totally aligned on that. I think…
264 00:32:01.060 ⇒ 00:32:19.530 Clint Dunn: look, like, obviously all the BI tools are building this stuff out, and, like, they’re gonna have great products. I think the differentiator for us is on two different fronts, that are… and they’re both kind of related, right? And the basic idea is, like, we are built for commerce brands.
265 00:32:19.710 ⇒ 00:32:37.039 Clint Dunn: And so, one half of that is, like, there is a lot of intelligence that we give the agent about commerce brands, about jargon, about best strategies, that it just knows out of the box. Like, this is actually a decent example.
266 00:32:37.550 ⇒ 00:32:43.509 Clint Dunn: It… it pulled out non-canceled orders without us prompting it to do so.
267 00:32:43.680 ⇒ 00:32:52.409 Clint Dunn: And that’s, like, an example where most brands I talk to want to report revenue for the last quarter on non-canceled
268 00:32:52.540 ⇒ 00:33:01.090 Clint Dunn: you know, subtotal revenue, and it’ll do that, you know, pull out that subtotal revenue, that post-discount revenue on its own. So…
269 00:33:01.480 ⇒ 00:33:12.650 Clint Dunn: there are a lot of things like that, that the agent is just built for commerce, in a way that, like, you know, HEX is not going to be. Another example of that is, like.
270 00:33:12.650 ⇒ 00:33:30.010 Clint Dunn: The agent was doing some survey data analysis, and so it was looking at, you know, how customers were responding to the survey, about how they heard about this brand that we’re working with. And a bunch of customers were starting to come in from,
271 00:33:30.340 ⇒ 00:33:31.480 Clint Dunn: podcasts.
272 00:33:31.750 ⇒ 00:33:54.519 Clint Dunn: And this is a supplement company, and so, like, the agent was like, hey, there’s a bunch of people coming in from podcasts, and the user was kind of like, okay, cool, tell me more about it. And it was like, actually, it’s coming… they’re coming in from, like, Huberman, they’re coming in from, like, this podcast and this podcast. The cool thing about these podcasts are there’s, like, 3 different ICPs, right? And so, like, it didn’t just, like, pull out the data report on the…
273 00:33:54.720 ⇒ 00:34:10.290 Clint Dunn: podcast itself, it started talking about the voice of each of those podcasts, and gave… you know, they basically used this… this agent, or this conversation, to build, like, a marketing budget. You know, they said, I have $150,000,
274 00:34:10.739 ⇒ 00:34:19.039 Clint Dunn: how should I spend it? Where should my ad reads go? It even came up with a script that Andrew Huberman should read versus Tim Ferriss.
275 00:34:19.050 ⇒ 00:34:37.439 Clint Dunn: So, you know, again, building for commerce alone gives us a lot of feature, optionality that we can build that the, like, data people just, you know, the BI tools won’t. And to that end, the second part is, we have, like, upsell add-ons for…
276 00:34:37.620 ⇒ 00:34:40.449 Clint Dunn: Third-party demographic information.
277 00:34:40.880 ⇒ 00:34:58.040 Clint Dunn: And so, you know, there are a lot of tools where these folks are, like, buying data from Axiom, and then they have very limited usage out of that Axiom data, because it’s… it’s just so cumbersome to do demographic analysis. It’s, like, perfect for an AI agent to do it, and so we can bolt that on super easily.
278 00:34:58.040 ⇒ 00:35:11.469 Uttam: Yeah, no, that’s actually a perfect use case. I mean, again, we have clients across SaaS and e-com and this, and we would just totally use you just for that. In fact, like, again, this is something that I… I don’t have a choice but to go with a vendor here, because…
279 00:35:12.140 ⇒ 00:35:12.840 Uttam: It’s…
280 00:35:12.960 ⇒ 00:35:20.370 Uttam: it’s just something that I… we can’t build ourselves, and I don’t want to go deep on, but the only people that, like, apart from…
281 00:35:20.580 ⇒ 00:35:26.289 Uttam: I mean, apart from just, like, random, kind of, like, Texas SQL API companies, like.
282 00:35:26.580 ⇒ 00:35:35.079 Uttam: who I just don’t, like, I don’t know, I just don’t really trust, and then, like, the big BI tools, like, there’s nobody. And so, if you guys are like, hey, we’re branding purely as just, like.
283 00:35:35.480 ⇒ 00:35:39.399 Uttam: and again, I’m oversimplifying, but just, like, this, like, chat with data for
284 00:35:39.710 ⇒ 00:35:50.139 Uttam: e-com, or commerce, then that’s what we would do. And so for us, again, like, we deploy a solution of tools in addition to, like, our people and process, so we would just
285 00:35:50.670 ⇒ 00:35:52.099 Uttam: Line you guys up with that.
286 00:35:52.100 ⇒ 00:35:52.510 Clint Dunn: Yeah.
287 00:35:52.510 ⇒ 00:35:56.600 Uttam: I think… I think the biggest… the best way to work with us, though.
288 00:35:56.600 ⇒ 00:35:59.480 Clint Dunn: Is that for many of our clients, we have to show.
289 00:35:59.480 ⇒ 00:36:03.349 Uttam: And then we sort of were like, okay, what do you think? You know, so…
290 00:36:03.350 ⇒ 00:36:03.810 Clint Dunn: Yeah, I’m…
291 00:36:03.810 ⇒ 00:36:11.730 Uttam: For example, yeah, as long as you’re okay with that, and all the vendors that we’ve been playing really nicely with have been okay, because we’ve converted a shit ton
292 00:36:11.760 ⇒ 00:36:27.130 Uttam: applies to them. Polyatomic is a good example where, like, never put pressure on us, and we’ve run, like, one or two month trials, and then ultimately they end up buying, and it’s like, for some vendors, they’re really picky, just because everybody’s getting burned by, like.
293 00:36:27.230 ⇒ 00:36:29.020 Uttam: SaaS vendors right now, so…
294 00:36:29.020 ⇒ 00:36:29.540 Clint Dunn: Of course.
295 00:36:29.540 ⇒ 00:36:36.619 Uttam: That’s… and then… but in that sense, also, it’s not like a… it’s not like you’re giving a free trial to a rando, like, we’re not,
296 00:36:37.040 ⇒ 00:36:42.809 Uttam: I’ll tell you every step of the way if this thing’s gonna go or not, but I think that’s where we could just use help
297 00:36:43.100 ⇒ 00:36:44.910 Uttam: And making the best case.
298 00:36:45.260 ⇒ 00:36:48.100 Uttam: And yeah, I mean, I feel.
299 00:36:48.100 ⇒ 00:36:48.880 Clint Dunn: Yeah.
300 00:36:48.880 ⇒ 00:36:50.359 Uttam: I feel fine about it.
301 00:36:50.360 ⇒ 00:36:54.580 Clint Dunn: Yeah, yeah, I mean, like, well, I want you to feel great about it, Utan.
302 00:36:57.230 ⇒ 00:37:07.830 Clint Dunn: I think, yeah, like, look, like, a couple, like, fast facts for you is we’re completely free for the rest of the year. We’re not, like, officially launching until next year, so…
303 00:37:08.540 ⇒ 00:37:15.539 Clint Dunn: Fast fact. That’s a fast fact for you. Another fast fact, we can be super flexible in the kind of, like, trial side of things. Okay.
304 00:37:15.540 ⇒ 00:37:30.199 Clint Dunn: You know, we have some, like, hard costs, with… between ETL and, like, obviously tokens are expensive and things like that, but, like, you know, I think I know you well enough, we can… we can write off a lot of that stuff.
305 00:37:30.200 ⇒ 00:37:30.790 Uttam: project.
306 00:37:30.790 ⇒ 00:37:43.469 Clint Dunn: Right? And the other thing is, I think, honestly, like, an ROI case for us is pretty easy and obvious. Like, I’m having all these pricing discussions right now, and I just basically go in, I go, look, you have, like.
307 00:37:43.730 ⇒ 00:37:51.420 Clint Dunn: you have 7 people from 6 departments using it, and they asked 200 questions this month. You know, like, from.
308 00:37:51.420 ⇒ 00:37:51.960 Uttam: Yeah, yeah.
309 00:37:51.960 ⇒ 00:38:02.749 Clint Dunn: perspective, it’s like, those are questions that are taken off your plate for custom development that probably… Yeah. And then, and then from their perspective, it’s like, you know, you just need to make decisions faster in a brand, and… For sure.
310 00:38:03.680 ⇒ 00:38:07.630 Clint Dunn: It’s hard to quantify that, but that is how they make more money.
311 00:38:08.220 ⇒ 00:38:08.810 Uttam: Okay.
312 00:38:09.140 ⇒ 00:38:09.850 Clint Dunn: Yeah.
313 00:38:10.250 ⇒ 00:38:10.960 Uttam: Okay.
314 00:38:11.390 ⇒ 00:38:12.190 Uttam: Cool.
315 00:38:12.620 ⇒ 00:38:17.950 Clint Dunn: Yeah, I don’t know what makes sense on your end, but, like, from our end,
316 00:38:18.170 ⇒ 00:38:32.220 Clint Dunn: like, I think it would be interesting if you are kind of friendly with a brand, and you have, kind of, like, Shopify access of some sort, just, like, plugging them in and giving it a spin with some actual data. That’s… that’s typically how we…
317 00:38:32.320 ⇒ 00:38:34.619 Clint Dunn: Have folks try us out and trial it.
318 00:38:35.030 ⇒ 00:38:35.600 Uttam: Yeah.
319 00:38:35.950 ⇒ 00:38:37.920 Uttam: Tell me, like, who,
320 00:38:38.470 ⇒ 00:38:43.410 Uttam: who ends up actually using it the most? Like, are you finding, like, people who are…
321 00:38:43.720 ⇒ 00:38:47.599 Uttam: used to be in, like, a lot of spreadsheets, are you finding it more leadership? Like…
322 00:38:48.230 ⇒ 00:38:52.130 Clint Dunn: It’s mostly, like, director-level, manager-level people.
323 00:38:52.990 ⇒ 00:39:08.830 Clint Dunn: And below. I actually find, like, directors like it the most, because they, right now, are, like, strategizing, a lot, and they have these kind of, like, off-the-wall questions that are not really supported by, like, the existing dashboarding or analytics engineering infrastructure.
324 00:39:08.830 ⇒ 00:39:09.390 Uttam: Oh.
325 00:39:09.660 ⇒ 00:39:19.099 Clint Dunn: So all of those kind of, like, crazy questions you get about, like, oh, well, what if we, like, you know, reduced our discounting rate by, like, XYZ?
326 00:39:19.100 ⇒ 00:39:27.110 Clint Dunn: And, like, all of that stuff, they just end up filtering to us, and… and, that’s, like, typically, but, like, you know, there’s not…
327 00:39:27.110 ⇒ 00:39:42.819 Clint Dunn: we have heads of CX that love us, we have, like, heads of merchandising that love us, we have heads of retention marketing, like, I don’t… I’m not sure there’s, like, a vertical or org, that I would go, but, like, in terms of level, I would say definitely, like, director and below is… is kind of our sweet spot right now.
328 00:39:43.070 ⇒ 00:39:43.730 Uttam: Okay.
329 00:39:46.120 ⇒ 00:39:49.460 Clint Dunn: I think C-suite gets too distracted.
330 00:39:49.460 ⇒ 00:39:50.010 Uttam: Yeah.
331 00:39:50.010 ⇒ 00:39:51.229 Clint Dunn: But different things, too.
332 00:39:51.230 ⇒ 00:39:55.390 Uttam: I mean, I guess, like, Awash, what do you think, like… Yeah, what do you think?
333 00:39:57.180 ⇒ 00:40:09.160 Awaish Kumar: Yeah, my question was more, like, where we want to connect it with, like, for example, for all times, do we wanna… we already built, kind of, DVD models and everything.
334 00:40:09.160 ⇒ 00:40:09.560 Clint Dunn: Yo.
335 00:40:10.260 ⇒ 00:40:20.930 Awaish Kumar: So, like, for us to, like, maybe connect it at the bottom of it, so we can maybe directly connect with our warehouse and marts-level tables.
336 00:40:21.410 ⇒ 00:40:22.100 Clint Dunn: Yeah.
337 00:40:22.670 ⇒ 00:40:24.949 Awaish Kumar: LinkedIn, so we can communicate with it.
338 00:40:25.400 ⇒ 00:40:35.909 Clint Dunn: Yeah, yeah, yeah. So, that’s, like, a Q1 feature for us, is the warehouse integration, which I think is, like, a no-brainer for us to work on. We also need some, like.
339 00:40:36.550 ⇒ 00:40:46.359 Clint Dunn: kind of custom work to make that integration happen, and so that’s, like, another great way for us to kind of partner, and you guys to upsell on your end.
340 00:40:46.770 ⇒ 00:40:48.070 Clint Dunn: So, I think…
341 00:40:48.550 ⇒ 00:40:52.759 Demilade Agboola: Quick question, what warehouses are you looking at integrating with?
342 00:40:53.460 ⇒ 00:40:55.560 Clint Dunn: I mean, we’re agnostic on that front.
343 00:40:55.800 ⇒ 00:40:56.850 Demilade Agboola: Oh, yeah, gotcha.
344 00:40:56.850 ⇒ 00:41:01.739 Clint Dunn: Yeah. Is there a particular one that you’d want to see, or that you’re curious about?
345 00:41:02.260 ⇒ 00:41:11.920 Demilade Agboola: No, not really. I just know that, like, different people use different warehouses, so just having an idea of, like, what this will be compatible with will just be very helpful.
346 00:41:12.090 ⇒ 00:41:19.830 Clint Dunn: Yeah, I mean, like, we would build it in such a way that that wouldn’t be an issue. We…
347 00:41:20.160 ⇒ 00:41:30.669 Clint Dunn: use Mother Duck internally. So, we’re pretty used to, like, dealing with different warehouses and kind of handling that complexity.
348 00:41:31.590 ⇒ 00:41:32.740 Demilade Agboola: Okay, sounds good.
349 00:41:32.740 ⇒ 00:41:42.120 Clint Dunn: Yeah, but I think, like, you know, for us to really work together and do the warehouse integration, it’ll take a few months, and so that’s why my recommendation is just to, like.
350 00:41:42.190 ⇒ 00:42:01.279 Clint Dunn: you know, connect something with a friendly brand, just to give you an idea of where our platform’s at, and you can give us a little bit of feedback that we can work on, and then going into the new year, we can kind of… if there’s, like, some opportunity to scope out that feature together on, like, a warehouse integration, we can work on that together.
351 00:42:03.680 ⇒ 00:42:04.630 Awaish Kumar: Okay, nope.
352 00:42:05.070 ⇒ 00:42:10.540 Demilade Agboola: Do you have a list of things you connect to right now? Like, present day?
353 00:42:11.780 ⇒ 00:42:16.110 Clint Dunn: Yeah, back to Utam. So, this is, like, our list right now.
354 00:42:18.270 ⇒ 00:42:25.809 Clint Dunn: We have… but we’re moving to a vendor for this, so we actually… we have a hidden page, we’re migrating.
355 00:42:25.920 ⇒ 00:42:29.639 Clint Dunn: to this new… Connector page.
356 00:42:30.260 ⇒ 00:42:41.280 Clint Dunn: This is rough, right now. This is… these are not… this is not the full list of connectors, but we’re planning to build about 80… 80… 82 connectors right now.
357 00:42:41.940 ⇒ 00:42:43.619 Uttam: Who’d you guys end up going with?
358 00:42:43.880 ⇒ 00:42:44.870 Clint Dunn: 5 tran.
359 00:42:45.180 ⇒ 00:42:45.980 Uttam: Oh, okay.
360 00:42:46.320 ⇒ 00:42:47.190 Uttam: I think so.
361 00:42:47.410 ⇒ 00:42:47.840 Clint Dunn: Yeah.
362 00:42:49.520 ⇒ 00:42:49.870 Demilade Agboola: So…
363 00:42:49.870 ⇒ 00:42:52.220 Uttam: Yeah, I mean, for me, I think it would be…
364 00:42:52.350 ⇒ 00:42:57.280 Uttam: Really helpful to consider the warehouse integration, because we just have a shit ton of model data, you know?
365 00:42:57.280 ⇒ 00:42:57.960 Clint Dunn: Of course.
366 00:42:58.160 ⇒ 00:43:03.539 Uttam: The other, maybe the other comment, and this, I told this to the… to the Omni…
367 00:43:04.340 ⇒ 00:43:21.519 Uttam: I don’t know, they had, like, 5 founders, or one of those dudes. I said… I said, look, I said, the benefit of my company versus you guys just coming in and slapping your BI tool, and people asking complicated questions that aren’t just, like, how many orders did I sell, is that I… our job is to collect
368 00:43:21.590 ⇒ 00:43:25.819 Uttam: context, and Brainforge is a huge, like.
369 00:43:27.320 ⇒ 00:43:33.879 Uttam: whatever the… I don’t know what the leading, vacuum brand is, but we are that for context. Like, we write.
370 00:43:34.080 ⇒ 00:43:34.540 Clint Dunn: Uber.
371 00:43:34.540 ⇒ 00:43:40.519 Uttam: record… we would… yeah, it was, like, more like Dyson, you know? I won’t say we’re… no bag, bagless.
372 00:43:40.520 ⇒ 00:43:40.930 Clint Dunn: Alright.
373 00:43:40.930 ⇒ 00:43:55.869 Uttam: We… we just write a ton of context, we have a ton of meetings saved, we do a lot of documentation, and, we do a lot of decks, and, like, we… we just have so much… so for me, what’s important is that we…
374 00:43:55.990 ⇒ 00:44:05.369 Uttam: put that right where the staging is. So, the thing I… the Omni solution, these guys were like, oh, add column descriptions.
375 00:44:05.740 ⇒ 00:44:07.420 Uttam: description, so, like, what table, so…
376 00:44:07.420 ⇒ 00:44:07.940 Clint Dunn: No.
377 00:44:07.940 ⇒ 00:44:13.100 Uttam: Yeah, this is where, like, I would… and I’m just… I hear that you… yeah, it sounds like you’ve thought about this, but I guess.
378 00:44:13.100 ⇒ 00:44:13.470 Clint Dunn: Yeah.
379 00:44:13.470 ⇒ 00:44:18.150 Uttam: for me, like, I would love to literally just… stuff, like.
380 00:44:18.520 ⇒ 00:44:24.640 Uttam: a ton of markdown files somewhere where it can sit next to your thing, and you can reference it. Okay.
381 00:44:24.640 ⇒ 00:44:34.779 Clint Dunn: Yeah, so this is, like, this is our PO… I didn’t share this at the beginning, because I wanted to not share slides, but, like, this is how we view the world.
382 00:44:36.030 ⇒ 00:44:42.480 Uttam: Yeah, so the right… the right piece is all… we have the whole thing. And none of our clients that we walk into have that on their own, by the way.
383 00:44:42.480 ⇒ 00:44:46.799 Clint Dunn: No. And none of them do this, like, news and benchmarking information either.
384 00:44:46.800 ⇒ 00:44:48.029 Uttam: Very true, very true.
385 00:44:48.030 ⇒ 00:44:51.849 Clint Dunn: You know, so, like, even really, like, BI tools are just this.
386 00:44:52.040 ⇒ 00:44:56.139 Clint Dunn: And we think that’s actually why data teams have struggled.
387 00:44:56.320 ⇒ 00:45:12.459 Clint Dunn: the… this is, like, kind of our hypothesis in the world. Like, to make an optimal decision, you need all three points of the context to triangulate the optimal decision. And, like, data teams usually only come in with this. Business folks only come in with this, typically, and so you have this, like, weird, like.
388 00:45:12.500 ⇒ 00:45:33.329 Clint Dunn: decision handoff between these two teams. And so that’s, like, a big kind of hypothesis of our model, and I didn’t talk about this as much, of our agent, is bringing in this, like, brand context from team meetings and documents and things like that, and using… the agent using that as a reference point to understand what’s actually going on and what’s being discussed.
389 00:45:34.460 ⇒ 00:45:49.809 Clint Dunn: Like, a great example is, like, all of our brands right now are doing 2026 planning, and they’re, like, forecasting, and they’re making, like, oh, we should invest more in Facebook, and it’s, like, absolutely preposterous to me that you’re gonna build some, like, analytics agent and not have the context.
390 00:45:49.810 ⇒ 00:45:54.210 Uttam: No, they’re… no, they’re not, but this is where, like, they’re… I mean, I don’t know, I just…
391 00:45:54.360 ⇒ 00:46:12.439 Uttam: I think they’re stupid, you can maybe tell me they’re not, but I think they… they… they think of themselves as the end-all, and I’m like, look, I have what you’re never gonna get, which is extremely great structured context. And this is where, like, for me, the… I… you know, in thinking about…
392 00:46:12.560 ⇒ 00:46:23.150 Uttam: I… Devilani and Awage can tell you, like, we’ve talked about, like, hey, we should just build our own agent, because we have the contacts on all these things, and there’s not gonna be a tool out there
393 00:46:23.200 ⇒ 00:46:33.809 Uttam: that survives without that. And… but at this moment, like, it’s… I don’t know, it’s just not something we can do. And so, for me, I’m… I just want to partner with a tool where we can bring
394 00:46:33.890 ⇒ 00:46:35.930 Uttam: A lot of that rich context.
395 00:46:36.030 ⇒ 00:46:52.189 Uttam: and then still get the best-in-class text-to-SQL retrieval, the analysis piece that’s really hard, and that’s what’s, you know, I think that’s where we develop, like, an incredible solution. Yeah.
396 00:46:52.580 ⇒ 00:47:07.379 Clint Dunn: Yeah, I mean, I’ve even talked about, like, this is way longer term and, like, totally off the wall, but, like, I’ve even talked about, like, injecting your brain, like, Brainforge’s brain and POV into our agent, and making that, like, its own, like, right?
397 00:47:07.380 ⇒ 00:47:08.050 Uttam: Definitely.
398 00:47:08.240 ⇒ 00:47:16.470 Clint Dunn: processes and, like, a POV on how things should be done, and, like, can we sell that as a brain that people can purchase within the product?
399 00:47:17.220 ⇒ 00:47:23.980 Uttam: Yeah, I mean, dude, like, at minimum, if you wanna… I mean, you could test this with our company data.
400 00:47:24.090 ⇒ 00:47:26.990 Uttam: But, like, at minimum, yeah, like, I think…
401 00:47:27.330 ⇒ 00:47:33.739 Uttam: I mean, again, like, our approach for everything is do the analysis and give a recommendation, because.
402 00:47:33.960 ⇒ 00:47:34.450 Clint Dunn: Yeah.
403 00:47:34.450 ⇒ 00:47:48.259 Uttam: for the most part, people, they’re like… this is the first time they’ve seen something broken down, like, oh, you guys did a… we did a segmentation analysis. But they’re not just like… we’re not just like, here’s the segmentation analysis, we’re like, here’s this, and this is probably what you should do.
404 00:47:48.260 ⇒ 00:47:49.370 Clint Dunn: Yeah. And…
405 00:47:49.400 ⇒ 00:48:04.769 Uttam: And both of those things really, really matter, but really, the recommendation is our… is the true alpha that we bring. And the analysis piece is where we’re already using Cursor, like, Amber on our team uses Cursor already, and we cooked right in some other duck, and our…
406 00:48:04.770 ⇒ 00:48:10.559 Uttam: issuing SQL and building analyses, she would just use your tool to start transitioning
407 00:48:10.670 ⇒ 00:48:25.170 Uttam: that over, and then… and then also, yeah, we would just… I would like… but the benefit is, we’re starting to write Markdown files with all the context, like, what was our original project plan? Yeah. What were their company goals.
408 00:48:25.190 ⇒ 00:48:31.059 Uttam: We record every meeting with them that we do. Like, we have all this stuff sitting, and.
409 00:48:31.060 ⇒ 00:48:31.400 Clint Dunn: Yup.
410 00:48:31.400 ⇒ 00:48:32.650 Uttam: Yeah, so…
411 00:48:32.800 ⇒ 00:48:46.119 Uttam: that would be really great to have, and again, I think if you can do it with us, then you could totally pitch this to companies, like, bring your own context. Yeah. That’s gonna take your stuff to the moon. Otherwise, like, the Omni product, they’re not… they don’t have that yet.
412 00:48:46.370 ⇒ 00:48:54.600 Uttam: maybe they’ll develop it, but I think that’s gonna be the next piece, is, like, going from these unstructured data to making it referenceable, you know.
413 00:48:54.600 ⇒ 00:49:02.379 Clint Dunn: Yeah, yeah, I think, you know, I think if they even really take that seriously, like, they’ll still miss this, like, third-party information that we can pull.
414 00:49:02.380 ⇒ 00:49:03.050 Uttam: Yes.
415 00:49:03.180 ⇒ 00:49:05.159 Clint Dunn: So, yeah, no.
416 00:49:05.160 ⇒ 00:49:13.279 Uttam: No, and they’re so wide, dude, there’s, like, there’s no way they do this. And also, again, I think the dashboard is a dying sport.
417 00:49:13.390 ⇒ 00:49:22.790 Uttam: Like, all in all, like, I think there’s probably gonna be some people that still use a dashboard, but, frankly, like, someone… an AI agent’s gonna tell you the number.
418 00:49:23.150 ⇒ 00:49:25.429 Uttam: Automatically, or on demand?
419 00:49:25.640 ⇒ 00:49:26.490 Clint Dunn: Right, yeah.
420 00:49:27.020 ⇒ 00:49:27.570 Uttam: Aye.
421 00:49:27.570 ⇒ 00:49:28.080 Clint Dunn: I mean…
422 00:49:28.080 ⇒ 00:49:30.439 Uttam: You know, so that’s how I feel things are going.
423 00:49:31.080 ⇒ 00:49:37.460 Clint Dunn: I… yeah, I mean, we’re totally aligned, like I said at the beginning of the call, like, we see the world pretty similarly, but .
424 00:49:37.460 ⇒ 00:49:49.209 Uttam: Well, dude, because I all… look, I tell you, out of all of our clients, I mean, maybe Eden is the only one where the dashboards are really… but we end up… we end up writing the paper every week about what’s going on anyways.
425 00:49:49.600 ⇒ 00:49:50.000 Clint Dunn: No.
426 00:49:50.000 ⇒ 00:49:52.899 Uttam: And… it’s…
427 00:49:53.330 ⇒ 00:50:06.889 Uttam: Yeah, I don’t know, and I think these guys kind of, like, don’t… and when I asked them, like, hey, can I just get your MCP, and not, like, your whole product? And, like, can you… can I use that? Like, can we structure our deal? They’re not, like, that interested in that. I’m like, okay. Yeah, yeah.
428 00:50:06.920 ⇒ 00:50:15.880 Clint Dunn: Yeah, I mean, I think what you said earlier is, like, we really strongly believe in this, like, triangulation of an optimal decision, and, like.
429 00:50:15.880 ⇒ 00:50:16.450 Uttam: Yeah.
430 00:50:16.450 ⇒ 00:50:39.499 Clint Dunn: we… we have to bring a lot of our own context and third-party information in, but, like, the really rich stuff is the conversations that are happening within the company and with their agencies and things like that. And so, like, if we can pull that in, then we actually can… can understand, like, what’s going on. The agent can sound and be just 50 times smarter. I think it’s…
431 00:50:39.540 ⇒ 00:50:55.660 Clint Dunn: preposterous that a lot of these agents are just, like, reporting on numbers, because, like, the obvious next step with AI is, like, it needs to actually integrate into decisions, it needs to be more autonomous, and it needs to, like, start to take over some workflows from people.
432 00:50:56.690 ⇒ 00:51:12.330 Uttam: Yeah, I mean, for us, we’re sort of recording it as, like, decision intelligence broadly, and for Brainforge, like, what we’re… I mean, like, I don’t know what else it is, but, like, basically we’re just supporting decisions, right? Like, and data or AI aside, like, that’s our job, and…
433 00:51:12.610 ⇒ 00:51:25.860 Uttam: what… what for us, like, we partner with all these vendors, but for the most part, we’re just structuring a lot of data and then putting it in somewhere. And then we have… our… our edge is gonna be on doing that, and then the analysis. Like, everything in the middle, I think, is, like…
434 00:51:26.610 ⇒ 00:51:29.270 Uttam: I don’t know, like, we’re gonna sort of commoditize a lot of that.
435 00:51:30.000 ⇒ 00:51:35.430 Clint Dunn: I, I didn’t need to do my sales deck, because you’re… you’re just, like, pitching my own slides to me.
436 00:51:35.430 ⇒ 00:51:39.169 Uttam: No, I mean, yeah, we’ve arrived at the same conclusion.
437 00:51:39.170 ⇒ 00:51:55.160 Clint Dunn: Yeah, yeah. Real quick, I’ll show you a couple of these. Like, our, like, initial kind of, like, sales conversation is, like, businesses are decision engines, and you need to get through this, like, feedback cycle faster, and you need to have better hypotheses from the start, and that’s, like, the faster, better decisions.
438 00:51:55.160 ⇒ 00:52:08.289 Clint Dunn: And, like, I actually took this from an earnings call from a customer that we were working with, and these were, like, just a handful of questions that were basically asked during the, you know, 45-minute earnings call.
439 00:52:08.290 ⇒ 00:52:19.999 Clint Dunn: And so that’s kind of where we get to on the optimal decision, is, like, businesses are just decision engines, and that’s very similar to what you were just saying, but that’s, like, our kind of overarching argument.
440 00:52:20.620 ⇒ 00:52:21.889 Uttam: Okay, okay, great.
441 00:52:21.890 ⇒ 00:52:22.550 Clint Dunn: Yeah.
442 00:52:22.890 ⇒ 00:52:35.659 Uttam: So, I mean, yeah, I think… I think we’re… this is, like… I mean, I think we’re really aligned. I actually also… maybe I can ask Patrick, but I also… we did some research, and I’m kind of wondering what the backend looks like, but I have some interesting…
443 00:52:35.920 ⇒ 00:52:42.679 Uttam: Tools to suggest if you guys are gonna start going into this route of trying to pull information from
444 00:52:43.040 ⇒ 00:52:48.670 Uttam: structure, semi-structured data, like documents and things like that. I have a really good company that we partner with.
445 00:52:48.790 ⇒ 00:52:52.599 Uttam: They all should consider. That’s, like, really probably the best people.
446 00:52:52.600 ⇒ 00:52:54.450 Clint Dunn: Okay, yeah, send it to us.
447 00:52:54.490 ⇒ 00:52:55.490 Uttam: Yeah, and .
448 00:52:55.490 ⇒ 00:52:59.100 Clint Dunn: We have a lot of the infrastructure out, but we don’t have a lot of the…
449 00:52:59.240 ⇒ 00:53:14.690 Clint Dunn: We haven’t built the connectors and things like that yet. So, like, I think if you guys, like, basically dropped some documents to us, we probably could integrate it pretty quickly, but we don’t have a more, like, product-oriented way to integrate it. That’s, like…
450 00:53:14.690 ⇒ 00:53:16.610 Uttam: Patrick, yeah, I’ll send you guys thing.
451 00:53:16.610 ⇒ 00:53:18.999 Clint Dunn: Yeah. They… these guys will… you, like…
452 00:53:19.290 ⇒ 00:53:37.820 Uttam: you could probably ship this document thing on top of their SDK, like, pretty fast. Okay. And it’s a really sick product, I’ll send it to you. That’s cool. Okay, so then let’s… I mean, I would love to, like, we should just try it. We should maybe… we can regroup on who to pick. Like, I don’t know, Awash, and then, like, Eden could be really great to try to…
453 00:53:38.370 ⇒ 00:53:39.659 Uttam: Do this for.
454 00:53:40.250 ⇒ 00:53:43.299 Uttam: I feel like we’re probably the closest
455 00:53:43.520 ⇒ 00:53:49.149 Uttam: them, and I think it would be something that… they’ve also asked us for this, I think, in the past, so…
456 00:53:49.150 ⇒ 00:53:49.810 Clint Dunn: Yeah.
457 00:53:50.380 ⇒ 00:53:55.520 Clint Dunn: Yeah, and we, like I said, you know, we’ll work really closely with you guys going into next year on, like, a…
458 00:53:55.830 ⇒ 00:54:05.400 Clint Dunn: Full warehouse integration and things like that, where, where we can actually, like, use the data and more of the context that you guys are doing with your own modeling work.
459 00:54:06.200 ⇒ 00:54:10.650 Uttam: I mean, yeah, and then also, if you… if we can just help build it alongside you, like, we’ll just do that.
460 00:54:10.650 ⇒ 00:54:11.620 Clint Dunn: Yeah, 100%.
461 00:54:11.620 ⇒ 00:54:14.860 Uttam: Because we have it all at S3, for the most part, I think.
462 00:54:15.220 ⇒ 00:54:22.400 Clint Dunn: Yeah, yeah, that’s, like, super easy for us, and it, you know, you… you might work well with my co-founder, I’m not sure.
463 00:54:25.140 ⇒ 00:54:32.210 Uttam: The AI team knows, I don’t know if anyone on this call knows him, but yeah, he… he needs someone.
464 00:54:34.600 ⇒ 00:54:40.349 Uttam: I finished a look back at that, so we started using it.
465 00:54:40.850 ⇒ 00:54:56.229 Clint Dunn: Yeah, alright, I gotta run, I gotta hard stop right now, but it was great chatting. Obviously, we have the Slack channel, so I’ll look out for any stuff there, but, yeah, just ping me when you guys are ready, and and we can kind of run through a trial for you.
466 00:54:57.070 ⇒ 00:54:58.439 Uttam: Okay, okay, perfect.
467 00:54:58.580 ⇒ 00:55:02.300 Clint Dunn: Alright, thanks, dude. Yup, talk to you. See y’all, nice to meet you.
468 00:55:03.370 ⇒ 00:55:04.010 Samuel Roberts: Thank you.
469 00:55:04.010 ⇒ 00:55:04.570 Demilade Agboola: Bye.