Meeting Title: Omni Data Modeling Handoff Sync Date: 2026-04-30 Meeting participants: Scratchpad Notetaker, Greg Stoutenburg, Nandika Jhunjhunwala, Caitlyn Vaughn, Advait Nandakumar Menon
WEBVTT
1 00:00:27.070 ⇒ 00:00:28.769 Greg Stoutenburg: Hey, Nautica, how’s it going today?
2 00:00:32.240 ⇒ 00:00:33.740 Nandika Jhunjhunwala: Hello, how are you?
3 00:00:33.740 ⇒ 00:00:35.790 Greg Stoutenburg: There we go, I’m great, how are you?
4 00:00:36.330 ⇒ 00:00:39.140 Nandika Jhunjhunwala: Good! Yeah, so I’m just setting up my…
5 00:00:39.140 ⇒ 00:00:39.750 Greg Stoutenburg: Yep.
6 00:00:39.960 ⇒ 00:00:42.120 Nandika Jhunjhunwala: Meeting or session.
7 00:00:42.120 ⇒ 00:00:44.230 Greg Stoutenburg: We’re all doing the same thing. No problem.
8 00:00:44.230 ⇒ 00:00:44.870 Nandika Jhunjhunwala: Yes.
9 00:00:47.320 ⇒ 00:00:48.360 Greg Stoutenburg: Okay.
10 00:00:49.460 ⇒ 00:00:52.049 Greg Stoutenburg: Let’s see, so I saw Lev Declined…
11 00:00:52.890 ⇒ 00:00:56.489 Greg Stoutenburg: Okay, so it looks like we’re waiting for Caitlin…
12 00:00:56.490 ⇒ 00:00:56.900 Nandika Jhunjhunwala: Yes.
13 00:00:56.900 ⇒ 00:00:57.480 Greg Stoutenburg: Awesome.
14 00:00:59.090 ⇒ 00:01:01.919 Greg Stoutenburg: And I think that will be the crew.
15 00:01:06.040 ⇒ 00:01:07.290 Caitlyn Vaughn: Hello!
16 00:01:07.290 ⇒ 00:01:08.530 Greg Stoutenburg: Hey, Caitlin, how’s it going?
17 00:01:08.530 ⇒ 00:01:10.090 Caitlyn Vaughn: Good, how are you?
18 00:01:10.090 ⇒ 00:01:11.690 Greg Stoutenburg: I am doing well.
19 00:01:12.060 ⇒ 00:01:16.199 Greg Stoutenburg: Bum… Save. This and following events.
20 00:01:17.350 ⇒ 00:01:22.810 Greg Stoutenburg: Done. Okay, cool. Yeah, I think we’re just waiting for Tom. Let me ping him real quick, and just confirm.
21 00:01:23.660 ⇒ 00:01:27.919 Greg Stoutenburg: Sometimes it just says he’s on, but it isn’t.
22 00:01:29.010 ⇒ 00:01:29.950 Caitlyn Vaughn: been caught.
23 00:01:34.570 ⇒ 00:01:36.609 Greg Stoutenburg: So the zombie is not on… okay.
24 00:01:37.240 ⇒ 00:01:42.210 Greg Stoutenburg: Yeah, well, just wait a second. How’s the weather today?
25 00:01:42.950 ⇒ 00:01:45.010 Greg Stoutenburg: I got some sun, so I want to talk about it.
26 00:01:45.160 ⇒ 00:01:45.640 Caitlyn Vaughn: You do.
27 00:01:45.640 ⇒ 00:01:46.860 Greg Stoutenburg: Yeah, yeah.
28 00:01:46.860 ⇒ 00:01:47.740 Nandika Jhunjhunwala: Nice.
29 00:01:47.980 ⇒ 00:01:49.850 Caitlyn Vaughn: You’ve sent your rain down here, honestly.
30 00:01:51.350 ⇒ 00:01:52.130 Greg Stoutenburg: Sorry.
31 00:01:52.130 ⇒ 00:01:54.950 Nandika Jhunjhunwala: No, same. It’s gloomy today in New York.
32 00:01:55.870 ⇒ 00:01:57.750 Caitlyn Vaughn: Is it finally warm, or no?
33 00:01:58.330 ⇒ 00:02:00.820 Nandika Jhunjhunwala: No. It’s okay. It’s like…
34 00:02:00.820 ⇒ 00:02:01.560 Caitlyn Vaughn: Yeah.
35 00:02:01.950 ⇒ 00:02:04.280 Nandika Jhunjhunwala: Late 40s, early 50s, sometimes they.
36 00:02:04.280 ⇒ 00:02:04.770 Caitlyn Vaughn: 15.
37 00:02:04.770 ⇒ 00:02:05.840 Nandika Jhunjhunwala: Yeah.
38 00:02:06.620 ⇒ 00:02:15.440 Caitlyn Vaughn: Last time I was in New York, like, a couple weeks ago, I was so mad that I had to, like, re-pull out all of my winter clothes.
39 00:02:15.470 ⇒ 00:02:30.670 Caitlyn Vaughn: And wear, like, sweaters again, and then I packed all winter, because it was cold, and then it shot up to, like, 85, and then I was just sweating, and our office has to be, like, winterized or summarized, because they’re all really old buildings in New York. Right.
40 00:02:30.670 ⇒ 00:02:31.230 Greg Stoutenburg: Yep.
41 00:02:31.230 ⇒ 00:02:38.470 Caitlyn Vaughn: It’s, like, all centralized heating, and the heat… the, like, heaters were still on for, like, the entire 85 degrees, so it was so hot in there.
42 00:02:38.470 ⇒ 00:02:47.449 Greg Stoutenburg: No good. Can’t have that. Yeah. Yeah, it was… there was one… there were, like, one or two really hot days, like, a week ago. And then, yeah, and then, like, winter came back.
43 00:02:47.650 ⇒ 00:02:48.020 Caitlyn Vaughn: Yeah.
44 00:02:48.140 ⇒ 00:02:54.800 Greg Stoutenburg: But yeah, that’s season. People say, oh, I love the seasons. I mean, like, not really. Like, I don’t like bugs. I like… so I like it when they show bugs.
45 00:02:54.800 ⇒ 00:02:55.210 Caitlyn Vaughn: Yeah.
46 00:02:55.210 ⇒ 00:02:59.060 Greg Stoutenburg: And, I like it when it’s good, like, I don’t know.
47 00:02:59.490 ⇒ 00:03:00.860 Greg Stoutenburg: Basically, 70.
48 00:03:01.140 ⇒ 00:03:01.830 Caitlyn Vaughn: That’s it.
49 00:03:01.830 ⇒ 00:03:08.699 Greg Stoutenburg: We’re gonna hold it there, and then ski season. Those are the only things I really need. Or 90 if I’m gonna be in a pool. That’s it.
50 00:03:08.700 ⇒ 00:03:26.770 Caitlyn Vaughn: Yeah, no, that’s unfair. I think you also just, like, start getting used to it, like, it’s, like, 100 here for 4 months, and the first, like, week of it is really miserable, and then you kind of just, like, start leaning into it. Like, you’re gonna lose all of your, you know, extra water weight, you’re gonna get really skinny, your skin’s gonna get nice.
51 00:03:27.320 ⇒ 00:03:28.709 Caitlyn Vaughn: Kind of crazy, but like…
52 00:03:28.710 ⇒ 00:03:29.710 Greg Stoutenburg: Trade-up.
53 00:03:29.710 ⇒ 00:03:30.370 Caitlyn Vaughn: off, you know?
54 00:03:30.370 ⇒ 00:03:32.560 Greg Stoutenburg: Where are you located, then? I thought you were in New York.
55 00:03:32.560 ⇒ 00:03:34.109 Caitlyn Vaughn: I’m in Austin, Texas.
56 00:03:34.110 ⇒ 00:03:34.890 Greg Stoutenburg: Oh!
57 00:03:35.610 ⇒ 00:03:41.940 Greg Stoutenburg: That explains some things. Tom’s like, oh yeah, you know, I saw Caitlin, I thought, like, I don’t know how, but okay.
58 00:03:41.940 ⇒ 00:03:42.339 Caitlyn Vaughn: I mean, a video.
59 00:03:42.340 ⇒ 00:03:43.670 Greg Stoutenburg: call.
60 00:03:43.670 ⇒ 00:03:47.510 Caitlyn Vaughn: Yeah, Utam, that’s how I know Utam. I know him from in person in Austin.
61 00:03:47.510 ⇒ 00:03:53.529 Greg Stoutenburg: Okay, cool. Well, that’s too bad that I was probably told that. That’s too bad that this has been recorded now, and
62 00:03:53.950 ⇒ 00:03:54.599 Greg Stoutenburg: that I… that I.
63 00:03:54.600 ⇒ 00:03:54.940 Caitlyn Vaughn: I love that.
64 00:03:54.940 ⇒ 00:03:58.069 Greg Stoutenburg: Oh, no, I’m just kidding, like, that I had to be told twice.
65 00:03:58.070 ⇒ 00:03:58.680 Caitlyn Vaughn: Okay.
66 00:03:59.010 ⇒ 00:04:02.819 Caitlyn Vaughn: No, you’re all good. Yeah, we started agencies at the same time, so I met Utah.
67 00:04:02.820 ⇒ 00:04:03.469 Greg Stoutenburg: Oh, neat.
68 00:04:03.720 ⇒ 00:04:15.390 Caitlyn Vaughn: maybe 3 years ago, maybe longer, 4 years ago, at, like, a VC dinner, and I got sat next to him, and then it ended up being, like, he’s the coolest, he’s the best.
69 00:04:15.390 ⇒ 00:04:21.860 Greg Stoutenburg: Yeah, yeah. I’ve, still not met in person, but from photos, it looks like he’s also quite tall.
70 00:04:21.860 ⇒ 00:04:22.869 Caitlyn Vaughn: It’s so tall!
71 00:04:22.870 ⇒ 00:04:23.200 Greg Stoutenburg: Like.
72 00:04:23.200 ⇒ 00:04:24.640 Caitlyn Vaughn: It was really surprising.
73 00:04:24.640 ⇒ 00:04:25.020 Greg Stoutenburg: Yeah.
74 00:04:25.400 ⇒ 00:04:25.930 Greg Stoutenburg: video.
75 00:04:25.930 ⇒ 00:04:26.640 Nandika Jhunjhunwala: Could not have gone.
76 00:04:26.980 ⇒ 00:04:27.860 Nandika Jhunjhunwala: either.
77 00:04:27.860 ⇒ 00:04:28.260 Greg Stoutenburg: No.
78 00:04:28.310 ⇒ 00:04:29.700 Caitlyn Vaughn: Yeah, it was probably, like…
79 00:04:29.700 ⇒ 00:04:30.950 Greg Stoutenburg: Alright, what’s everybody here?
80 00:04:30.950 ⇒ 00:04:33.690 Caitlyn Vaughn: I think he’s probably, like, 6’3”. He’s very, very tall.
81 00:04:34.060 ⇒ 00:04:40.550 Greg Stoutenburg: Yeah, I gotta prepare for that, yeah, because I’ll, I’ll finally get to meet him in person in DC in two weeks.
82 00:04:40.550 ⇒ 00:04:40.940 Caitlyn Vaughn: Huh?
83 00:04:40.940 ⇒ 00:04:54.149 Greg Stoutenburg: And then, last, I guess, I don’t know, last thing on that, and then I’ll actually, like, do this meeting, is I had a… there’s another company I worked at for, like, a year and a half before I met people in person, and I’m about 5’10”, so, you know, not short, but…
84 00:04:54.150 ⇒ 00:04:56.030 Caitlyn Vaughn: I was taller than I thought you were, too!
85 00:04:56.030 ⇒ 00:05:07.290 Greg Stoutenburg: Yeah, how about it? Yeah, well, it’s because of the standing desk, I was just bringing it with me, you know? Yeah, yeah. But I went and met the company in person, and I was like, I cannot believe it. I’m like… I’m, like, by 3 inches the shortest man in the company.
86 00:05:07.290 ⇒ 00:05:08.040 Caitlyn Vaughn: What?
87 00:05:08.040 ⇒ 00:05:13.469 Greg Stoutenburg: I was like… like, I didn’t know I was working with all giants, and it was all of them, so that was weird.
88 00:05:13.470 ⇒ 00:05:14.130 Caitlyn Vaughn: Interesting.
89 00:05:14.130 ⇒ 00:05:16.920 Greg Stoutenburg: Yeah, yeah, British, British giants.
90 00:05:16.920 ⇒ 00:05:25.960 Caitlyn Vaughn: all of the tech companies I’ve worked at have had very tall men, and it’s all men, usually, and they’re just… that is so true. It’s very surprising, I think.
91 00:05:25.960 ⇒ 00:05:26.340 Greg Stoutenburg: Yeah.
92 00:05:26.340 ⇒ 00:05:27.809 Nandika Jhunjhunwala: Even a deep walk, I think every…
93 00:05:27.810 ⇒ 00:05:28.310 Caitlyn Vaughn: at least time.
94 00:05:28.740 ⇒ 00:05:30.339 Caitlyn Vaughn: Everyone’s really tall!
95 00:05:30.340 ⇒ 00:05:31.200 Nandika Jhunjhunwala: Wow.
96 00:05:31.200 ⇒ 00:05:33.569 Caitlyn Vaughn: Yeah, I mean, Nante, you’re pretty tall, right?
97 00:05:33.900 ⇒ 00:05:35.249 Nandika Jhunjhunwala: Yeah, boy, I’m.
98 00:05:35.250 ⇒ 00:05:36.089 Caitlyn Vaughn: Like, 5 seconds.
99 00:05:36.090 ⇒ 00:05:37.300 Nandika Jhunjhunwala: 7, yeah.
100 00:05:37.300 ⇒ 00:05:40.649 Greg Stoutenburg: Yeah, alright, okay, alright, respect, yeah, respect.
101 00:05:40.650 ⇒ 00:05:42.120 Nandika Jhunjhunwala: We’re both on the same time.
102 00:05:42.120 ⇒ 00:05:47.169 Caitlyn Vaughn: But then, every time I’m in the office and I wear, like, heels or anything, I feel like a giant.
103 00:05:48.820 ⇒ 00:05:56.669 Caitlyn Vaughn: We’re, like, 5’10”, walking around the office, and, like, for most of the men, it’s fine, but then there’s, like, some people, like Laura, she’s, like.
104 00:05:56.950 ⇒ 00:06:02.210 Caitlyn Vaughn: 5’2”. She’s so big, so I just, like, tower over her.
105 00:06:02.210 ⇒ 00:06:02.780 Nandika Jhunjhunwala: 15.
106 00:06:02.780 ⇒ 00:06:03.810 Caitlyn Vaughn: Right, I’m like…
107 00:06:03.940 ⇒ 00:06:05.900 Greg Stoutenburg: I’m the boss now.
108 00:06:06.730 ⇒ 00:06:23.180 Greg Stoutenburg: Okay, good. Cool. Alright, well, let’s, let’s dive in. So, I’ll do… I’ll just… I’m just gonna do a sort of abbreviated take here, because, you know, the focus is really on the handoff and QAing the things that are already in flight for, you know, for the next two weeks, and we’re just… we’re making progress on all those things, so…
109 00:06:23.350 ⇒ 00:06:26.750 Greg Stoutenburg: Let’s just, let’s go, why is it not?
110 00:06:29.080 ⇒ 00:06:30.540 Greg Stoutenburg: Seven last time, didn’t it?
111 00:06:31.450 ⇒ 00:06:33.850 Greg Stoutenburg: I’m hitting the slideshow button, and it’s not doing it.
112 00:06:33.850 ⇒ 00:06:34.590 Caitlyn Vaughn: Fresh.
113 00:06:41.180 ⇒ 00:06:42.589 Caitlyn Vaughn: Hmm, that’s fine.
114 00:06:42.750 ⇒ 00:06:46.289 Greg Stoutenburg: Well, alright. Yeah. You know, lean in.
115 00:06:46.460 ⇒ 00:06:47.060 Caitlyn Vaughn: Yep.
116 00:06:47.750 ⇒ 00:06:48.620 Greg Stoutenburg: Okay.
117 00:06:49.680 ⇒ 00:06:59.730 Greg Stoutenburg: So, okay, pivoted handoff mode shared that roadmap, which is represented here. We’ll come back to this in a moment. The archive…
118 00:06:59.870 ⇒ 00:07:12.889 Greg Stoutenburg: for old topics that expose raw data that you’re able to interact with through Blobby, that PR has been merged, and the dashboards that were built on them have been deprecated, they’ve been taken out of view. So…
119 00:07:13.090 ⇒ 00:07:24.290 Greg Stoutenburg: there’s no… there are no dashboards now that are built on that that anyone’s gonna stumble into, and any questions asked through Blobby will be using the data sources that are fresh, and the ones that have been curated.
120 00:07:24.290 ⇒ 00:07:28.740 Caitlyn Vaughn: Really, so to clarify, it was the dashboards built on the raw data?
121 00:07:29.060 ⇒ 00:07:29.690 Greg Stoutenburg: Yes.
122 00:07:29.690 ⇒ 00:07:30.400 Caitlyn Vaughn: Okay, amazing.
123 00:07:30.400 ⇒ 00:07:46.869 Greg Stoutenburg: because the dashboards built on raw data before, before we had Polyatomic, before we had, Plane or Pylon, and so the way that it was explained to me is, they were like, yeah, Greg, these are, these were, like, pilot phase dashboards, basically, that were set up to introduce the team, and,
124 00:07:46.910 ⇒ 00:07:56.969 Greg Stoutenburg: But, since they were still in view, and so that’s how you’re able to see that. But the way that Blobby works is, even if we had just archived those, or put them in a folder, and just written on the folder, don’t look at these,
125 00:07:57.300 ⇒ 00:08:11.930 Greg Stoutenburg: The way that Blobby works is if you type in a query, and it identifies a particular topic as, yeah, exactly, as relevant to that, it’s gonna pull data from there. So, we had to, not only get rid of the dashboards, but then deprecate the topic, and so that won’t happen anymore.
126 00:08:12.110 ⇒ 00:08:30.820 Caitlyn Vaughn: Yeah, I guess it was my understanding, like, this week I’ve been getting as caught up as I can on Omni and data modeling and, you know, the full thing so that we can take it over, but it’s my understanding that for Omni and for Blobby, it’s only pulling from topics? So were there topics created from the raw data?
127 00:08:30.980 ⇒ 00:08:44.520 Greg Stoutenburg: Yes, there was a… there was a… there was at least one topic that was used in that initial pilot phase that had raw data as a source, and so that’s why the topic exposed it. Yep. Cool.
128 00:08:44.520 ⇒ 00:08:47.500 Nandika Jhunjhunwala: Is that the Salesforce one, I’m assuming?
129 00:08:48.370 ⇒ 00:08:51.440 Nandika Jhunjhunwala: It was from… that the topic was from raw data.
130 00:08:51.970 ⇒ 00:08:54.349 Greg Stoutenburg: Advay, can you speak to that? It was,
131 00:08:54.770 ⇒ 00:09:04.590 Greg Stoutenburg: Yeah. By the way, have we done the intro here? I can’t… I cannot recall. Okay, yeah, alright, thought so. But he wasn’t on the meeting, so that was the mistake I’d made. Yeah, Advay, can you speak to that?
132 00:09:05.840 ⇒ 00:09:08.240 Advait Nandakumar Menon: Can you please repeat that? What did you ask?
133 00:09:08.700 ⇒ 00:09:16.909 Nandika Jhunjhunwala: Yeah, so I think I was asking, I think Brett was mentioning that there was some topics created from raw data directly.
134 00:09:17.490 ⇒ 00:09:23.650 Nandika Jhunjhunwala: And that was deprecated data, maybe? And that was getting pulled into Omni when people were querying it.
135 00:09:24.050 ⇒ 00:09:28.179 Nandika Jhunjhunwala: Was that the Salesforce one, or was it a different topic?
136 00:09:28.760 ⇒ 00:09:46.779 Advait Nandakumar Menon: It was the Salesforce one, like, it was a raw export, and it was stale, and it was a one-time export that we pulled into Omni, and Blobby was basically referring to that topic and underlying views and tables to give us the answers, which is basically stale data at this point, so…
137 00:09:47.180 ⇒ 00:09:49.390 Advait Nandakumar Menon: Which is what we have removed and replicated.
138 00:09:51.060 ⇒ 00:09:58.210 Nandika Jhunjhunwala: The other question, I’m so sorry. That makes sense, thank you. I think Caitlin and I were chatting about this, but…
139 00:09:59.120 ⇒ 00:10:17.600 Nandika Jhunjhunwala: so they’re, like, flat tables and dimension tables, and I know some topics have, like, multiple dimension tables, and some of them are, like, bridge dimension tables that, like, join to dimensions, or, like, fact tables to dimension tables. But when we query Blobby, like, from what we were understanding, is that
140 00:10:17.840 ⇒ 00:10:33.490 Nandika Jhunjhunwala: it reads one dimension table, and that’s what it uses as its base to, like, join other fact tables, and please correct me if I’m wrong, but I’m assuming having multiple dimension tables in one topic can confuse AI? And so, if you have tables like
141 00:10:33.510 ⇒ 00:10:46.759 Nandika Jhunjhunwala: dimensions workflows by integration, and, like, dimension workflows by integration daily, that’s gonna be confusing for the AI to query off of, or are we, like, putting in guardrails against that, if I’m understanding correctly?
142 00:10:47.570 ⇒ 00:10:53.519 Advait Nandakumar Menon: So the way Topic is built right now is not on the fact and dimension tables, like, the raw fact.
143 00:10:53.520 ⇒ 00:10:54.000 Nandika Jhunjhunwala: I didn’t.
144 00:10:54.000 ⇒ 00:10:57.780 Advait Nandakumar Menon: It is the dbt modeling we have done. We have aggregated
145 00:10:57.920 ⇒ 00:11:15.209 Advait Nandakumar Menon: layers of business logic applied to the Salesforce tables and a couple of… whatever you mentioned, basically. So, the topic is built on top of those dbt tables, and the dbt basically has the logic behind all the joints and whatever.
146 00:11:15.350 ⇒ 00:11:29.979 Advait Nandakumar Menon: But to answer a question, yeah, it’s… you can join a couple of different fact and dimension tables within the topic, and you can enrich the topic with guardrails or specific instructions to Blobby or the AI to
147 00:11:30.070 ⇒ 00:11:39.900 Advait Nandakumar Menon: Maybe look at a particular query you want using sample queries, or any sort of instructions, basically, or context you want to give to the AI.
148 00:11:42.460 ⇒ 00:11:43.659 Advait Nandakumar Menon: Does that make sense?
149 00:11:44.050 ⇒ 00:11:46.280 Nandika Jhunjhunwala: Yeah, sorry, I’m a little confused, because…
150 00:11:46.420 ⇒ 00:11:49.880 Nandika Jhunjhunwala: If we have similar-sounding dimension tables that are
151 00:11:50.110 ⇒ 00:11:53.359 Nandika Jhunjhunwala: Aggregated at, like, daily level or, like, weekly level.
152 00:11:53.640 ⇒ 00:12:09.389 Nandika Jhunjhunwala: how should we query Bobby correctly so that it’s reading off the correct table? Because, like, while Caitlin and I have spent time diving into the data, not everybody will, and then we just want to make sure that it’s always going to be pulling from the correct table.
153 00:12:09.670 ⇒ 00:12:10.529 Nandika Jhunjhunwala: And not using.
154 00:12:10.530 ⇒ 00:12:11.040 Advait Nandakumar Menon: Yes.
155 00:12:11.040 ⇒ 00:12:13.170 Nandika Jhunjhunwala: different tailored words, like baseline query.
156 00:12:14.010 ⇒ 00:12:26.650 Advait Nandakumar Menon: Yeah, so I think one thing that even we can do is, like, update the topic to make sure we give explicit instruction to Bobby when to refer to the daily table, to the weekly table, or whatever, so that
157 00:12:26.770 ⇒ 00:12:45.400 Advait Nandakumar Menon: When the user asks something, Blobby is able to infer that automatically and direct it to it. But the other thing is also that when you’re asking Blobby some question, it’s always better to select the topic, that you’re going to use than letting Blobby decide which topic
158 00:12:45.550 ⇒ 00:12:51.210 Advait Nandakumar Menon: it takes up. Because, AI is, as you know, it’s…
159 00:12:51.290 ⇒ 00:13:06.510 Advait Nandakumar Menon: Experimental, it’s non-deterministic, it won’t give you the same answer always, so it’s better to select a topic once, and the topic should have explicit instructions, like the daily view, or the weekly view, or whatever, it needs to be updated with that.
160 00:13:06.930 ⇒ 00:13:12.000 Greg Stoutenburg: Have you seen that, Nandica? If you go to use Bobby, you can actually click here, and choose which…
161 00:13:12.000 ⇒ 00:13:13.750 Nandika Jhunjhunwala: Oh, I’m totally aware. Okay.
162 00:13:13.750 ⇒ 00:13:14.310 Greg Stoutenburg: Alright.
163 00:13:14.580 ⇒ 00:13:18.229 Nandika Jhunjhunwala: But, can you, sorry, can you click within customer success just for a second?
164 00:13:18.540 ⇒ 00:13:23.849 Nandika Jhunjhunwala: So if you click on Workflows Created by Table figure.
165 00:13:24.520 ⇒ 00:13:25.630 Greg Stoutenburg: I trigger it, yep.
166 00:13:25.630 ⇒ 00:13:31.669 Nandika Jhunjhunwala: Yeah, so for that table, too, I think that’s a topic, right, within customer success?
167 00:13:31.670 ⇒ 00:13:32.460 Greg Stoutenburg: Yeah, these are all topics.
168 00:13:32.460 ⇒ 00:13:32.930 Nandika Jhunjhunwala: the next day.
169 00:13:32.930 ⇒ 00:13:34.870 Greg Stoutenburg: Anything you select here is a topic, yep.
170 00:13:35.710 ⇒ 00:13:41.100 Nandika Jhunjhunwala: So is that a particular table, or is that a combination of tables within customer success? Like…
171 00:13:41.680 ⇒ 00:13:44.289 Nandika Jhunjhunwala: Those are a few tables, right? This is a subtopic.
172 00:13:45.610 ⇒ 00:13:49.340 Advait Nandakumar Menon: Customer success is a folder, it’s not a topic, it’s just a folder, so…
173 00:13:49.550 ⇒ 00:13:50.169 Nandika Jhunjhunwala: Got it.
174 00:13:50.170 ⇒ 00:13:59.130 Advait Nandakumar Menon: see within the customer success folder is the topics, and the topic can have one base view or table, and we can join in other tables as well to it.
175 00:14:00.090 ⇒ 00:14:01.160 Nandika Jhunjhunwala: within a topic.
176 00:14:01.800 ⇒ 00:14:02.640 Advait Nandakumar Menon: Yep, yep.
177 00:14:02.780 ⇒ 00:14:07.659 Greg Stoutenburg: Yes. Yep. A topic is just a curated join of tables with certain parameters.
178 00:14:08.830 ⇒ 00:14:15.229 Nandika Jhunjhunwala: And so, I think Caitlin and I were looking at this briefly, or like I was, and we were going back and forth as well.
179 00:14:15.840 ⇒ 00:14:30.020 Nandika Jhunjhunwala: trying to, like, resolve our confusion and question, but there’s, like, a few, like… I think there’s workflow created by Trigger, and there’s workflow created by Trigger Daily, and there’s, like, a few other, like, variations, in that topic, as far as I remember.
180 00:14:30.100 ⇒ 00:14:43.270 Nandika Jhunjhunwala: So, when someone within customer success is querying it and asking for, like, different aggregations, like, how are we gonna make sure, like, what table it’s gonna read from and if it’s going to be accurate? Like, that’s more so my question.
181 00:14:43.850 ⇒ 00:14:44.520 Nandika Jhunjhunwala: I don’t know if that.
182 00:14:44.520 ⇒ 00:14:45.170 Advait Nandakumar Menon: I’m so…
183 00:14:45.170 ⇒ 00:14:47.149 Nandika Jhunjhunwala: Off, like, a different tangent here, or…
184 00:14:47.350 ⇒ 00:14:57.650 Caitlyn Vaughn: Yeah, so let’s maybe come back to this at the end, because I think you and I have a ton of questions about this, and we could probably back into, like, a wider framing of how we want this modeling to look, and then…
185 00:14:57.850 ⇒ 00:14:58.480 Nandika Jhunjhunwala: Okay.
186 00:14:58.480 ⇒ 00:15:01.170 Caitlyn Vaughn: Those questions will probably make a little bit more sense.
187 00:15:01.170 ⇒ 00:15:01.860 Nandika Jhunjhunwala: Sorry.
188 00:15:02.260 ⇒ 00:15:07.010 Caitlyn Vaughn: That’s okay, but Gray, is there anything else on the presentation side that we should be looking at?
189 00:15:07.320 ⇒ 00:15:23.190 Greg Stoutenburg: So, let’s just, I mean, basically, let’s just move forward. I can do all these updates and things, error recon is done, BDR dashboard is, in progress. There’s a PR that’s under review right now that’s needed for a couple of the pieces of feedback. The rest has already been, been implemented.
190 00:15:24.610 ⇒ 00:15:32.829 Caitlyn Vaughn: Okay, so for the BDR dashboard, I did get a question about that today. So, you said that there’s, like, a few last things to finish off with that?
191 00:15:32.870 ⇒ 00:15:45.320 Greg Stoutenburg: There’s one, yeah, it’s nearly all actioned. There’s just one PR that has to be merged, because there’s just a little bit of modeling work that had to be done in order to accommodate it. It wasn’t just, you know, visual.
192 00:15:45.970 ⇒ 00:15:47.690 Caitlyn Vaughn: Okay, what is it exactly?
193 00:15:48.530 ⇒ 00:15:50.379 Greg Stoutenburg: Aude, can you speak to that? Because you made the PR.
194 00:15:50.380 ⇒ 00:15:57.380 Advait Nandakumar Menon: The one, Pierre, you’re referring to the CS reporting and enablement dashboard, Greg.
195 00:15:57.380 ⇒ 00:15:59.800 Greg Stoutenburg: I thought you had one for the BDR dashboard as well.
196 00:16:00.160 ⇒ 00:16:07.750 Advait Nandakumar Menon: The BDR dashboard, couple of the tickets need some modeling changes, which, Demi is looking into, and we need to create PS for that.
197 00:16:07.750 ⇒ 00:16:16.010 Greg Stoutenburg: Okay, alright, so that PR is not in, my mistake. The customer reporting and enablement one, that’s the one that had a PR required, and it is in review.
198 00:16:16.360 ⇒ 00:16:17.010 Caitlyn Vaughn: Okay, cool.
199 00:16:17.010 ⇒ 00:16:17.330 Advait Nandakumar Menon: Yeah.
200 00:16:17.330 ⇒ 00:16:22.049 Caitlyn Vaughn: the BDR dashboard still… the PRs still need to be created.
201 00:16:23.820 ⇒ 00:16:24.530 Greg Stoutenburg: Yes.
202 00:16:24.850 ⇒ 00:16:25.840 Advait Nandakumar Menon: Yes, yeah.
203 00:16:26.330 ⇒ 00:16:34.170 Caitlyn Vaughn: Okay, would love to have that dashboard wrapped up by tomorrow. I think this has been going on for… for a while, so…
204 00:16:36.000 ⇒ 00:16:40.100 Greg Stoutenburg: Yeah, this one, I think we got… well, I think we did the review of this one last Wednesday?
205 00:16:41.120 ⇒ 00:16:42.220 Greg Stoutenburg: I think? Yeah.
206 00:16:42.370 ⇒ 00:16:43.200 Greg Stoutenburg: Okay.
207 00:16:43.520 ⇒ 00:16:54.610 Greg Stoutenburg: All right, heard. Yeah, we’ll… we’ll do that, and, being able to… being able to bring in another team member for the DBT training is, is helping to offset some of the workload for that.
208 00:16:55.850 ⇒ 00:17:01.250 Greg Stoutenburg: Okay, just discussed… Most of those…
209 00:17:02.850 ⇒ 00:17:27.200 Greg Stoutenburg: For ARR, future discrepancies can be controlled in Salesforce. So, the issue is that we had made modeling decisions so that there were, you know, calculated fields and things like that, so that what showed up is an ARR contribution dependent upon status plus some logic. The logic has effectively been removed, the logic is now just default to Salesforce, and so if there’s a discrepancy around ARR,
210 00:17:27.200 ⇒ 00:17:31.509 Greg Stoutenburg: you can just change the field in Salesforce by tweaking the renewal date. So…
211 00:17:31.510 ⇒ 00:17:33.670 Greg Stoutenburg: Something will be marked as churned.
212 00:17:33.760 ⇒ 00:17:45.799 Greg Stoutenburg: When something that’s closed one passes the renewal date. So then it’s, you know, if you want to move that date forward, then, that’ll… that’ll make it active and contribute that ARR.
213 00:17:49.680 ⇒ 00:18:05.100 Greg Stoutenburg: Demi’s reviewing where there were choices made around things like, you know, again, the calculated field from dbt that marked something as stale over the course of a 7-day period, splitting those into modifiable grains so it’s easier for you to manage those dashboards.
214 00:18:06.770 ⇒ 00:18:10.999 Greg Stoutenburg: Yeah, we just talked about this, Nautica and Lev are taking over the productivity dashboard.
215 00:18:12.230 ⇒ 00:18:14.760 Greg Stoutenburg: the… handoff?
216 00:18:15.430 ⇒ 00:18:21.299 Greg Stoutenburg: project, which is marked down here, puts us at being able to wrap it up
217 00:18:21.530 ⇒ 00:18:33.689 Greg Stoutenburg: by May 15th, which is there. And so, from here, we’ll then resume that marketing attribution dashboard work, and should be able to, you know, we’ll have everything delivered by the end of May.
218 00:18:34.140 ⇒ 00:18:39.830 Greg Stoutenburg: Yeah, customer productivity, well, I guess I can just mark that one off, because you’re taking that over.
219 00:18:40.020 ⇒ 00:18:42.880 Greg Stoutenburg: So, anyway, that’s where we are.
220 00:18:43.390 ⇒ 00:18:48.219 Greg Stoutenburg: Yep, that is… so, we’re on track there. Do you feel good about how that is going so far?
221 00:18:48.980 ⇒ 00:18:57.369 Caitlyn Vaughn: Yeah, I think, so we’ve actually been digging into the, like, how we would actually build this on our side, and, like.
222 00:18:57.620 ⇒ 00:19:07.620 Caitlyn Vaughn: what would make sense for us from, like, a modeling and strategy standpoint, since there’s kind of a lot going on. And honestly, before I started digging into this, the only real…
223 00:19:07.920 ⇒ 00:19:24.479 Caitlyn Vaughn: like, sense I had of things going right or wrong was if I used Blobby to try to create something, and then we weren’t able to create anything, right? This is basically the most downstream of the work that has been done. So, I had created that doc. Did you look through it by chance?
224 00:19:24.480 ⇒ 00:19:25.090 Greg Stoutenburg: Yes.
225 00:19:25.600 ⇒ 00:19:38.090 Caitlyn Vaughn: Okay, awesome. So, for that, is there anything inside of there that your team, like, disagrees with, or is, like, maybe a principle that shouldn’t exist that we’re saying should exist?
226 00:19:38.610 ⇒ 00:19:51.919 Greg Stoutenburg: I mean, we didn’t have any, like, major conceptual differences or anything like that. I mean, that seems sensible. We’re actioning it as well. Just today, Abdate and I were talking about, part of the…
227 00:19:51.920 ⇒ 00:20:03.909 Greg Stoutenburg: part of the preparation for the handoff for the Omni training will be just incorporating those pieces that you had for the audit checklist, and making sure that those are touched upon. So, yeah, I mean, that seems sensible. I don’t…
228 00:20:03.910 ⇒ 00:20:11.990 Greg Stoutenburg: Utam was going to write some comments, I think that he had written some comments in there, but, I mean, there wasn’t anything that sort of was, you know, a flag.
229 00:20:12.730 ⇒ 00:20:18.569 Greg Stoutenburg: Yeah, and I shared as well with Brian, who will be doing the DBT training with us in 40 minutes.
230 00:20:18.570 ⇒ 00:20:35.719 Caitlyn Vaughn: Okay, cool. Yeah, I think the only thing that Utam flagged was that… I think we’re doing a lot of the, like, defining in dbt versus in Omni, so I think we want that moved over to Omni just so that it’s, like, a clear… the semantic layer should always live,
231 00:20:35.850 ⇒ 00:20:50.410 Caitlyn Vaughn: in Omni and not in the dbt code. But there’s, like, there’s quite a few things as we started going through this, and, like, I don’t have all of the context, obviously, not as much as you guys have, because you’ve been working on this, but there’s some things that I’ve been noticing…
232 00:20:50.610 ⇒ 00:20:59.119 Caitlyn Vaughn: That I just don’t quite understand, so maybe either some explanation would be good, or maybe we could talk about it and fix it.
233 00:20:59.120 ⇒ 00:21:16.879 Greg Stoutenburg: Yeah. Yeah. Part of the handoff will be identifying those places where there was work done in dbt that, you know, limits you in this way. So, the documentation is in progress for that, and you’ll know what those are, and in places where it just shouldn’t be there, that’s something… that’s a sweep that Demi is doing right now.
234 00:21:17.030 ⇒ 00:21:36.819 Caitlyn Vaughn: Yep. Okay, so I’ll just start talking through some of the things that I’ve noticed. Okay. And maybe, Advent, this might be relevant to you, and you could probably advise as to, like, why these decisions were made. But basically, what I landed on is the, like, the Kimball style of having, like, the fact and dimension tables.
235 00:21:36.830 ⇒ 00:21:44.049 Caitlyn Vaughn: relationship is, like, the cleanest way that we can model our data that will also work for Blobby.
236 00:21:44.630 ⇒ 00:21:45.799 Caitlyn Vaughn: Is that correct?
237 00:21:47.500 ⇒ 00:21:53.389 Advait Nandakumar Menon: That’s correct, but there is this other, thought process as well, like…
238 00:21:53.590 ⇒ 00:22:03.269 Advait Nandakumar Menon: there’s this deep integration between dbt and Omni, like, whatever basic metrics or predefined metrics that you don’t see changing constantly, you can…
239 00:22:03.280 ⇒ 00:22:17.329 Advait Nandakumar Menon: have that model within DBT, and as the business changes, as your needs changes, as the data changes, you can do the same changes within Omni and push that back to dbt as well. So there is an integration wherein
240 00:22:17.430 ⇒ 00:22:33.029 Advait Nandakumar Menon: Basically, you can do the changes within dbt, push it to Omni, and do it the other way as well. So there’s a shared model connection between the both. That’s basically the main ideology behind Omni as well.
241 00:22:34.750 ⇒ 00:22:35.590 Caitlyn Vaughn: Okay.
242 00:22:35.720 ⇒ 00:22:37.160 Caitlyn Vaughn: Interesting.
243 00:22:37.900 ⇒ 00:22:38.799 Caitlyn Vaughn: I think
244 00:22:40.070 ⇒ 00:22:58.840 Caitlyn Vaughn: That’s probably fine. I think we would probably prefer it in Omni, just so that it’s all in one place, like, description, semantics, just from, like, a clean, we-know-where-everything-lives perspective, plus the dbt, the only thing that we can see is, like, the output in…
245 00:22:59.000 ⇒ 00:23:04.869 Caitlyn Vaughn: GitHub, right? We don’t have access to, like, your specific dbt account,
246 00:23:05.220 ⇒ 00:23:08.000 Caitlyn Vaughn: So Omni’s probably, like, the best place for that to live.
247 00:23:09.170 ⇒ 00:23:10.460 Advait Nandakumar Menon: Okay, let’s jump in.
248 00:23:10.460 ⇒ 00:23:18.509 Nandika Jhunjhunwala: Are you currently defining anything in Omni, or is that all in dbt?
249 00:23:19.730 ⇒ 00:23:31.899 Advait Nandakumar Menon: Most of it is in dbt, like, simple aggregations, like sun, average, or whatever that we can do within Omni, but most of the modeling is done in dbt now, which is handled by Demi.
250 00:23:32.930 ⇒ 00:23:37.660 Nandika Jhunjhunwala: So… I guess, like, for the semantic layers, like Caitlin is mentioning,
251 00:23:38.510 ⇒ 00:23:41.940 Nandika Jhunjhunwala: Can you also, like, maybe later walk us through how you would…
252 00:23:42.280 ⇒ 00:23:54.819 Nandika Jhunjhunwala: do changes in Omni and push it to dbt. I know that the Omni UI UX, like, not very, like, favorable for, like, doing those, like, modeling changes, and that’s, like, what Demi has said to me previously as well.
253 00:23:56.540 ⇒ 00:24:14.150 Nandika Jhunjhunwala: So, like, for your recommendation, like, would love to know more about, like, how that integration works, and if it would actually be correct for us to, like, move it all to, like, Omni or DBT, or how we could surface it in a way that’s easily flexible for us, from, like, a modeling perspective in the future.
254 00:24:14.970 ⇒ 00:24:19.119 Advait Nandakumar Menon: Yeah, there is this, I think…
255 00:24:19.250 ⇒ 00:24:23.830 Advait Nandakumar Menon: training guide that sent you the other day from Omni, so that space.
256 00:24:23.830 ⇒ 00:24:24.150 Nandika Jhunjhunwala: Yeah.
257 00:24:24.150 ⇒ 00:24:26.820 Advait Nandakumar Menon: a little about the dbt integration as well.
258 00:24:26.970 ⇒ 00:24:28.529 Advait Nandakumar Menon: So… And I’ll tip on.
259 00:24:28.530 ⇒ 00:24:34.249 Greg Stoutenburg: the trainer for tomorrow as well, and say that, you know, we’d like to take some time and have them show that.
260 00:24:35.590 ⇒ 00:24:36.230 Advait Nandakumar Menon: Yeah.
261 00:24:36.850 ⇒ 00:24:37.440 Greg Stoutenburg: Yep.
262 00:24:38.820 ⇒ 00:24:45.240 Caitlyn Vaughn: Okay, yeah, that’s one piece of it, and then… another piece is… let’s see if I share my screen…
263 00:24:45.350 ⇒ 00:24:47.900 Caitlyn Vaughn: As I’m looking through…
264 00:24:48.630 ⇒ 00:24:56.090 Caitlyn Vaughn: like, in Omni, the topics, I can see the base view, which looks like it’s, like, several
265 00:24:56.190 ⇒ 00:25:02.620 Caitlyn Vaughn: Maybe fact tables combined, or several tables combined, and, like, no dimension joins?
266 00:25:04.090 ⇒ 00:25:05.470 Caitlyn Vaughn: Is that what I’m seeing?
267 00:25:06.810 ⇒ 00:25:17.110 Advait Nandakumar Menon: Yeah, that is correct, and if we go back to the workflow created by Trigger, the one you were asking about, Nandika, it’s just taking the daily
268 00:25:17.510 ⇒ 00:25:28.239 Advait Nandakumar Menon: table, it’s not taking anything else, so if you ask Bloggie some question, it will be pointing to the daily table and not the monthly, or whatever else. So, it depends on the definition of the topic here, so…
269 00:25:28.650 ⇒ 00:25:35.910 Advait Nandakumar Menon: This topic here… probably combines a lot of other tables behind the scenes in the dbt logic,
270 00:25:36.140 ⇒ 00:25:39.990 Advait Nandakumar Menon: Yeah, you’re right. Basically, there are no other joints here.
271 00:25:40.360 ⇒ 00:25:57.870 Caitlyn Vaughn: Okay, and then I also see this RPT, which I think is just, like, a predefined aggregate of, like, time and data shoved in there for a chart, right? Instead of it being, like, fact and dimension tables. Can you explain to me what exactly? Is that… is that the right definition?
272 00:25:58.300 ⇒ 00:26:08.680 Advait Nandakumar Menon: Yeah, like you said, it’s a combination of different fact and dimension tables that has been… the raw logic for that, should reside in dbt, and that’s equal code.
273 00:26:08.810 ⇒ 00:26:20.530 Advait Nandakumar Menon: which was used to come up with this table, will be in dbt, and we are pulling that aggregated table into Omni, and that’s what’s being used in this topic here.
274 00:26:20.820 ⇒ 00:26:36.049 Caitlyn Vaughn: Okay, so why would we do this versus have, like, a base fact table plus the join dimensions, and then have, like, measures on top of it? Like, are there any measures inside of here, or are those measures accounted for in the RPT?
275 00:26:36.960 ⇒ 00:26:37.570 Caitlyn Vaughn: File.
276 00:26:37.570 ⇒ 00:26:48.830 Advait Nandakumar Menon: It… it should be accounted for in this table, that table that’s created. I would say, like, again, there are two schools of thought here.
277 00:26:49.960 ⇒ 00:26:59.129 Advait Nandakumar Menon: we can model everything within dbt and use the BR layer just for the reporting purpose as a way to expose the data, or you can…
278 00:26:59.520 ⇒ 00:27:03.909 Advait Nandakumar Menon: do all the modeling, the changes, everything within the BI layer, and…
279 00:27:04.290 ⇒ 00:27:10.100 Advait Nandakumar Menon: just maybe expose the data as well. So, what we have adopted here is the…
280 00:27:10.280 ⇒ 00:27:12.440 Advait Nandakumar Menon: Standard engineering principle, where it’s…
281 00:27:12.560 ⇒ 00:27:23.189 Advait Nandakumar Menon: ideally, it’s better to have all the modeling and everything within the dbt layer, and expose the reporting models, whatever is required, into the BI layer, which is Omni in this case.
282 00:27:25.760 ⇒ 00:27:41.339 Caitlyn Vaughn: I think for this, though, like, as I’m thinking through when we were doing the dashboarding for the BDR dashboard, and we want to change it from, like, 7 to, you know, 30 days, or, like, if I’m asking blobby questions and it’s giving me the wrong answers, I feel like…
283 00:27:41.870 ⇒ 00:28:05.519 Caitlyn Vaughn: and I could be wrong, but it feels like this model of having these, like, intermediate tables, where all of the data is, like, pre-aggregated is super limiting. It, like, helps us get to the, like, the chart faster, right? We can see the data, it’s modeled exactly how we said that we wanted it, but if any tweaks need to be made, then we have to go back into dbt and, like, remodel this table instead of…
284 00:28:06.040 ⇒ 00:28:09.379 Caitlyn Vaughn: You know, having, like, a much more flexible,
285 00:28:09.620 ⇒ 00:28:27.379 Caitlyn Vaughn: like, joins between our tables plus some kind of, like, predefined measures that we have included on the BI level, right? Because we can always add measures, or take away measures, or, like, change them. But with this, this takes, like, quite a bit of changing on the back end.
286 00:28:27.590 ⇒ 00:28:32.539 Caitlyn Vaughn: So, as I’m thinking through, like, us taking this over, I don’t know if, like, this specific…
287 00:28:33.450 ⇒ 00:28:36.929 Caitlyn Vaughn: Like, infrastructure would work for us, or make sense for us.
288 00:28:40.010 ⇒ 00:28:42.729 Advait Nandakumar Menon: Yeah, I think… Yeah, go ahead.
289 00:28:43.400 ⇒ 00:28:45.229 Greg Stoutenburg: Oh, yeah, I mean, I was just going to say, if there’s…
290 00:28:45.230 ⇒ 00:28:45.840 Advait Nandakumar Menon: Yeah.
291 00:28:45.840 ⇒ 00:28:52.320 Greg Stoutenburg: Yeah, if there’s a field that’s been calculated where you can’t touch it, that’s… that’s just a field that
292 00:28:53.070 ⇒ 00:29:07.169 Greg Stoutenburg: we shouldn’t do that for, right? Like, the 7-day stale thing, right? That’s… I don’t think that that directly speaks to, like, the overall strategy, that’s more just like, you know, hey, this particular instance of it impedes usability, so let’s get that out of there.
293 00:29:07.850 ⇒ 00:29:09.869 Caitlyn Vaughn: I think you’re going to find…
294 00:29:10.160 ⇒ 00:29:23.619 Greg Stoutenburg: Yeah, I… so I… and I think, actually, and, you know, let’s… let’s actually regroup on this in an hour and a half. I mean, I think you’re going to find that dbt is… is more flexible and easier to work with than maybe is,
295 00:29:23.620 ⇒ 00:29:28.989 Greg Stoutenburg: sort of being thought of here. I think, you know, it’s,
296 00:29:28.990 ⇒ 00:29:45.000 Greg Stoutenburg: you know, yeah, in following with the principle that Advait outlined, I think, it makes sense for these things to be in DBT. It’s just a matter of, you know, making sure it’s the right ones, and that you feel confident that you can make the edits that you need to. And so, you know, that’s what the training and handoff will help with.
297 00:29:45.320 ⇒ 00:30:01.919 Caitlyn Vaughn: Yeah, it’s not even that I don’t think we should be using dbt, I do think we should be using dbt, and I think modeling the tables in there makes a ton of sense. Like, I don’t know how else we would do it, really, but what my concern is, is these, like, RPT tables seem to be…
298 00:30:02.000 ⇒ 00:30:13.059 Caitlyn Vaughn: like, a lot of decisions are made inside of them that disallows us to have the flexibility we need when we’re using that data inside of Omni, or, like, we want to actually use that data later for something else.
299 00:30:14.710 ⇒ 00:30:19.020 Greg Stoutenburg: I mean, I do understand the… I understand what you’re saying in principle. Are there…
300 00:30:19.500 ⇒ 00:30:30.309 Greg Stoutenburg: Are there particular examples you have in mind where, because of the way that the table is modeled in dbt, it really is preventing you from accessing something in Omni?
301 00:30:30.920 ⇒ 00:30:41.670 Caitlyn Vaughn: Yes, I have a ton of examples. I mean, even this in and of itself is monthly, right? Which is a pattern that we should not have. There should be… it should just be, like, renewals.
302 00:30:42.180 ⇒ 00:30:57.099 Caitlyn Vaughn: Or components, and then we have measures inside of here that say, you know, here’s a pre-aggregated total monthly, we take all of this whole column and add it up or sum it up, and that would be, you know, what our monthly would be.
303 00:30:57.110 ⇒ 00:31:02.790 Caitlyn Vaughn: Or we take, you know, 30 days, and that’s monthly, versus it being, like, hard-coded inside of here.
304 00:31:02.890 ⇒ 00:31:08.300 Caitlyn Vaughn: Let me look at another one. There’s also, like, there’s no joins in here, right?
305 00:31:10.310 ⇒ 00:31:29.950 Greg Stoutenburg: Yeah, and some of that is standard. The smallest topic increment is a single table for Omni. So that can actually be advantageous. When there are distinct types of questions that would be asked of a particular table, then you can just make a single table, you know, join free topic.
306 00:31:31.390 ⇒ 00:31:43.509 Caitlyn Vaughn: Yeah, I… I understand that, it’s still quite limiting. Like, I don’t think that there should be any… there shouldn’t be any topic that only has one singular fact table and zero dimension tables joined to it.
307 00:31:45.210 ⇒ 00:31:50.419 Caitlyn Vaughn: In principle. Like, this is daily, right? And then if I’m looking through…
308 00:31:50.530 ⇒ 00:31:55.619 Caitlyn Vaughn: like, executive, there’s no joints, there’s no AI labels inside of here.
309 00:31:56.770 ⇒ 00:31:59.650 Caitlyn Vaughn: Once again, like, unit of time.
310 00:32:00.350 ⇒ 00:32:04.200 Caitlyn Vaughn: So, I don’t know, just going through this, it’s like, it’s looking like…
311 00:32:04.760 ⇒ 00:32:17.620 Caitlyn Vaughn: probably not the modeling that would be most conducive for us. So, the reason why I’m bringing this up is because I would love for us to fix this, and just, like, remodel it in a way where we can use it modularly going forward.
312 00:32:17.620 ⇒ 00:32:28.129 Caitlyn Vaughn: Yeah. I think one of our priorities should be, like, conforming the DIM accounts, so we know, like, you guys know the resolution layer way better than us, right?
313 00:32:29.300 ⇒ 00:32:36.150 Caitlyn Vaughn: So, if we can at least get, like, the DIM accounts and fact tables joined correctly, and, like, get away from this…
314 00:32:36.620 ⇒ 00:32:55.590 Caitlyn Vaughn: what is it? RPC tables, RPT tables? Yeah. That’s probably a pattern we’d like to see, and then also making sure that there’s measures defined on all of our, like, critical fact tables. Especially the ARR one, I’ve noticed. There’s, like, almost 1400 lines of code in that table, as opposed to the other tables have, like.
315 00:32:55.590 ⇒ 00:32:59.659 Caitlyn Vaughn: Max 40 lines of code, which seems like its own…
316 00:32:59.800 ⇒ 00:33:12.890 Caitlyn Vaughn: Kind of thing. But we’re missing, like, the AI context layers, those all need to be filled out. Any kind of, like, relationship tests on fax and DIM tables should be probably, like, manually tested.
317 00:33:12.970 ⇒ 00:33:19.869 Caitlyn Vaughn: And then we also have no, like, freshness blocks on any of our sources that I could see.
318 00:33:21.530 ⇒ 00:33:41.249 Greg Stoutenburg: Yeah, so let us… let us review this, with that feedback in mind, and… and come up with a menu for you, and sort of, like, we can… we can present some options and say, you know, we could do it this way, and here would be the implications, and as an alternative, if we do it this way, there’s some of the implications, and go from there. I think that’s something that we should just take on.
319 00:33:41.250 ⇒ 00:33:45.580 Greg Stoutenburg: So that you’re able to have the level of flexibility and visibility that you’re looking for.
320 00:33:46.120 ⇒ 00:33:48.109 Caitlyn Vaughn: Yeah,
321 00:33:48.110 ⇒ 00:34:07.030 Caitlyn Vaughn: That makes sense. I think that is a good idea. I think we probably, at this point, want to be, like, a little bit more pointed, and I think we’ve done enough work on our side to figure out, like, how we actually do want the data. So why don’t we outline for you, like, what’s going to make the most sense? What would be helpful from your perspective is to, like, consult on it, like.
322 00:34:07.030 ⇒ 00:34:16.940 Caitlyn Vaughn: us making these decisions means the consequences are X, Y, and Z. Yeah. And, like, does that make sense for us? And then if it does, we can go forward, and if not, then we can make some changes.
323 00:34:16.940 ⇒ 00:34:17.290 Greg Stoutenburg: Yep.
324 00:34:17.290 ⇒ 00:34:21.069 Caitlyn Vaughn: Nadica, do you have anything else that you wanted to add in here?
325 00:34:21.780 ⇒ 00:34:27.790 Nandika Jhunjhunwala: The other thing, when Demi shared that schema with, like, those definitions of those tables.
326 00:34:27.790 ⇒ 00:34:28.500 Caitlyn Vaughn: Yes.
327 00:34:28.500 ⇒ 00:34:33.229 Nandika Jhunjhunwala: A lot of them, just to be very frank, sounded like gibberish.
328 00:34:33.679 ⇒ 00:34:35.929 Nandika Jhunjhunwala: Like, we had to go in and look at…
329 00:34:36.150 ⇒ 00:34:50.059 Nandika Jhunjhunwala: the columns to figure out what that data exactly was in that table. So I think it would be really helpful for you to fix what data is exactly in that table, and give us more context, because some of them
330 00:34:50.670 ⇒ 00:34:58.749 Nandika Jhunjhunwala: is not, like, jargon we use internally or, like, generally to refer to those tables. So there’s, like, some mismatch there, because.
331 00:34:58.750 ⇒ 00:34:59.280 Greg Stoutenburg: Yeah.
332 00:34:59.280 ⇒ 00:35:01.920 Nandika Jhunjhunwala: I think they’re named properly, yeah.
333 00:35:01.920 ⇒ 00:35:02.620 Greg Stoutenburg: Mmm.
334 00:35:03.090 ⇒ 00:35:04.980 Caitlyn Vaughn: Do you have that doc, Nautica?
335 00:35:05.840 ⇒ 00:35:09.799 Nandika Jhunjhunwala: Let me just pull it up, actually.
336 00:35:20.790 ⇒ 00:35:22.190 Nandika Jhunjhunwala: Yeah, let me show my screen.
337 00:35:22.520 ⇒ 00:35:23.160 Caitlyn Vaughn: Yeah.
338 00:35:29.120 ⇒ 00:35:30.270 Nandika Jhunjhunwala: Yay!
339 00:35:32.170 ⇒ 00:35:34.110 Nandika Jhunjhunwala: So, for example, here…
340 00:35:37.740 ⇒ 00:35:40.410 Nandika Jhunjhunwala: There’s, like, Dim Team Salesforce account.
341 00:35:41.440 ⇒ 00:35:45.570 Nandika Jhunjhunwala: It says one product team matched to one Salesforce account when matched.
342 00:35:45.860 ⇒ 00:35:48.220 Nandika Jhunjhunwala: There’s DIM Customer Salesforce.
343 00:35:48.690 ⇒ 00:35:52.969 Nandika Jhunjhunwala: one account with CRM and lifecycle fields.
344 00:35:55.260 ⇒ 00:36:06.809 Nandika Jhunjhunwala: like, all of these definitions were sort of hard for us to, like, figure out. I think one of them, these are, like, a bridge dimension table. So, like, having that context there would be helpful.
345 00:36:06.860 ⇒ 00:36:17.329 Nandika Jhunjhunwala: and being more appointed and referring to what that table exactly does contain. I think this one is just a bunch of IDs and foreign keys and primary keys for storing.
346 00:36:17.500 ⇒ 00:36:24.330 Nandika Jhunjhunwala: And, like, for this table example, you say one customer role from Salesforce or QuickBooks.
347 00:36:24.570 ⇒ 00:36:28.269 Nandika Jhunjhunwala: Plus hyperlane billing when laned.
348 00:36:28.500 ⇒ 00:36:33.960 Nandika Jhunjhunwala: Like, I wasn’t sure what this was exactly, like, are you, like, joining on multiple dimension tables to create this one?
349 00:36:34.230 ⇒ 00:36:37.160 Nandika Jhunjhunwala: That would be super helpful to see.
350 00:36:39.350 ⇒ 00:36:42.969 Caitlyn Vaughn: And I think even more, like, seeing this is…
351 00:36:43.090 ⇒ 00:36:54.650 Caitlyn Vaughn: there’s a lot of, like, duplicate dimension tables, or even fact tables, where I would imagine it would be very difficult for Blobby or for Claude Code to, like.
352 00:36:54.720 ⇒ 00:37:10.459 Caitlyn Vaughn: go through these dim tables and, like, pick the right data, because there’s, like, 3 different definitions of what a customer is, and how many customers we have, and, like, they’re all being defined in terms of, like, Salesforce or Hyperline customers, where there should just be, like.
353 00:37:10.470 ⇒ 00:37:17.600 Caitlyn Vaughn: dimension customer table, right? Where it has the attributes that can be shared with any fact table.
354 00:37:17.770 ⇒ 00:37:25.430 Caitlyn Vaughn: Versus there being, like, 3 or 4 different DIM or fact tables. It just… it is a pattern that I think is creating
355 00:37:25.630 ⇒ 00:37:27.010 Caitlyn Vaughn: some…
356 00:37:27.150 ⇒ 00:37:44.670 Caitlyn Vaughn: incorrect outputs in Blobby. This is, like, another pattern that we saw, where it’s like, I don’t even know what half of these are, and then it looks like the definitions are either the same, or, like, trying to measure or hold the same amount of data on the same topic, which isn’t quite… it’s not gonna make sense for us.
357 00:37:45.660 ⇒ 00:37:53.820 Nandika Jhunjhunwala: The other thing is, like, for example here, it says company-wide usage metrics, like, being more specific as to, like, what
358 00:37:54.050 ⇒ 00:38:02.669 Nandika Jhunjhunwala: usage metrics we’re technically using here. It’s totally unclear, like, just looking at it as a snapshot.
359 00:38:03.590 ⇒ 00:38:08.020 Nandika Jhunjhunwala: And that you have, like, this demarcation between, like, team and, like.
360 00:38:08.760 ⇒ 00:38:26.140 Nandika Jhunjhunwala: company and, like, account and customer, I think those are all, like, interchangeable words for now. So just, like, maybe defining how you’re using these terms in a way that would make sense for us to read the schema or documentation that you share. Okay. That would be great.
361 00:38:26.140 ⇒ 00:38:31.930 Nandika Jhunjhunwala: And, like, this is… this is the other thing that I was referring to. There’s…
362 00:38:31.970 ⇒ 00:38:35.300 Nandika Jhunjhunwala: Team users monthly, there’s…
363 00:38:36.140 ⇒ 00:38:40.919 Nandika Jhunjhunwala: I think I’m not sure this needed to be a table with, like, monthly aggregation.
364 00:38:41.100 ⇒ 00:38:43.110 Nandika Jhunjhunwala: Same with, like.
365 00:38:43.110 ⇒ 00:38:44.630 Caitlyn Vaughn: No table should have…
366 00:38:44.920 ⇒ 00:39:03.220 Caitlyn Vaughn: time or unit measurements on it. It shouldn’t be, like, it shouldn’t be source… raw source specific, it shouldn’t be time-bound, it should just be the data on a specific… on a specific grain, like, the lowest grain we need, and we can always roll it up, and the measurements should be,
367 00:39:03.470 ⇒ 00:39:06.499 Caitlyn Vaughn: Added in the topic, and not in… not in the table.
368 00:39:09.020 ⇒ 00:39:10.060 Greg Stoutenburg: Heard, okay.
369 00:39:12.340 ⇒ 00:39:13.500 Greg Stoutenburg: Yeah, okay.
370 00:39:13.790 ⇒ 00:39:17.439 Greg Stoutenburg: I think, yeah, sorry, Nanaka, you can keep going.
371 00:39:18.140 ⇒ 00:39:18.910 Nandika Jhunjhunwala: No.
372 00:39:18.910 ⇒ 00:39:19.660 Greg Stoutenburg: They’re going, oh my god.
373 00:39:19.850 ⇒ 00:39:22.280 Nandika Jhunjhunwala: Yeah, yeah, I think that was it all.
374 00:39:22.280 ⇒ 00:39:23.440 Greg Stoutenburg: Okay, yeah, I mean, I…
375 00:39:24.080 ⇒ 00:39:34.729 Greg Stoutenburg: Yeah, I think the next piece will just be, like, consolidating that feedback and saying, like, hey, you know, here’s what they’re… here’s what they’re looking for for the handoff. And I can…
376 00:39:34.730 ⇒ 00:39:44.819 Greg Stoutenburg: I can raise that internally at the same time as you come up with what you want to see, and then we can, you know, as you suggested, consult on implications of particular decisions.
377 00:39:44.980 ⇒ 00:39:55.109 Caitlyn Vaughn: Yeah, yeah, and as we’re going through this, it’s, like, me reading first principles about this, and then going through and looking at, you know, the work that we’ve done, so if there is…
378 00:39:55.250 ⇒ 00:40:06.319 Caitlyn Vaughn: things that are different, that are different for a reason, and, like, they should be, that’s fine. I just… I’m not understanding, like, why they’re so off when you guys are coming in as this, like, data expert.
379 00:40:06.320 ⇒ 00:40:07.170 Greg Stoutenburg: Yeah, of course, yes.
380 00:40:07.170 ⇒ 00:40:09.100 Caitlyn Vaughn: This is the output, and I’m like, what is going on?
381 00:40:09.100 ⇒ 00:40:27.629 Greg Stoutenburg: Yeah, yeah, no, I think… I mean, I think something that, is on us to do a better job of getting out in front of is, like, articulating those principles before we build anything. Like, hey, just so you know, if you’re interested in this, right? I mean, not every client is interested in it, they just want to see their reports and have them work, and they don’t… they’re like, DBT, stop. Too far.
382 00:40:28.000 ⇒ 00:40:36.490 Greg Stoutenburg: But, you know, for someone like you all, who is, to be, you know, really transparent about what
383 00:40:36.540 ⇒ 00:41:00.179 Greg Stoutenburg: principles we’re following and where we’d make exceptions and deviations and why we would do that is, you know, that’s just work that we have to do. So, now we’re doing it after, rather than before, but, we’ll, you know, we’ll do it. So, let us come up with, let us, you know, let us get you those principled takes, and also, like I said, just share your feedback on this with the team, so that we can, do things like.
384 00:41:00.200 ⇒ 00:41:03.630 Greg Stoutenburg: get date measures out of tables.
385 00:41:04.130 ⇒ 00:41:05.660 Greg Stoutenburg: That’s work that we can do, yeah.
386 00:41:05.660 ⇒ 00:41:13.709 Caitlyn Vaughn: Yeah, definitely. Yeah, we’ll send you, like, a clear list, so it’s outlined, like, how we actually want to see the data, and what needs to get done.
387 00:41:13.710 ⇒ 00:41:15.950 Greg Stoutenburg: Is that something that’s already in progress on your end?
388 00:41:16.080 ⇒ 00:41:17.030 Caitlyn Vaughn: What do you mean?
389 00:41:17.030 ⇒ 00:41:21.190 Greg Stoutenburg: Sorry, like, documenting that, like you said, you know, you’re going to have what you want to do. Okay, alright.
390 00:41:21.190 ⇒ 00:41:30.930 Caitlyn Vaughn: Yeah, I created the audit doc, and then I audited all of the work that’s been done, so I can send over the report if you want, but I think it would be easier for me to just send a, like, a simple list of what needs to get changed.
391 00:41:31.900 ⇒ 00:41:33.830 Caitlyn Vaughn: I think the other thing is, like.
392 00:41:34.410 ⇒ 00:41:39.129 Caitlyn Vaughn: like, I think this is a great example of, I need you guys to be…
393 00:41:39.780 ⇒ 00:41:48.519 Caitlyn Vaughn: like, consulting on our decision-making, and, like, on our opinions as a company. Like, for example, when I asked for…
394 00:41:48.520 ⇒ 00:42:03.270 Caitlyn Vaughn: like, to view… to be able to view raw data side by side in Omni, and then you guys… you guys did that for us. You, like, you were like, yes, absolutely, and then you made it work, but, like, the consequence of that was that Blobby is now pulling from, like, raw data, which is a terrible pattern, right?
395 00:42:03.270 ⇒ 00:42:03.840 Greg Stoutenburg: Yeah.
396 00:42:03.840 ⇒ 00:42:07.300 Caitlyn Vaughn: Like, I just haven’t seen a lot of these…
397 00:42:08.280 ⇒ 00:42:11.520 Caitlyn Vaughn: like, downstream thing, so I think we need to just…
398 00:42:11.760 ⇒ 00:42:30.319 Caitlyn Vaughn: like, align on, we want… what we want from the data infrastructure is to be able to, like, use Blobby to generate different dashboards, see data really clearly, have it be flexible, have it not be time-bound, so that we can use it going forward, and that we’re not having to, like, redo dbt work.
399 00:42:30.320 ⇒ 00:42:34.030 Caitlyn Vaughn: Like, that is, like, our genesis of what we’re kind of moving towards.
400 00:42:34.030 ⇒ 00:42:34.400 Greg Stoutenburg: Yeah.
401 00:42:34.400 ⇒ 00:42:57.629 Caitlyn Vaughn: So, Greg, I think you’re, like, the good first principles person. I think as we, like, over the next couple of weeks are, like, working through this, I would love for you to be able to be like, hey, this doesn’t make sense for the things that you outlined, and, like, this is the consequences of that, even if I’m saying, like, we want to see raw data, and you being like, hey, that’s not a great pattern for Omni because of XYZ, that would be helpful for me to learn even faster.
402 00:42:57.720 ⇒ 00:42:58.640 Greg Stoutenburg: Sure.
403 00:42:59.460 ⇒ 00:42:59.939 Greg Stoutenburg: Yeah, heard.
404 00:42:59.940 ⇒ 00:43:03.830 Caitlyn Vaughn: I think from our CEO, we have 2 weeks.
405 00:43:04.330 ⇒ 00:43:19.130 Caitlyn Vaughn: to finish up this project. It was basically like, okay, this has been dragging on, and I don’t want to, like, give another month to this. We have two weeks, so as of May 15th, we have to have this all wrapped up and passed off to the default team internally.
406 00:43:19.490 ⇒ 00:43:20.070 Greg Stoutenburg: Okay.
407 00:43:20.410 ⇒ 00:43:26.189 Greg Stoutenburg: Okay. Is that when the contract ends, or that’s just when he wants to have final reports in?
408 00:43:26.410 ⇒ 00:43:32.530 Caitlyn Vaughn: That’s what… I think our contract technically goes through the end of the year, just we’re not having to, like, re-sign it.
409 00:43:32.530 ⇒ 00:43:35.229 Greg Stoutenburg: But he’s just saying that’s when he wants, yeah, that’s when he wants the.
410 00:43:35.230 ⇒ 00:43:36.280 Caitlyn Vaughn: Yeah.
411 00:43:36.280 ⇒ 00:43:37.100 Greg Stoutenburg: Yeah, okay.
412 00:43:37.750 ⇒ 00:43:43.899 Caitlyn Vaughn: So we have a two-week sprint, which is a good, timeline for us to rally behind and get as much done as we can in that time.
413 00:43:43.900 ⇒ 00:43:57.189 Greg Stoutenburg: Yep, okay. Okay, heard. Yeah, let me, let me, you know, consult internally, and, and, and raise these items, and, yeah, I mean, identify anything that looks like it’s…
414 00:43:57.350 ⇒ 00:44:01.139 Greg Stoutenburg: A risk or a challenge, and we can… we’ll communicate about that.
415 00:44:01.140 ⇒ 00:44:01.890 Caitlyn Vaughn: Cool, amazing.
416 00:44:02.230 ⇒ 00:44:02.799 Caitlyn Vaughn: Thank you so much!
417 00:44:02.800 ⇒ 00:44:05.890 Nandika Jhunjhunwala: I think, Advit, do you have something to say?
418 00:44:06.460 ⇒ 00:44:18.329 Advait Nandakumar Menon: Yeah, I want to say one thing, probably, Greg, this is to us, but there is the dbt integration setting or something we have to enable wherein
419 00:44:18.430 ⇒ 00:44:33.969 Advait Nandakumar Menon: we can basically code in dbt within Omni itself, and that’s how you push the changes between Omni and dbt. I think we should probably enable this for these guys here, Greg, and that might help them to just look into the dbt code and.
420 00:44:33.970 ⇒ 00:44:34.530 Greg Stoutenburg: I think so.
421 00:44:34.530 ⇒ 00:44:37.970 Advait Nandakumar Menon: You guys can just change it on the go, I feel.
422 00:44:38.530 ⇒ 00:44:39.110 Greg Stoutenburg: Yeah.
423 00:44:39.340 ⇒ 00:44:43.219 Greg Stoutenburg: Amazing. Yeah, let’s, yeah, let’s make sure,
424 00:44:43.740 ⇒ 00:44:56.429 Greg Stoutenburg: So I’ve communicated with Brian about, like, where the team is at and what you’re up to and things like that. Let’s also just make sure we’re super clear as you prepare the, OmniHandoff Docs, that we outline how to do that.
425 00:44:56.990 ⇒ 00:44:58.139 Advait Nandakumar Menon: Yep. Yep.
426 00:44:58.430 ⇒ 00:45:12.640 Caitlyn Vaughn: Yeah, I think documentation’s gonna be, like, the best thing that you can do for us, over the next couple weeks, and just us hacking away and being able to ask you questions while we still have you, and fix as much as we can to be in the format that will be best for us.
427 00:45:12.640 ⇒ 00:45:14.370 Greg Stoutenburg: Yep, yep, yeah, that sounds good.
428 00:45:14.700 ⇒ 00:45:23.159 Nandika Jhunjhunwala: The other thing I want to add, I’m so sorry, like, with this documentation, if we’re, like, saying, like, this is how we think it should be, and
429 00:45:23.240 ⇒ 00:45:41.479 Nandika Jhunjhunwala: going back to first principles, if you think, no, actually, like, this is the reason we put it there, and we really believe that it should be there, like, please do push back, so we can all, like, sort of, like, discuss and talk through these, like, strategies and solutions that you’re coming up with, and really feel good about what we’re seeing. So, totally, like.
430 00:45:41.800 ⇒ 00:45:47.719 Nandika Jhunjhunwala: Feel free to push back if you think, like, no, this is, like, it should be there, this is why you put it, so you’ll be like, okay, yeah.
431 00:45:47.720 ⇒ 00:45:55.640 Greg Stoutenburg: Yeah, yeah, consult, in other words, right? Yes, totally. Yep. Yeah, sounds good. That’s, that’s, that’s in the job description.
432 00:45:56.220 ⇒ 00:46:02.970 Greg Stoutenburg: Okay, cool. I actually, I just… I think… I take it that was it on that, on that topic, right? Okay, I was gonna just…
433 00:46:02.970 ⇒ 00:46:24.619 Greg Stoutenburg: move on, just to raise something I noticed, and I think I’ve separately mentioned this to both of you, but I was digging in on it again. I was looking around at some of the rest of, you know, what we can call, like, the growth stack, and noticing, like, some things that I weren’t sure if they were… if they were just stable, and, you know, everything is going great as they are, or, like, if they’ve been deprioritized. So, for example, like,
434 00:46:24.770 ⇒ 00:46:25.919 Greg Stoutenburg: Well, I made a…
435 00:46:27.790 ⇒ 00:46:47.629 Greg Stoutenburg: made a little appendix here, right? Like, here’s… here’s what’s coming into segment, and here’s the destinations. A lot of these, the connections are disabled, or nothing’s gone to it in a very long time. In some cases, like, Amplitude, last login is, like, a couple of weeks ago, so it doesn’t look like anyone’s really done
436 00:46:48.090 ⇒ 00:47:00.150 Greg Stoutenburg: going in there, Customer I.O. has, I think, I think it was zero active campaigns, like, nothing’s been… looks like nothing’s been sent in a month. And I was just, you know, just thinking through, like, I understand
437 00:47:00.420 ⇒ 00:47:10.219 Greg Stoutenburg: the priority here is that we’re doing all this reporting for growth purposes, and along those same lines, I was thinking, you know, there are these other pieces of the stack that I wasn’t sure if they were…
438 00:47:10.540 ⇒ 00:47:13.979 Greg Stoutenburg: Again, just stable or deprioritized, or if there’s a…
439 00:47:14.190 ⇒ 00:47:18.000 Greg Stoutenburg: If there’s a pain there, or, you know, what’s going on there.
440 00:47:18.260 ⇒ 00:47:37.630 Caitlyn Vaughn: Yeah, so for Segment, we do still use Segment, but in, like, a pretty specific way. I think it’s really mainly feeding into loops, and pushing our, like, our website data out. We’re not using the Amplitude anymore, we switched over to PostHog a couple months back, so we’ve gone away. I think we ended our contract with them.
441 00:47:37.630 ⇒ 00:47:49.909 Caitlyn Vaughn: And then… what else did you say? Oh, Customer I.O. We switched from Customer I.O. to Email Bison, maybe 6 months ago. So I’d be surprised if anyone was in there. I’d probably need to, like, cancel our subscription, to be honest.
442 00:47:49.910 ⇒ 00:47:50.240 Nandika Jhunjhunwala: Right.
443 00:47:50.610 ⇒ 00:47:51.249 Greg Stoutenburg: I think we must…
444 00:47:51.250 ⇒ 00:47:53.170 Nandika Jhunjhunwala: from SmartV Team, and I soon.
445 00:47:53.170 ⇒ 00:47:53.700 Caitlyn Vaughn: That’s.
446 00:47:53.700 ⇒ 00:47:55.430 Nandika Jhunjhunwala: is for, like, marketing purposes.
447 00:47:55.430 ⇒ 00:47:55.770 Caitlyn Vaughn: Interesting.
448 00:47:55.770 ⇒ 00:48:03.679 Nandika Jhunjhunwala: that we have, but I don’t think, like, our head of market is currently using it, but I think we will always have that.
449 00:48:03.960 ⇒ 00:48:05.789 Nandika Jhunjhunwala: It’s part of our growth stack.
450 00:48:05.870 ⇒ 00:48:23.699 Greg Stoutenburg: Yeah, so, like, this is disabled, there’s nothing… there’s no mention of Bison. Posthog is not connected as a destination. Your amplitude is still on, I was just in it. It has some logins, but it doesn’t appear that anything’s, like, being done or created.
451 00:48:24.160 ⇒ 00:48:28.429 Greg Stoutenburg: or loops… I didn’t go…
452 00:48:28.570 ⇒ 00:48:30.940 Greg Stoutenburg: I didn’t go beyond… oh, I hate it when it does that.
453 00:48:31.650 ⇒ 00:48:33.380 Greg Stoutenburg: I’m just gonna do it again.
454 00:48:35.130 ⇒ 00:48:36.330 Greg Stoutenburg: Yeah, yeah, come on.
455 00:48:38.220 ⇒ 00:48:42.690 Greg Stoutenburg: So, we can see… The mappings, so it’s sending, you know.
456 00:48:42.970 ⇒ 00:48:47.329 Greg Stoutenburg: There’s a track event set up, but even that is disabled, so…
457 00:48:47.330 ⇒ 00:48:48.860 Caitlyn Vaughn: Yeah, we disabled it.
458 00:48:48.860 ⇒ 00:48:50.240 Greg Stoutenburg: Yeah, okay.
459 00:48:50.530 ⇒ 00:48:51.680 Greg Stoutenburg: Yeah, okay.
460 00:48:52.290 ⇒ 00:48:55.399 Greg Stoutenburg: Yeah, so I guess I just… I just wasn’t sure, you know, what…
461 00:48:55.530 ⇒ 00:49:01.270 Greg Stoutenburg: What the status of these tools was, or, you know, if you wanted,
462 00:49:01.520 ⇒ 00:49:09.779 Greg Stoutenburg: you know, help setting them up to optimize, you know, that lifecycle marketing, those growth touches, things like that. So just wanted to raise that.
463 00:49:10.040 ⇒ 00:49:28.739 Caitlyn Vaughn: Yeah, I appreciate that. We’ve given you, I think, all the tools that we’re really using. I did set up Customer I.O. a very long time ago for marketing stuff, but we’re not really doing any yet until the new product is out, so all of it is, like, kind of irrelevant at the moment, until we get out the new product, which seems back.
464 00:49:28.740 ⇒ 00:49:37.810 Greg Stoutenburg: Yeah, I mean, that is a nice thing about… I almost said about Phoenix, about, segment, is that once… once you’re ready, you just put that snippet on
465 00:49:38.170 ⇒ 00:49:47.229 Greg Stoutenburg: product, and then here you go. All that stuff is now going to those destinations, so that’s pretty cool. Okay, all right. Well, yeah, just wanted to raise that and check in on it.
466 00:49:47.560 ⇒ 00:49:56.289 Caitlyn Vaughn: Okay, cool. Well, I appreciate you, keeping an eye on everything. I think we’re in a good place. I think we will be in a great place in two weeks, and I really appreciate you guys.
467 00:49:56.290 ⇒ 00:49:57.530 Greg Stoutenburg: Yep. Alright, thanks a lot.
468 00:49:57.530 ⇒ 00:49:58.149 Nandika Jhunjhunwala: Thank you so much.
469 00:49:58.150 ⇒ 00:50:04.699 Greg Stoutenburg: Let’s go back, huddle, and regroup. And, see you in 10 minutes.
470 00:50:04.700 ⇒ 00:50:05.900 Caitlyn Vaughn: Bye. Bye.
Tickets Created
Generated by post-call-to-tickets on 2026-05-01
- DEF-624 Align Default dbt model to Kimball-style facts, dimensions, and conformed DIM accounts
- DEF-625 Store Default models at lowest necessary grain; define time windows and measures in Omni (not only in RPT tables)
- DEF-626 Fill Omni AI context for critical Default topics and validate fact–dim relationships (incl. manual tests)
- DEF-627 Revise Default schema documentation: plain-language definitions, glossary, and table contents