Meeting Title: Architecture Session: Insomnia daily reporting Date: 2025-10-28 Meeting participants: Uttam Kumaran, Casie Aviles, Mustafa Raja, Samuel Roberts, Demilade Agboola, Awaish Kumar
WEBVTT
1 00:00:33.500 ⇒ 00:00:34.590 Uttam Kumaran: Hey, Casey.
2 00:00:36.790 ⇒ 00:00:38.020 Casie Aviles: Hey, Laton.
3 00:00:38.260 ⇒ 00:00:39.310 Uttam Kumaran: Hey, how are you?
4 00:00:40.570 ⇒ 00:00:45.159 Casie Aviles: Alright, yep, doing good, just… Looking at the fig jam again.
5 00:00:45.680 ⇒ 00:00:46.470 Uttam Kumaran: Nice.
6 00:01:10.700 ⇒ 00:01:12.229 Samuel Roberts: Here we go, okay.
7 00:01:15.110 ⇒ 00:01:16.010 Uttam Kumaran: Hi, guys.
8 00:01:17.190 ⇒ 00:01:18.509 Samuel Roberts: Yo, how’s it going?
9 00:01:20.240 ⇒ 00:01:21.300 Uttam Kumaran: I’m good
10 00:01:22.420 ⇒ 00:01:30.209 Uttam Kumaran: I just did not like those updates, like, I… that’s not… they’re not helpful for anybody, and I’m like, what are we doing?
11 00:01:30.470 ⇒ 00:01:40.210 Samuel Roberts: Yeah, I was glad you said something, because I didn’t want to be like, I don’t, like… we confirmed Postgres is… er, what was it, basically just Postgres? I was like, that was not, like… I don’t know.
12 00:01:40.210 ⇒ 00:01:45.039 Uttam Kumaran: Well, it’s just lazy. I’m like, look, we’re an AI company, if you copy-paste AI, everybody’s gonna know.
13 00:01:45.040 ⇒ 00:01:46.749 Samuel Roberts: Exactly, yeah, yeah.
14 00:02:01.950 ⇒ 00:02:05.530 Samuel Roberts: I’m written on… Movement closer.
15 00:02:05.820 ⇒ 00:02:07.290 Samuel Roberts: Montimilate?
16 00:02:08.610 ⇒ 00:02:10.350 Samuel Roberts: And we will be…
17 00:02:12.530 ⇒ 00:02:15.160 Uttam Kumaran: Yeah, you mind pinging them in the Insomnia channel?
18 00:02:15.160 ⇒ 00:02:16.090 Samuel Roberts: Yeah, totally.
19 00:03:01.310 ⇒ 00:03:03.880 Samuel Roberts: Do we wait for them, or should we start chatting, or what are you thinking?
20 00:03:03.880 ⇒ 00:03:09.510 Uttam Kumaran: Yeah, let’s wait, like, 1 minute, a couple minutes. I’d like to have them there.
21 00:03:10.860 ⇒ 00:03:15.840 Samuel Roberts: Yeah, I had a couple questions about some of this stuff, like that one. For example.
22 00:03:16.630 ⇒ 00:03:25.340 Casie Aviles: So, this just means… Is that… for this… Sources?
23 00:03:26.220 ⇒ 00:03:30.629 Casie Aviles: We do the daily update, and then we also do a 7-day backfill.
24 00:03:30.960 ⇒ 00:03:35.719 Samuel Roberts: Oh, okay, so that wasn’t, like, this happens and then triggers these things, that’s why I was… it’s just… these ones…
25 00:03:36.320 ⇒ 00:03:38.409 Samuel Roberts: Every day, that’s what happens? Okay.
26 00:03:38.980 ⇒ 00:03:44.500 Samuel Roberts: Sorry, I thought it was a flow. No, you’re good, you’re good. Now I… that makes more sense. I just wasn’t 100%.
27 00:03:45.580 ⇒ 00:03:49.660 Samuel Roberts: I didn’t know what was flowing where based on those, but that’s good. Okay.
28 00:03:51.750 ⇒ 00:03:52.490 Casie Aviles: Okay.
29 00:03:54.880 ⇒ 00:03:58.070 Casie Aviles: Yeah, I think most of us are here.
30 00:03:58.190 ⇒ 00:04:02.450 Casie Aviles: So I guess… Yeah, I’ll just go ahead and…
31 00:04:03.290 ⇒ 00:04:05.970 Casie Aviles: Give a high-level overview, like we did.
32 00:04:07.870 ⇒ 00:04:12.799 Casie Aviles: So, for this client, we have, Insonia Cupe, so…
33 00:04:13.010 ⇒ 00:04:16.880 Casie Aviles: One of the tasks that we’re doing for them is…
34 00:04:17.290 ⇒ 00:04:21.629 Casie Aviles: To be able to, help them with their marketing report.
35 00:04:22.240 ⇒ 00:04:32.160 Casie Aviles: So, for that, they have, multiple, sources, so I’m just gonna focus on… this,
36 00:04:32.320 ⇒ 00:04:35.990 Casie Aviles: photo for now. So they have this spreadsheet.
37 00:04:36.490 ⇒ 00:04:43.830 Casie Aviles: And… So they track, basically, several, data for…
38 00:04:44.560 ⇒ 00:04:51.979 Casie Aviles: For their marketing, so they have for paid media, which are from Google and Meta, and then they have owned marketing.
39 00:04:53.020 ⇒ 00:04:56.200 Casie Aviles: they also have, FDA promotions, so…
40 00:04:56.320 ⇒ 00:04:58.850 Casie Aviles: Which are from DoorDash and Uber Eats.
41 00:05:00.320 ⇒ 00:05:07.659 Casie Aviles: They also have, like, a daily sales flash that gets sent, so… How that looks.
42 00:05:07.870 ⇒ 00:05:12.340 Casie Aviles: Is… It gets sent through, an inbox.
43 00:05:13.090 ⇒ 00:05:16.380 Casie Aviles: Through, through email, and it ends up in an inbox.
44 00:05:17.420 ⇒ 00:05:22.670 Casie Aviles: For example, this one… This is,
45 00:05:22.850 ⇒ 00:05:27.000 Casie Aviles: Basically, that for… this is the sheet that we…
46 00:05:28.000 ⇒ 00:05:33.440 Casie Aviles: Yeah, that we use to get the data for these sections.
47 00:05:33.670 ⇒ 00:05:38.680 Casie Aviles: So, bottom line is, they have a lot of, sources for their data.
48 00:05:39.030 ⇒ 00:05:44.529 Casie Aviles: And where we come in here is to build out the automation.
49 00:05:46.120 ⇒ 00:05:47.960 Casie Aviles: Such that we’re able to get
50 00:05:48.100 ⇒ 00:05:55.710 Casie Aviles: The data from each of the sources, and then just, let that… ideally, we should… we can just let that run
51 00:05:56.090 ⇒ 00:06:03.719 Casie Aviles: Daily and… That’s… To help them optimize, like, this, the reporting process.
52 00:06:04.100 ⇒ 00:06:08.090 Casie Aviles: So, yeah, that’s the overall… that’s a high level…
53 00:06:08.820 ⇒ 00:06:11.179 Casie Aviles: Overview of what we’re trying to do.
54 00:06:12.010 ⇒ 00:06:15.539 Casie Aviles: So, yeah, let me know if it’s clear so far, and…
55 00:06:16.230 ⇒ 00:06:18.959 Casie Aviles: If, you already have any questions.
56 00:06:24.260 ⇒ 00:06:31.960 Samuel Roberts: Well, so I guess my first question is, like, so we’re kind of… this was basically mapping to what they already had, right? We were just trying to automate what people were…
57 00:06:32.060 ⇒ 00:06:33.600 Samuel Roberts: doing before.
58 00:06:34.260 ⇒ 00:06:36.260 Casie Aviles: Yes, okay.
59 00:06:36.900 ⇒ 00:06:39.660 Casie Aviles: So, yeah, to also give you a sense of…
60 00:06:40.000 ⇒ 00:06:48.339 Casie Aviles: what they were doing before, but so it’s all manual, so they would have to, for example, they would go here, they would go and click
61 00:06:48.910 ⇒ 00:06:50.040 Casie Aviles: This one…
62 00:06:50.590 ⇒ 00:06:56.729 Casie Aviles: And they would just have to get, like, the data that they need, they’ll apply a formula, or they’ll have to go to DoorDash.
63 00:06:56.990 ⇒ 00:07:00.780 Casie Aviles: They have to log in. Yeah, so, pretty much that.
64 00:07:01.170 ⇒ 00:07:02.779 Casie Aviles: Then they have to get this.
65 00:07:03.340 ⇒ 00:07:07.039 Casie Aviles: So that’s what they used to do before.
66 00:07:07.480 ⇒ 00:07:10.139 Casie Aviles: We came into…
67 00:07:10.510 ⇒ 00:07:18.409 Casie Aviles: Ideally, we wanted to optimize that process, and we’ve built out some automations that, has helped us…
68 00:07:18.740 ⇒ 00:07:21.319 Casie Aviles: Shave off some time in that process, but…
69 00:07:21.920 ⇒ 00:07:25.470 Casie Aviles: There are definitely a lot of room for improvement.
70 00:07:25.910 ⇒ 00:07:26.960 Samuel Roberts: And…
71 00:07:27.480 ⇒ 00:07:30.609 Casie Aviles: There’s definitely, yeah, there’s definitely room for improvement, and
72 00:07:31.050 ⇒ 00:07:35.560 Casie Aviles: So I’m just gonna go here to, like, the common issues, so…
73 00:07:35.960 ⇒ 00:07:38.080 Samuel Roberts: Yeah. Or, like, also the goals…
74 00:07:40.270 ⇒ 00:07:43.810 Casie Aviles: So one of the goals that we want is to have resiliency.
75 00:07:44.490 ⇒ 00:07:55.070 Casie Aviles: And, you know, the automations for some of them, we don’t have, like, API endpoints. I think that was already established in the past, where
76 00:07:57.040 ⇒ 00:08:02.950 Casie Aviles: Ideally, if we had API endpoints, it would be much easier to get the data out from these sources.
77 00:08:04.320 ⇒ 00:08:08.280 Casie Aviles: Right. But yeah, the issue is that
78 00:08:08.780 ⇒ 00:08:11.330 Casie Aviles: For some of the sources, like.
79 00:08:11.780 ⇒ 00:08:15.899 Casie Aviles: DoorDash, like, mainly these ones, Uber.
80 00:08:16.260 ⇒ 00:08:19.010 Casie Aviles: But we couldn’t get, like, API access.
81 00:08:19.250 ⇒ 00:08:26.469 Casie Aviles: So, kind of our workaround is to use the browser automations, which I would say are…
82 00:08:27.080 ⇒ 00:08:33.179 Casie Aviles: They work, but they also break pretty often, so that’s one of the things that
83 00:08:35.190 ⇒ 00:08:37.020 Casie Aviles: That’s one of the issues that I would…
84 00:08:37.730 ⇒ 00:08:41.090 Casie Aviles: I would… that I face with this whole setup.
85 00:08:43.990 ⇒ 00:08:48.179 Casie Aviles: And… Yeah, so that’s pretty much one of the issues that I have.
86 00:08:48.340 ⇒ 00:08:55.260 Casie Aviles: And then also another thing that we listed down here is redundancy. So, while we do have all this data.
87 00:08:55.530 ⇒ 00:08:59.450 Casie Aviles: It’s not really usable for analytics, so…
88 00:09:02.720 ⇒ 00:09:11.270 Casie Aviles: Yeah, so what ends up happening is, people also export from the platforms. They export, like, they create reports.
89 00:09:11.930 ⇒ 00:09:16.439 Casie Aviles: And then that’s where they… that’s what they use for the analysis part.
90 00:09:16.880 ⇒ 00:09:17.840 Uttam Kumaran: I think.
91 00:09:17.840 ⇒ 00:09:24.470 Casie Aviles: That’s… mostly, like, most of the issues that are faced
92 00:09:25.310 ⇒ 00:09:27.240 Casie Aviles: For, given this setup.
93 00:09:28.900 ⇒ 00:09:36.379 Casie Aviles: Yeah, also, the reason why we’re doing this for 7 days, like a 7-day backfill is…
94 00:09:37.170 ⇒ 00:09:45.270 Casie Aviles: We wanted to see, kind of, how… Like, how the… The data changes.
95 00:09:45.510 ⇒ 00:09:55.229 Casie Aviles: per day, I believe, and… That’s not something we can… Immediately report on… Given this setup, so…
96 00:09:55.780 ⇒ 00:09:59.970 Casie Aviles: I think that’s… yeah, overall, that’s… It,
97 00:10:00.910 ⇒ 00:10:03.999 Casie Aviles: I think maybe one thing is…
98 00:10:04.160 ⇒ 00:10:12.050 Casie Aviles: should we continue with this kind of, setup that they have? Do we change this, or…
99 00:10:12.880 ⇒ 00:10:18.980 Casie Aviles: Those are some of the questions, overarching questions, that we have
100 00:10:21.690 ⇒ 00:10:26.459 Casie Aviles: Yeah, like, for some of the existing solutions, are these, like, the best
101 00:10:27.980 ⇒ 00:10:33.049 Casie Aviles: Is this, like, the best approach for that, or is there something else we could try, you know?
102 00:10:33.820 ⇒ 00:10:37.709 Casie Aviles: So I think, yeah, I’ll pause there for now.
103 00:10:37.840 ⇒ 00:10:38.690 Casie Aviles: And…
104 00:10:38.720 ⇒ 00:10:39.970 Uttam Kumaran: Let me know if…
105 00:10:41.350 ⇒ 00:10:44.379 Casie Aviles: And everything’s clear so far, and… yeah.
106 00:10:54.300 ⇒ 00:11:00.599 Samuel Roberts: Well, my initial thought is, I think this is kind of what I was wondering a little bit earlier, before, I wasn’t sure how much their…
107 00:11:02.520 ⇒ 00:11:07.840 Samuel Roberts: they like, or want to stick with this. Like, we were just clearly trying to map onto what they already had, and I think…
108 00:11:08.330 ⇒ 00:11:13.769 Samuel Roberts: from what I’ve heard now, we have a little more flexibility to potentially change up this daily impact scorecard.
109 00:11:14.270 ⇒ 00:11:15.509 Samuel Roberts: Is that true?
110 00:11:15.880 ⇒ 00:11:18.829 Samuel Roberts: That’s part of the re-architecting talk that we’re having? Okay.
111 00:11:20.020 ⇒ 00:11:25.660 Samuel Roberts: Good, okay. Because I think there’s got to be better ways, and I’m glad we have some more data people here, because I’m curious
112 00:11:25.760 ⇒ 00:11:33.730 Samuel Roberts: this backfill process, I think I understand, because it… we’re getting snapshots every day. Is that how this is working? Like, when you pull the data from…
113 00:11:34.640 ⇒ 00:11:35.720 Samuel Roberts: say…
114 00:11:35.720 ⇒ 00:11:36.290 Casie Aviles: Yeah.
115 00:11:36.290 ⇒ 00:11:38.899 Samuel Roberts: DoorDash, or Uber Ads, or Uber Eats.
116 00:11:39.300 ⇒ 00:11:41.260 Samuel Roberts: That’s just, like, giving you, like, what it…
117 00:11:41.440 ⇒ 00:11:44.779 Samuel Roberts: Currently is at that morning, right?
118 00:11:45.290 ⇒ 00:11:46.000 Casie Aviles: Yes.
119 00:11:46.340 ⇒ 00:11:51.450 Samuel Roberts: And there’s no historical… so we’re just doing that manually. So I feel like there’s got to be a better way to keep track of that.
120 00:11:51.800 ⇒ 00:11:59.369 Samuel Roberts: With a timestamp, and over time, rather than having to do all this Backfilling as well, but…
121 00:12:03.540 ⇒ 00:12:11.230 Samuel Roberts: I know, Casey, you and I have talked a bunch about this, like, the browser automation stuff, and the fact that these… that DoorDash, Uber, and Uber Eats don’t really have…
122 00:12:11.530 ⇒ 00:12:16.549 Samuel Roberts: or don’t have APIs that we have access to, at least. They have… they sort of seem to have an API.
123 00:12:16.780 ⇒ 00:12:19.549 Samuel Roberts: I’m curious, like, what…
124 00:12:19.710 ⇒ 00:12:26.249 Samuel Roberts: how fragile are these, is sort of my first question. Like, are they… is it every day you’re in there doing something to fix it, or…
125 00:12:28.700 ⇒ 00:12:36.599 Casie Aviles: I would say, like, weekly, there are some things that get changed because of how, like, campaigns are being
126 00:12:37.140 ⇒ 00:12:44.660 Casie Aviles: Deployed, so that’s one thing, so I have to, like, do another…
127 00:12:45.960 ⇒ 00:12:51.170 Casie Aviles: So, like, I considered just one… Very linear set of…
128 00:12:51.440 ⇒ 00:12:58.449 Casie Aviles: processes, like, I would… you would have to just go and click here, and then click that, but then sometimes it changes.
129 00:12:58.680 ⇒ 00:13:06.360 Casie Aviles: That’s one thing, and then the other one… It’s like, for authentication, so…
130 00:13:07.280 ⇒ 00:13:07.750 Samuel Roberts: Right.
131 00:13:07.750 ⇒ 00:13:12.030 Casie Aviles: typically expires, and I have to, like,
132 00:13:13.750 ⇒ 00:13:21.539 Casie Aviles: like, the workaround that I have is to just load in the cookies, that I have logged in.
133 00:13:21.860 ⇒ 00:13:28.389 Casie Aviles: So I would just use this, I would export the cookies here, and then I would try to load that into the browser context.
134 00:13:28.800 ⇒ 00:13:30.689 Casie Aviles: And I have to do that, like…
135 00:13:31.840 ⇒ 00:13:36.850 Casie Aviles: I was wondering, like, if there’s, like, a better way to handle that,
136 00:13:38.330 ⇒ 00:13:44.560 Casie Aviles: Given… given our setup, like, is there a way that… such that We can just.
137 00:13:44.560 ⇒ 00:13:45.460 Awaish Kumar: Oh, dude.
138 00:13:46.050 ⇒ 00:13:47.340 Casie Aviles: Yeah.
139 00:13:47.770 ⇒ 00:13:50.480 Casie Aviles: There must be ways to refresh your tokens.
140 00:13:50.730 ⇒ 00:13:52.139 Awaish Kumar: Using API?
141 00:13:56.900 ⇒ 00:14:04.040 Awaish Kumar: Like, if we are using browser-based, like, Why then we have to… Use any other browser.
142 00:14:07.410 ⇒ 00:14:10.759 Samuel Roberts: Is BrowserBase able to do the login itself and, like, re-authenticate?
143 00:14:11.560 ⇒ 00:14:12.220 Uttam Kumaran: Yeah.
144 00:14:13.590 ⇒ 00:14:21.359 Casie Aviles: Mmm, yeah, so for browser-based, I believe… Let me recall what… How it works, but…
145 00:14:23.730 ⇒ 00:14:27.880 Casie Aviles: Yeah, I think that’s something I can take a look at once more,
146 00:14:31.240 ⇒ 00:14:32.270 Casie Aviles: Oh, okay.
147 00:14:36.820 ⇒ 00:14:39.530 Awaish Kumar: So, is the issue with the meta resolved?
148 00:14:42.100 ⇒ 00:14:51.390 Casie Aviles: Yeah, so for method, the issue there is just… is access, and we’re still looking through… The local report?
149 00:14:55.840 ⇒ 00:14:59.050 Awaish Kumar: I mean, that issue is not resolved.
150 00:14:59.380 ⇒ 00:15:00.590 Uttam Kumaran: No, it’s not resolved.
151 00:15:20.910 ⇒ 00:15:25.099 Awaish Kumar: Like, one of the things I have, like, thought
152 00:15:25.620 ⇒ 00:15:28.580 Awaish Kumar: About it in the past, like…
153 00:15:28.850 ⇒ 00:15:35.669 Awaish Kumar: To generate these scripts, like, these sheets, through Python scripts.
154 00:15:35.870 ⇒ 00:15:37.810 Awaish Kumar: Like, a fully…
155 00:15:38.690 ⇒ 00:15:48.639 Awaish Kumar: we have a script which basically have the layouts and everything, and it just populates everything in the… in the sheet. And to do that, the only…
156 00:15:49.170 ⇒ 00:15:55.399 Awaish Kumar: Only thing is that, like, there are some fields which are being filled out by us in that
157 00:15:55.560 ⇒ 00:16:00.069 Awaish Kumar: The spreadsheet, and for some of it, we don’t know the data source.
158 00:16:00.650 ⇒ 00:16:09.360 Awaish Kumar: And to do, like, to fully automate that process, you need to… we first need to identify all the data sources.
159 00:16:09.550 ⇒ 00:16:12.870 Awaish Kumar: From where the data is coming from for this…
160 00:16:13.390 ⇒ 00:16:22.490 Awaish Kumar: chargers also in cases, and then when we have all the data sources, then we can basically
161 00:16:23.690 ⇒ 00:16:29.410 Awaish Kumar: like, even create the Python script to have this exact layout, and just write
162 00:16:30.240 ⇒ 00:16:32.679 Awaish Kumar: No, you don’t do any manual phase.
163 00:16:36.650 ⇒ 00:16:38.200 Casie Aviles: Okay. Yeah.
164 00:16:38.420 ⇒ 00:16:39.260 Casie Aviles: So…
165 00:16:39.260 ⇒ 00:16:49.029 Awaish Kumar: Yeah, and for that, we need all the data sources that I don’t know if we even have a knowledge of other data sources, other different fees which are there.
166 00:16:51.900 ⇒ 00:17:00.760 Casie Aviles: Yes, for the other data sources, we don’t really touch those, like, for example, the promotional pack, these ones that are not in red.
167 00:17:02.860 ⇒ 00:17:04.140 Samuel Roberts: Someone else is filling those in.
168 00:17:04.339 ⇒ 00:17:05.540 Samuel Roberts: From other sources?
169 00:17:06.760 ⇒ 00:17:13.210 Casie Aviles: Yes, and for, like… There are, like, some existing, formulas for that.
170 00:17:13.349 ⇒ 00:17:14.240 Casie Aviles: So…
171 00:17:17.980 ⇒ 00:17:20.360 Casie Aviles: Yeah, we can… we can also take a look.
172 00:17:28.890 ⇒ 00:17:33.750 Casie Aviles: Yeah, like, there are some formulas here, for example, for some of the other parts, so…
173 00:17:34.100 ⇒ 00:17:37.899 Casie Aviles: The ones in red are the only ones that we’re constantly updating.
174 00:17:38.680 ⇒ 00:17:42.480 Casie Aviles: So these ones that I’ve… highlighted.
175 00:17:48.510 ⇒ 00:17:58.030 Samuel Roberts: Yeah, I mean, I think… I think it was getting to the… a good point that, like, what we’re… what we’re actually responsible for, like, we need… we should probably have a better sense of, like.
176 00:17:58.300 ⇒ 00:18:02.689 Samuel Roberts: Not a better sense, but, like, a better model of what it is, and then we can store it
177 00:18:02.970 ⇒ 00:18:10.069 Samuel Roberts: somewhere, and not necessarily here. We can then, if they want it filled here, we can fill it there, but rather have some…
178 00:18:10.690 ⇒ 00:18:16.940 Samuel Roberts: Ongoing historical record that is… Not just in these sheets, is my thought.
179 00:18:18.820 ⇒ 00:18:22.989 Uttam Kumaran: Well, yeah, I mean, my first point here is, like, we should be dumping stuff into.
180 00:18:22.990 ⇒ 00:18:23.830 Samuel Roberts: Exactly.
181 00:18:23.830 ⇒ 00:18:24.839 Uttam Kumaran: a data lake.
182 00:18:26.170 ⇒ 00:18:26.720 Uttam Kumaran: Right?
183 00:18:26.720 ⇒ 00:18:37.840 Awaish Kumar: Right now, I think what’s happening is we are getting data from different sources to some Google Sheets, and then we copy from those Google Sheets to final spreadsheet.
184 00:18:38.470 ⇒ 00:18:41.990 Uttam Kumaran: Yeah, but, like, yeah, that’s exactly it, yeah. But, like…
185 00:18:42.100 ⇒ 00:18:45.270 Uttam Kumaran: We don’t need to, like, we… one is…
186 00:18:45.870 ⇒ 00:18:53.079 Uttam Kumaran: I would at least like us to copy-paste into the final Google Sheet in, like, in a way that’s more of, like, copy-paste a bunch of raw data
187 00:18:54.080 ⇒ 00:18:55.859 Uttam Kumaran: Versus, like.
188 00:18:55.980 ⇒ 00:19:11.549 Uttam Kumaran: do it in one sheet, then move it over to the next sheet. Like, ideally, at minimum, this process should be just take a raw source, paste it in the new Google… in the last level Google Sheet, and then we just build formulas. Like, that’s the minimum.
189 00:19:12.200 ⇒ 00:19:12.920 Samuel Roberts: Right.
190 00:19:19.260 ⇒ 00:19:26.919 Samuel Roberts: Yeah, I mean, there’s, you know, there’s other issues with, like, the browser animation stuff, but I think at the base, like, there’s a lot of… where is it? Hold on, I’m looking at the…
191 00:19:27.320 ⇒ 00:19:32.670 Samuel Roberts: Video and not the actual… Awesome. A lot of the Go ahead.
192 00:19:32.970 ⇒ 00:19:39.819 Demilade Agboola: I was gonna say, my Zoom was acting up initially, but I head’s off about backfilling, and I’m wondering, like, why are we backfilling?
193 00:19:41.750 ⇒ 00:19:46.539 Casie Aviles: Yeah, so I believe that’s for what they call the 7-day attribution.
194 00:19:47.710 ⇒ 00:19:52.400 Casie Aviles: So, like I did mention, they wanted to kind of see, like, how the…
195 00:19:53.010 ⇒ 00:19:59.350 Casie Aviles: Like, a specific campaign they deployed is performing, like, for a week, for a week, like.
196 00:19:59.700 ⇒ 00:20:03.730 Casie Aviles: So they would have, like, snapshots per day, and they want to be able to see that.
197 00:20:03.860 ⇒ 00:20:06.140 Casie Aviles: And see if it’s increasing…
198 00:20:06.630 ⇒ 00:20:08.729 Casie Aviles: Like, for the revenue, or… yeah.
199 00:20:09.440 ⇒ 00:20:12.409 Casie Aviles: That’s, like, my rough understanding of that.
200 00:20:12.980 ⇒ 00:20:16.289 Casie Aviles: Of the… of why we do the 7-day backfill.
201 00:20:19.310 ⇒ 00:20:21.549 Demilade Agboola: Do they do, like, weekly reports of…
202 00:20:21.550 ⇒ 00:20:23.900 Casie Aviles: Yeah, they do weekly reports of that.
203 00:20:25.670 ⇒ 00:20:27.109 Demilade Agboola: Y-yes, but…
204 00:20:27.370 ⇒ 00:20:34.729 Demilade Agboola: My question is, how… where does the data live, and why are we backfilling, is my question. Like, do we get it every day?
205 00:20:34.860 ⇒ 00:20:43.539 Demilade Agboola: Or is it a thing of, like, they only get it once every 7 days, so… I’m curious as to why that backfilling process is needed, in terms.
206 00:20:43.540 ⇒ 00:20:52.080 Casie Aviles: Yeah, we do it… We do it every day, and the reason is, like, the… for example.
207 00:20:52.300 ⇒ 00:20:54.250 Casie Aviles: What would we need to go about?
208 00:20:54.680 ⇒ 00:20:59.909 Casie Aviles: So, for example, the email, if you get it for, like, today, sometimes it would just be…
209 00:21:01.040 ⇒ 00:21:02.710 Casie Aviles: At a lower value.
210 00:21:05.210 ⇒ 00:21:09.639 Uttam Kumaran: It takes time, it takes time to, like, basically fill in.
211 00:21:10.120 ⇒ 00:21:16.420 Samuel Roberts: Right, so, like, it’s just a snapshot at any given time, and so when we do it on that… at that morning, we’re gonna get whatever.
212 00:21:16.620 ⇒ 00:21:18.040 Demilade Agboola: You want a lot of money?
213 00:21:19.380 ⇒ 00:21:20.190 Samuel Roberts: I’m sorry?
214 00:21:20.370 ⇒ 00:21:22.639 Demilade Agboola: You get what it was at that morning, basically.
215 00:21:22.940 ⇒ 00:21:37.880 Samuel Roberts: Right, and so every morning, it’s just all the new campaigns are new, and all the campaigns that are a few days older, and so, like, it’s just… there’s not a good way to get better data, I think, out of this, except for doing it that way, because we don’t have access to the APIs.
216 00:21:37.880 ⇒ 00:21:40.470 Demilade Agboola: Basically, 4. Oh, gotcha.
217 00:21:40.700 ⇒ 00:21:49.730 Samuel Roberts: DoorDash, Uber Ads, and Uber Eats. So it’s really just pulling from those… Those, their dashboards online.
218 00:21:51.510 ⇒ 00:21:53.820 Samuel Roberts: But even with that, I feel like if we’re gonna just…
219 00:21:54.050 ⇒ 00:22:01.939 Samuel Roberts: export this every… if we could just do this, run something every day that is pulling all of those, putting it into, like, a data lake, like you’re saying, UTOM.
220 00:22:02.440 ⇒ 00:22:07.729 Samuel Roberts: Does that… After 7 days, Take care of doing any kind of backfill that way?
221 00:22:07.990 ⇒ 00:22:08.670 Uttam Kumaran: Yeah.
222 00:22:09.000 ⇒ 00:22:13.070 Demilade Agboola: Yeah, because we usually run a script, if we have, like, a script run on it, we should be able to do that.
223 00:22:14.070 ⇒ 00:22:14.660 Uttam Kumaran: Yeah.
224 00:22:14.970 ⇒ 00:22:15.790 Uttam Kumaran: I will.
225 00:22:18.610 ⇒ 00:22:26.710 Samuel Roberts: So then I think the… my… from my… like, the browser automation is its own kind of issue with not having the APIs, but we can…
226 00:22:26.870 ⇒ 00:22:30.449 Samuel Roberts: work on that, I guess, a little bit, but it’s still gonna be a…
227 00:22:30.740 ⇒ 00:22:35.669 Samuel Roberts: weak point, I feel like, in general, but the rest of it, I feel like it should just be going…
228 00:22:36.030 ⇒ 00:22:38.020 Samuel Roberts: Into a source that we can then…
229 00:22:38.180 ⇒ 00:22:41.709 Samuel Roberts: do whatever we want with after the fact. Like, if they’re not…
230 00:22:42.190 ⇒ 00:22:46.479 Samuel Roberts: You know, if it has to go into that spreadsheet, great, but a lot of this, like, all these manual processes here…
231 00:22:46.760 ⇒ 00:22:47.850 Samuel Roberts: seem like…
232 00:22:48.850 ⇒ 00:22:53.629 Samuel Roberts: it should just be going into one source that we can pull from. I mean, I don’t know.
233 00:22:55.700 ⇒ 00:23:02.100 Uttam Kumaran: Yeah, I mean, I guess I… I mean, I’m happy to chime in, but I guess I’m more looking for, like, Demolade and Awish, like, what do you guys suggest?
234 00:23:02.310 ⇒ 00:23:03.410 Uttam Kumaran: We do here.
235 00:23:06.960 ⇒ 00:23:07.610 Awaish Kumar: Yeah.
236 00:23:10.110 ⇒ 00:23:13.529 Awaish Kumar: Like, I, I, like, I…
237 00:23:13.930 ⇒ 00:23:27.230 Awaish Kumar: Well, what I saw is, like, the scale of data isn’t that big, right? So what they are basically… what’s happening is the data comes from a Google, for example.
238 00:23:27.670 ⇒ 00:23:32.420 Awaish Kumar: Google Ads, and then it’s… If it goes,
239 00:23:33.140 ⇒ 00:23:36.079 Awaish Kumar: to a Google Sheet that is just, like, a…
240 00:23:36.290 ⇒ 00:23:41.479 Awaish Kumar: like, single row, single values which are being copied over. There’s not, not, like…
241 00:23:42.620 ⇒ 00:23:46.170 Awaish Kumar: No, like, transformations or anything after that.
242 00:23:47.640 ⇒ 00:23:51.369 Awaish Kumar: So I don’t know why we want to move to data lag.
243 00:23:52.360 ⇒ 00:23:54.660 Awaish Kumar: But what I would… Suggest…
244 00:23:54.660 ⇒ 00:23:58.229 Uttam Kumaran: It’s because we’re gonna use this data for other stuff, right? Like…
245 00:23:59.460 ⇒ 00:24:07.479 Uttam Kumaran: the data from Braze, the data from DoorDash, like, we don’t only want to use it for this, like, it needs to be in some environment where we can query.
246 00:24:08.040 ⇒ 00:24:11.590 Uttam Kumaran: Gotta think a little bit bigger, like, the… that…
247 00:24:11.780 ⇒ 00:24:16.419 Uttam Kumaran: The output you’re seeing in terms of the trackers is something we have to support.
248 00:24:16.560 ⇒ 00:24:17.300 Uttam Kumaran: But…
249 00:24:17.430 ⇒ 00:24:30.600 Uttam Kumaran: building a one-to-one pipeline with just one report is useless, right? Like, why don’t we dump all of it into a data lake, build, like, the raw export they need for one report, and shove that into the report.
250 00:24:30.970 ⇒ 00:24:35.559 Uttam Kumaran: and then build… build, like, logic on top of that in Excel.
251 00:24:36.040 ⇒ 00:24:45.879 Uttam Kumaran: Right? And then that way, we’re gonna need to run more analysis on FDA, on paid media, on all this stuff anyways, eventually, so we need all this data.
252 00:24:47.690 ⇒ 00:24:48.589 Awaish Kumar: Go on.
253 00:24:49.940 ⇒ 00:24:50.245 Awaish Kumar: Oh.
254 00:24:50.550 ⇒ 00:24:53.629 Demilade Agboola: My feeling is, if progressive, if we need all this data.
255 00:24:54.420 ⇒ 00:24:56.779 Demilade Agboola: But we still don’t have access to the APIs.
256 00:24:57.530 ⇒ 00:24:58.699 Demilade Agboola: It appears we might need.
257 00:24:58.700 ⇒ 00:25:00.860 Uttam Kumaran: No, no, there’s no APIs for these.
258 00:25:02.210 ⇒ 00:25:10.640 Samuel Roberts: Yeah, they’re, like, they have… they have, like, some documentation of, like, you know, testing thing, but it’s not anything that, like, just, we don’t have API, there’s no API to hit.
259 00:25:11.540 ⇒ 00:25:13.420 Uttam Kumaran: Yeah, so we have to scrape and bring it in.
260 00:25:13.420 ⇒ 00:25:14.720 Samuel Roberts: Which is… yeah.
261 00:25:15.910 ⇒ 00:25:20.520 Awaish Kumar: So, like, some of the data we are already storing in, snowflake.
262 00:25:20.940 ⇒ 00:25:22.179 Awaish Kumar: Yeah.
263 00:25:22.180 ⇒ 00:25:25.059 Uttam Kumaran: But again, that was just decided randomly, like.
264 00:25:25.160 ⇒ 00:25:29.029 Samuel Roberts: This is in our snowflake, which is a huge mistake.
265 00:25:29.220 ⇒ 00:25:32.880 Uttam Kumaran: Like, we can’t have client data, in our stuff.
266 00:25:33.060 ⇒ 00:25:34.990 Uttam Kumaran: So this is where, like, basically.
267 00:25:34.990 ⇒ 00:25:35.960 Samuel Roberts: It’s like…
268 00:25:35.970 ⇒ 00:25:42.660 Uttam Kumaran: This and this, like, this whole middle piece needs to be, like, a data warehouse or something.
269 00:25:42.660 ⇒ 00:25:43.450 Samuel Roberts: Yeah.
270 00:25:43.750 ⇒ 00:25:48.310 Uttam Kumaran: And I need to go get… I need to go get… I need to just have an architecture of what we’re proposing
271 00:25:48.450 ⇒ 00:25:59.010 Uttam Kumaran: And we should, like, again, this is where, like, what I’m kind of hoping to hear is that, like, we land everything somewhere, and then we move it out to wherever it’s needed, right? This is just one customer.
272 00:25:59.310 ⇒ 00:26:04.050 Uttam Kumaran: Like, Amber is another customer, Demolata is another customer of this source data.
273 00:26:06.240 ⇒ 00:26:07.660 Awaish Kumar: Yeah, I get it.
274 00:26:08.160 ⇒ 00:26:08.960 Demilade Agboola: Yeah.
275 00:26:09.680 ⇒ 00:26:13.290 Demilade Agboola: So I think at this point, if we have a list of…
276 00:26:15.230 ⇒ 00:26:17.450 Demilade Agboola: If you have a list of all the sources.
277 00:26:17.700 ⇒ 00:26:21.850 Demilade Agboola: We can basically say, like, try and split it into the ones we have to scrape.
278 00:26:22.340 ⇒ 00:26:25.709 Awaish Kumar: And the ones that we have, too, that we can ingest directly.
279 00:26:26.730 ⇒ 00:26:31.459 Demilade Agboola: Once we have that, like, list, we can try to create the pipelines and
280 00:26:32.030 ⇒ 00:26:37.890 Demilade Agboola: like, both of them. The ones that we can, like, ingest directly into the warehouse, and the ones that we need to scrape.
281 00:26:38.440 ⇒ 00:26:43.880 Demilade Agboola: And then… I had to see…
282 00:26:46.320 ⇒ 00:26:54.089 Demilade Agboola: Because it appears that the things… some things downstream are just basically aggregations of the things that… of the different sources.
283 00:26:56.110 ⇒ 00:26:58.279 Demilade Agboola: would I be right with that assumption?
284 00:27:01.130 ⇒ 00:27:06.699 Demilade Agboola: So things like the spreadsheets that we are trying to get data from, they seem to be, like.
285 00:27:07.890 ⇒ 00:27:13.929 Demilade Agboola: like, Daily Impact scorecard, for instance, it appears to be… An application.
286 00:27:15.690 ⇒ 00:27:16.390 Casie Aviles: Yes.
287 00:27:16.950 ⇒ 00:27:21.410 Demilade Agboola: Yeah, so if we get the source data, we don’t necessarily need to…
288 00:27:23.430 ⇒ 00:27:30.680 Demilade Agboola: We would have the information we need for the day, like, to create the daily impact scorecard by ourselves, and provide the data that people need for that.
289 00:27:33.230 ⇒ 00:27:40.749 Casie Aviles: Yeah, yeah, that makes sense. So, like, yeah, we just get, like, the data, and then, yeah, that’s what Otong was mentioned earlier.
290 00:27:40.750 ⇒ 00:27:48.459 Uttam Kumaran: Can we… can we work on, like, a new diagram, like, while we’re on the call? Like, let’s just make one, Casey, like, to the right of this.
291 00:27:48.800 ⇒ 00:27:55.760 Uttam Kumaran: And I think, like, this is where I think Demolata, in a way, can we… can I just get your help and see, like, can we architect?
292 00:27:56.050 ⇒ 00:27:58.539 Uttam Kumaran: Like, what the new version could look like, basically?
293 00:27:58.900 ⇒ 00:28:03.249 Uttam Kumaran: Casey, you can just… you can literally just copy-paste this whole thing to the right, and we can just edit it.
294 00:28:03.250 ⇒ 00:28:03.840 Samuel Roberts: Yeah.
295 00:28:05.010 ⇒ 00:28:12.210 Awaish Kumar: Yeah, what I… what I can, suggest is that we are already, using polyatomic.
296 00:28:12.400 ⇒ 00:28:15.419 Awaish Kumar: For some of the connectors, and we can move
297 00:28:15.530 ⇒ 00:28:19.130 Awaish Kumar: Almost, all the characters which we can
298 00:28:19.600 ⇒ 00:28:22.339 Awaish Kumar: For which we can use polyatomic, we should use that.
299 00:28:22.510 ⇒ 00:28:26.249 Awaish Kumar: And only the script part, which is, like, DoorDash or UberAds.
300 00:28:26.370 ⇒ 00:28:45.170 Awaish Kumar: we can have our own automations, because that’s web… that’s scrapping the data. And after that, we can use MotherDuck. I’m not going… thinking of data lakes here, because the kind of data we are getting is… isn’t that big, and also is structured
301 00:28:45.270 ⇒ 00:28:51.520 Awaish Kumar: And, is being used. So I don’t see any of the extra data with… Which, which,
302 00:28:51.840 ⇒ 00:29:00.290 Awaish Kumar: isn’t being used and will be used in future, or something like that. We don’t have a schema of read requirements right now, and the volume of data, isn’t…
303 00:29:00.530 ⇒ 00:29:05.450 Uttam Kumaran: But, like, well, I guess, like, this is where I’m, like, concerned about, like, I don’t… I… I’m… I care about…
304 00:29:05.770 ⇒ 00:29:11.270 Uttam Kumaran: For example, if Demolade gets an ask to, like, start modeling this for other use cases.
305 00:29:12.330 ⇒ 00:29:15.920 Uttam Kumaran: like, what are we gonna do, right? We need to have the data saved somewhere.
306 00:29:16.200 ⇒ 00:29:19.920 Awaish Kumar: Yeah, yeah, I’m suggesting having mother dog, or warehouse.
307 00:29:20.080 ⇒ 00:29:20.850 Uttam Kumaran: Okay.
308 00:29:21.070 ⇒ 00:29:21.800 Awaish Kumar: Listen.
309 00:29:22.000 ⇒ 00:29:23.120 Uttam Kumaran: I see what you mean, I see what you mean.
310 00:29:23.120 ⇒ 00:29:30.679 Awaish Kumar: scope of the data, and it is all structured data. That’s why it can be easily… easily can live in Mother Duck.
311 00:29:30.940 ⇒ 00:29:35.620 Awaish Kumar: Without having, like, extra complexity of managing a data lake there.
312 00:29:36.620 ⇒ 00:29:37.910 Uttam Kumaran: I see what you mean, okay.
313 00:29:39.500 ⇒ 00:29:44.509 Samuel Roberts: Can I give a little clarification, like, what the difference there is? Because I don’t… just not being a data guy, I don’t necessarily understand the, like…
314 00:29:44.700 ⇒ 00:29:46.970 Samuel Roberts: What makes a lake versus, like, something a mother duck?
315 00:29:47.300 ⇒ 00:29:49.060 Samuel Roberts: is just… how…
316 00:29:49.060 ⇒ 00:29:49.500 Awaish Kumar: Oh, no.
317 00:29:49.500 ⇒ 00:29:50.940 Samuel Roberts: Well, yeah, what is the difference there?
318 00:29:52.380 ⇒ 00:30:02.439 Awaish Kumar: Like, when we have structured data, and most of it being in use, we use something like data… some databases.
319 00:30:02.740 ⇒ 00:30:12.870 Awaish Kumar: And so, that’s, like, mother death. And when you have, like, a lot of data, hundreds of terabytes of data, which isn’t being used.
320 00:30:12.980 ⇒ 00:30:14.190 Awaish Kumar: Got it.
321 00:30:14.550 ⇒ 00:30:16.850 Awaish Kumar: And maybe some of it
322 00:30:17.100 ⇒ 00:30:25.419 Awaish Kumar: gets used by someone in the future, so kind of schema on read, like, when we need it, we will get the data. That’s just keeping the…
323 00:30:25.420 ⇒ 00:30:25.980 Samuel Roberts: Got it.
324 00:30:25.980 ⇒ 00:30:36.600 Awaish Kumar: all the historical data there, and without… if anyone… without knowing if anyone is using or not. So, yeah. So, in that case, we need data lakes.
325 00:30:36.710 ⇒ 00:30:45.590 Awaish Kumar: But then after… Yeah, and then, like, when we use DataLex, then we still need some data warehouse, because
326 00:30:45.780 ⇒ 00:30:52.860 Awaish Kumar: For… with the data which will be used for analysis or anything, then it will move from data lake to
327 00:30:52.990 ⇒ 00:30:54.729 Awaish Kumar: Some data models.
328 00:30:55.420 ⇒ 00:30:59.500 Samuel Roberts: Got it. Okay, yeah, it’s terminology I’m not as familiar with, I guess, but that makes sense.
329 00:31:02.850 ⇒ 00:31:07.789 Awaish Kumar: So, yeah, mother duck, and the Polytomic, and the…
330 00:31:08.100 ⇒ 00:31:14.370 Awaish Kumar: The VIN million scripts which we already have, And…
331 00:31:15.220 ⇒ 00:31:18.800 Demilade Agboola: So, we don’t want to use Snowflake for them, because they’re a Snowflake.
332 00:31:19.390 ⇒ 00:31:26.600 Uttam Kumaran: It’s… no, no, it’s our Snowflake, so I think, Casey, correct me if I’m wrong, you just shoved it into Snowflake, because we just have to build it really quick, right?
333 00:31:27.340 ⇒ 00:31:28.090 Awaish Kumar: Yeah, yeah.
334 00:31:28.090 ⇒ 00:31:31.690 Uttam Kumaran: So we need to, one, we need to move to completely their infrastructure.
335 00:31:31.900 ⇒ 00:31:38.820 Uttam Kumaran: Right, so right now we’re using our polyatomic, our snowflake, we can’t… like, our internal stuff, we can’t use that anymore.
336 00:31:39.670 ⇒ 00:31:41.769 Demilade Agboola: And they use Mother Dog by themselves?
337 00:31:41.770 ⇒ 00:31:43.760 Uttam Kumaran: That’s our proposal.
338 00:31:44.100 ⇒ 00:31:45.550 Awaish Kumar: Well, that’s proposed.
339 00:31:45.800 ⇒ 00:31:47.620 Uttam Kumaran: They have no data warehouse at all.
340 00:31:47.920 ⇒ 00:31:50.620 Demilade Agboola: Is there any reason why I won’t mother talk over Snowflake?
341 00:31:50.940 ⇒ 00:31:57.890 Uttam Kumaran: No reason. It’s just… It’s just simple, I guess. Yeah, we can propose…
342 00:31:58.350 ⇒ 00:32:00.839 Uttam Kumaran: We can propose Snowflake if you want, I just…
343 00:32:01.240 ⇒ 00:32:03.920 Uttam Kumaran: I don’t know, I guess my overall thinking is, like, it’s…
344 00:32:04.110 ⇒ 00:32:08.320 Uttam Kumaran: It’s kind of overcomplicated for just simple tables, right? Like, what do you think?
345 00:32:09.710 ⇒ 00:32:16.490 Awaish Kumar: Yeah, and I know that we have very limited data, And,
346 00:32:18.150 ⇒ 00:32:24.630 Awaish Kumar: Yeah, it can easily… yeah, but they’re, like, really simple to load files and stuff like that.
347 00:32:26.380 ⇒ 00:32:28.919 Demilade Agboola: Yeah, I think, yeah, I think Mother Duck is fine.
348 00:32:29.990 ⇒ 00:32:35.730 Uttam Kumaran: Yeah, and then, yeah, I agree, and then if… once everything gets into dbt anyways, it’s… it’ll be chill.
349 00:32:37.950 ⇒ 00:32:42.580 Uttam Kumaran: like, the UI isn’t great, but, like, we can put a… some layer on top of that, so…
350 00:32:43.970 ⇒ 00:32:47.139 Awaish Kumar: And what volume of data, do we want to use DB?
351 00:32:47.140 ⇒ 00:32:49.559 Demilade Agboola: Or we could just use stop procedure.
352 00:32:51.050 ⇒ 00:32:55.920 Uttam Kumaran: The only point I would say against store procedures, I just want to have it as code somewhere, dude.
353 00:32:59.250 ⇒ 00:33:01.600 Demilade Agboola: I’m sure, but it’s just, like, the volume…
354 00:33:02.240 ⇒ 00:33:10.189 Demilade Agboola: I guess the volume is what I’m just looking at. It’s really… it doesn’t seem like a lot of data, and the… doesn’t seem like there’ll be a lot of business logic to apply.
355 00:33:10.680 ⇒ 00:33:16.899 Uttam Kumaran: Yeah, I don’t… I don’t mind you doing a store procedures, but, like, I just want to make sure that we have it in a version-controlled environment.
356 00:33:17.400 ⇒ 00:33:17.929 Uttam Kumaran: You know?
357 00:33:18.760 ⇒ 00:33:25.070 Uttam Kumaran: So, like, the easiest thing you could do is you could just run, using GitHub Actions,
358 00:33:25.200 ⇒ 00:33:29.950 Uttam Kumaran: Or, yeah, I don’t know, I feel like… Yeah, either way…
359 00:33:29.950 ⇒ 00:33:31.429 Awaish Kumar: double what for DBTQ.
360 00:33:32.810 ⇒ 00:33:34.680 Uttam Kumaran: Yeah, maybe we just keep it consistent.
361 00:33:35.000 ⇒ 00:33:43.920 Awaish Kumar: Yeah, it’s like, we can easily, like, put all the code in one place, in a standardized format, and version controlled.
362 00:33:45.020 ⇒ 00:33:49.669 Demilade Agboola: Yeah, like, yeah, I haven’t issues with DBT, it’s just, I guess the skin is just pretty…
363 00:33:50.940 ⇒ 00:33:51.690 Demilade Agboola: Fish and soup.
364 00:33:51.690 ⇒ 00:33:59.160 Uttam Kumaran: But this is just the start. I think this is just… this is just the start. Like, I… we’re gonna start to uncover more and more stuff about them, so… it’ll get bigger.
365 00:33:59.160 ⇒ 00:33:59.810 Demilade Agboola: Good.
366 00:33:59.960 ⇒ 00:34:01.300 Demilade Agboola: Did that make sense, then?
367 00:34:02.350 ⇒ 00:34:07.719 Awaish Kumar: And then, like, are we not suggesting, recommending any BI tool?
368 00:34:09.250 ⇒ 00:34:12.789 Uttam Kumaran: In order for, like, we have to first support these.
369 00:34:13.060 ⇒ 00:34:16.939 Uttam Kumaran: Before I can go… we can go think about making a recommendation.
370 00:34:17.340 ⇒ 00:34:18.239 Awaish Kumar: Right.
371 00:34:18.440 ⇒ 00:34:27.330 Uttam Kumaran: Because if I go and change that, then we’re changing so many things. So at this point, all I want to do is support these, and then we can go make a recommendation for that, too.
372 00:34:29.989 ⇒ 00:34:31.679 Uttam Kumaran: Because it looks like they have Looker.
373 00:34:32.570 ⇒ 00:34:34.500 Awaish Kumar: Yeah, but right now, we are, like…
374 00:34:34.730 ⇒ 00:34:39.789 Awaish Kumar: To them, we are… we are already, like, have this flow, right?
375 00:34:40.239 ⇒ 00:34:43.479 Awaish Kumar: For them, like, we already have sheets which are working for them.
376 00:34:43.639 ⇒ 00:34:46.259 Awaish Kumar: Data is coming in, it’s being populated.
377 00:34:51.199 ⇒ 00:34:54.119 Uttam Kumaran: Yeah. No, no, no, but where it’s manual right now.
378 00:34:55.989 ⇒ 00:35:01.389 Awaish Kumar: like, none of… like, we… we have semi… semi-automated, I would say, but for them…
379 00:35:01.390 ⇒ 00:35:05.430 Uttam Kumaran: What are you gonna put the BI tool on? You’re gonna put the BI tool on the Google Sheets?
380 00:35:07.660 ⇒ 00:35:08.600 Awaish Kumar: What I… what I’m…
381 00:35:08.600 ⇒ 00:35:09.119 Demilade Agboola: on a duck.
382 00:35:10.320 ⇒ 00:35:22.850 Awaish Kumar: Yeah. When we have data in Mother Duck, we are going to have some Tableau reports, which are going to maybe… Tableau or Looker, whatever, like, which are going to basically show some of the tables which they are seeing in Google Sheet.
383 00:35:27.420 ⇒ 00:35:29.820 Samuel Roberts: Yeah, I think the idea is that… Oh, go ahead.
384 00:35:29.820 ⇒ 00:35:38.170 Demilade Agboola: Instead of us using the… The cheats that they’re currently using as the final end-all decision maker.
385 00:35:38.280 ⇒ 00:35:41.059 Demilade Agboola: We can replace that with the BI tool.
386 00:35:41.060 ⇒ 00:35:42.960 Awaish Kumar: I was really having a good time with that doc.
387 00:35:43.190 ⇒ 00:35:44.800 Demilade Agboola: In an easy-to-use format.
388 00:35:45.630 ⇒ 00:35:53.810 Samuel Roberts: Yeah, because one of the other issues was that they were pulling reports anyway from these sources because they’re not getting everything they need out of these spreadsheets, it seems.
389 00:35:54.380 ⇒ 00:35:55.649 Samuel Roberts: Is that true?
390 00:35:56.340 ⇒ 00:36:00.310 Uttam Kumaran: Well, I just can’t, like, you guys have to put it… put yourself in the client’s shoes, like…
391 00:36:00.330 ⇒ 00:36:03.170 Samuel Roberts: You, we can’t change multiple things at once.
392 00:36:03.240 ⇒ 00:36:10.659 Uttam Kumaran: So our first thing is, right now, we’re still errors with just updating the sheets frequently.
393 00:36:10.790 ⇒ 00:36:13.009 Uttam Kumaran: Let’s just solve the first problem.
394 00:36:13.240 ⇒ 00:36:17.070 Uttam Kumaran: Then we can start to diagnose, do they have a BI reporting issue?
395 00:36:17.210 ⇒ 00:36:23.970 Uttam Kumaran: Like, but even if we solve that, we still need this stuff, so… that’s my… that’s what I… that’s what I’m saying.
396 00:36:30.750 ⇒ 00:36:31.910 Awaish Kumar: Who is.
397 00:36:32.730 ⇒ 00:36:36.740 Uttam Kumaran: And also, like, I guess, Casey, they have a Looker? They have Looker?
398 00:36:37.650 ⇒ 00:36:39.340 Casie Aviles: Yeah, for.
399 00:36:39.790 ⇒ 00:36:41.640 Uttam Kumaran: Is it Looker Studio?
400 00:36:43.490 ⇒ 00:36:47.699 Casie Aviles: I don’t know, let me show… yeah, Looker Studio here.
401 00:36:48.210 ⇒ 00:36:49.289 Demilade Agboola: Yeah, it looks serious trash.
402 00:36:49.970 ⇒ 00:36:51.580 Uttam Kumaran: And what is this sitting on?
403 00:36:51.580 ⇒ 00:36:53.680 Samuel Roberts: Yeah, what is, what is this?
404 00:36:55.080 ⇒ 00:36:57.660 Samuel Roberts: But where is this data coming from, you know?
405 00:36:58.950 ⇒ 00:37:01.140 Casie Aviles: What is this, what has been coming out of here?
406 00:37:03.890 ⇒ 00:37:09.670 Casie Aviles: This is, like, this is the Google data. Google data, and then metadata, and then from Facebook.
407 00:37:09.670 ⇒ 00:37:12.250 Demilade Agboola: Can you scroll through the different pages?
408 00:37:12.500 ⇒ 00:37:17.110 Demilade Agboola: Opt on top left, and you, like, scroll to the top left.
409 00:37:17.110 ⇒ 00:37:18.200 Samuel Roberts: So, yeah.
410 00:37:22.000 ⇒ 00:37:24.889 Demilade Agboola: Oh, no, no, no, I’m sorry, like, can you say, yeah, that…
411 00:37:26.410 ⇒ 00:37:30.369 Casie Aviles: Oh, it’s in… sorry, it’s in… It’s in Filipino.
412 00:37:31.340 ⇒ 00:37:32.270 Demilade Agboola: Yeah.
413 00:37:33.200 ⇒ 00:37:41.529 Demilade Agboola: I just want to see the different pages, so there’s 9 out of 10, and you see, like, what’s… it’s just to be sure, it’s only, like, the Google data. Google Ads seems to just be Google Ads data.
414 00:37:43.270 ⇒ 00:37:46.330 Demilade Agboola: Which makes sense, because it’s a quick connection.
415 00:37:46.330 ⇒ 00:37:47.330 Samuel Roberts: Right…
416 00:37:47.330 ⇒ 00:37:49.260 Demilade Agboola: It’s all within the Google ecosystem.
417 00:37:53.950 ⇒ 00:37:59.740 Uttam Kumaran: Oh, okay, so this is probably, like, all Google Ads outputs, right?
418 00:38:00.150 ⇒ 00:38:01.280 Uttam Kumaran: Perfect, nothing, anything else.
419 00:38:01.900 ⇒ 00:38:05.929 Demilade Agboola: That’s what I wanted to just be sure. Yeah, I just wanted to show what else they had.
420 00:38:09.870 ⇒ 00:38:12.569 Uttam Kumaran: Okay, so can we complete these,
421 00:38:14.330 ⇒ 00:38:17.320 Uttam Kumaran: Can we complete this, these…
422 00:38:18.290 ⇒ 00:38:21.800 Uttam Kumaran: Diagram, just for the new version.
423 00:38:22.070 ⇒ 00:38:24.489 Uttam Kumaran: I’m just, like, finishing this up.
424 00:38:25.210 ⇒ 00:38:27.820 Uttam Kumaran: So… Yeah, let’s…
425 00:38:27.820 ⇒ 00:38:30.680 Awaish Kumar: They are bringing in Facebook data here somewhere.
426 00:38:30.680 ⇒ 00:38:37.439 Demilade Agboola: Yeah, I need to ask, but see, this is the thing, I need to… I need to have a clear view before I go ask them for access.
427 00:38:38.230 ⇒ 00:38:44.459 Uttam Kumaran: So I need to share, like, hey, here’s how we’re doing it now, like, this is what it’s gonna impact if I can get direct access. But, like, again.
428 00:38:44.670 ⇒ 00:38:50.960 Uttam Kumaran: if I need… if we need that, then we have to have a polyatomic decision. So, like, I can’t do just too many moving parts right now.
429 00:38:52.440 ⇒ 00:38:57.599 Demilade Agboola: It appears… so, like, if you go back to the diagram, Like, the architectural diagram.
430 00:38:57.820 ⇒ 00:39:07.290 Demilade Agboola: I think if we… once we have the list of all the sources, the ones that need… that can be ingested directly, we can set that up with Polytomic, do Mother Doc.
431 00:39:08.070 ⇒ 00:39:14.049 Demilade Agboola: And the ones that don’t, we might need to set up scripts with DAGs that run at a certain time.
432 00:39:14.770 ⇒ 00:39:15.690 Awaish Kumar: Oh, we already have…
433 00:39:15.690 ⇒ 00:39:16.350 Demilade Agboola: Interesting.
434 00:39:17.830 ⇒ 00:39:23.980 Awaish Kumar: Yeah, some of it, we are already using polyatomic, and for some of it, We already have scripts.
435 00:39:25.100 ⇒ 00:39:31.909 Awaish Kumar: But right now, we are pointing to a Google Sheet, we just need to change the destination part to MotherBank.
436 00:39:33.440 ⇒ 00:39:35.299 Demilade Agboola: Okay, that makes sense then, mate.
437 00:39:35.520 ⇒ 00:39:37.390 Demilade Agboola: That’s basically what we need to do.
438 00:39:38.110 ⇒ 00:39:43.579 Demilade Agboola: And let’s just make everything land in Mother Doc, in the appropriate schemas for each of them.
439 00:39:44.890 ⇒ 00:39:46.850 Demilade Agboola: We can have the raw data.
440 00:39:50.130 ⇒ 00:39:55.310 Uttam Kumaran: So, for Braze Currents, Like, this… it’s gonna be like this, right, Sam?
441 00:39:55.850 ⇒ 00:40:04.819 Samuel Roberts: Yeah, it just outputs to a few different things. S3 was the one that I had access to, so that’s where it went. They have some documentation, I’m not sure exactly.
442 00:40:04.820 ⇒ 00:40:06.070 Uttam Kumaran: Where else can I go?
443 00:40:06.730 ⇒ 00:40:10.060 Samuel Roberts: Let me… Okay.
444 00:40:14.210 ⇒ 00:40:16.770 Samuel Roberts: Let’s see, their partners were…
445 00:40:20.190 ⇒ 00:40:27.020 Samuel Roberts: Available partners… Yeah, S3, Google Cloud Storage, or Azure Bob Storage.
446 00:40:27.220 ⇒ 00:40:34.650 Samuel Roberts: And then there was other things that they talked about platforms to, like, look at it with and behave, yeah. But it was basically just, like, yeah, just dumping it out.
447 00:40:35.160 ⇒ 00:40:37.079 Demilade Agboola: That’s a length.
448 00:40:39.760 ⇒ 00:40:40.720 Uttam Kumaran: Yeah.
449 00:40:41.250 ⇒ 00:40:43.430 Uttam Kumaran: Like, they don’t connect to a direct warehouse.
450 00:40:43.430 ⇒ 00:40:44.470 Demilade Agboola: one option there.
451 00:40:44.470 ⇒ 00:40:49.620 Samuel Roberts: Yeah, no, they had… they had something, where they talked about how they used it, and it looked like they…
452 00:40:50.930 ⇒ 00:40:55.460 Samuel Roberts: they went from S3… to Snowflake.
453 00:40:57.430 ⇒ 00:41:03.490 Demilade Agboola: I mean, definitely, you can always go from the leak to warehouse, but it’s just a bit surprised that there isn’t a direct connection to…
454 00:41:06.590 ⇒ 00:41:07.840 Demilade Agboola: me warehoused.
455 00:41:10.450 ⇒ 00:41:13.859 Demilade Agboola: Well, that’s fine. So, effectively, once we have that data in
456 00:41:14.000 ⇒ 00:41:17.430 Demilade Agboola: Another source would have to be… to occurrence to…
457 00:41:17.710 ⇒ 00:41:20.599 Demilade Agboola: S3, and then S3 to the warehouse as well.
458 00:41:21.660 ⇒ 00:41:22.200 Samuel Roberts: Yeah.
459 00:41:25.910 ⇒ 00:41:30.390 Uttam Kumaran: Oh, but Casey, Uber… Uber Ads? Oh, this is just DoorDash, right?
460 00:41:30.450 ⇒ 00:41:31.519 Samuel Roberts: Yeah. I guess.
461 00:41:32.440 ⇒ 00:41:32.880 Casie Aviles: Yeah, yeah.
462 00:41:38.660 ⇒ 00:41:39.709 Uttam Kumaran: Is this the case?
463 00:41:40.350 ⇒ 00:41:41.440 Samuel Roberts: Yes, yes.
464 00:41:41.440 ⇒ 00:41:43.180 Uttam Kumaran: And then what’s going through Polyatomic?
465 00:41:44.830 ⇒ 00:41:45.420 Casie Aviles: Google Ads.
466 00:41:45.420 ⇒ 00:41:47.070 Samuel Roberts: Atomic Google, yeah.
467 00:41:47.070 ⇒ 00:41:48.949 Uttam Kumaran: But then what’s going through Looker Report?
468 00:41:49.800 ⇒ 00:41:51.710 Uttam Kumaran: Wait, oh, wait, the what?
469 00:41:54.680 ⇒ 00:41:56.649 Uttam Kumaran: Like, how does this get connected here?
470 00:41:57.420 ⇒ 00:41:59.359 Samuel Roberts: On the other one, it says removal.
471 00:42:00.490 ⇒ 00:42:01.280 Samuel Roberts: to apologize.
472 00:42:01.280 ⇒ 00:42:01.650 Casie Aviles: I can’.
473 00:42:01.650 ⇒ 00:42:03.250 Samuel Roberts: the Looker, yeah.
474 00:42:03.750 ⇒ 00:42:04.229 Uttam Kumaran: it’s redundant.
475 00:42:04.230 ⇒ 00:42:05.930 Casie Aviles: That’s what happens, yeah.
476 00:42:06.560 ⇒ 00:42:08.710 Casie Aviles: But then they also have the local report.
477 00:42:10.520 ⇒ 00:42:11.880 Uttam Kumaran: And we’re using both?
478 00:42:13.860 ⇒ 00:42:20.910 Casie Aviles: We’re using the Google export from Polyatomic, but for Facebook, we’re using the local report.
479 00:42:22.220 ⇒ 00:42:25.730 Uttam Kumaran: I see. Okay, so then I’m gonna delete this one.
480 00:42:27.110 ⇒ 00:42:28.180 Uttam Kumaran: Is that fair?
481 00:42:29.140 ⇒ 00:42:30.170 Casie Aviles: Yes, yes.
482 00:42:31.770 ⇒ 00:42:34.499 Awaish Kumar: What is this browser-based generate report?
483 00:42:36.220 ⇒ 00:42:38.719 Casie Aviles: Yeah, so what happens here is…
484 00:42:40.980 ⇒ 00:42:45.100 Casie Aviles: Yeah, let me show you the Uber Eats manager. So, we have Google…
485 00:42:46.960 ⇒ 00:42:53.830 Casie Aviles: Basically, I’m kind of mimicking the manual process of going to… reports…
486 00:42:53.940 ⇒ 00:42:56.200 Casie Aviles: And then it would create a report.
487 00:42:56.480 ⇒ 00:43:02.450 Casie Aviles: So the browser-based session would… Click this with… make sure this is…
488 00:43:04.020 ⇒ 00:43:07.899 Uttam Kumaran: Like, dude, browser base is, like, literally just headless browser-a-ish.
489 00:43:08.750 ⇒ 00:43:09.630 Awaish Kumar: Yeah, yeah.
490 00:43:10.070 ⇒ 00:43:10.660 Uttam Kumaran: Yeah.
491 00:43:11.480 ⇒ 00:43:17.229 Casie Aviles: Yeah, and then it just generates the report, basically. Like, there’s one… there’s two tasks for browser-based.
492 00:43:17.430 ⇒ 00:43:20.060 Awaish Kumar: One says generate report, another one…
493 00:43:20.530 ⇒ 00:43:29.659 Casie Aviles: Like… Yeah, because this kind of takes some time until it’s downloadable, so…
494 00:43:29.660 ⇒ 00:43:32.400 Awaish Kumar: Okay, other one is just copy-paste the…
495 00:43:35.630 ⇒ 00:43:39.610 Casie Aviles: Yeah, the others, they don’t really generate the report.
496 00:43:41.120 ⇒ 00:43:49.040 Casie Aviles: Okay. They basically scrape But I’m thinking if, yeah, maybe we should be doing the reports instead.
497 00:43:49.860 ⇒ 00:43:55.010 Casie Aviles: So we could… Get, like, a better, like, more of the data out.
498 00:43:55.230 ⇒ 00:43:59.639 Casie Aviles: Because if we’re just scraping it, It’s just gonna go through…
499 00:43:59.860 ⇒ 00:44:03.250 Casie Aviles: whichever selector that I specify, and…
500 00:44:03.790 ⇒ 00:44:04.450 Samuel Roberts: Hmm.
501 00:44:04.450 ⇒ 00:44:05.010 Casie Aviles: That’s between.
502 00:44:05.010 ⇒ 00:44:09.719 Samuel Roberts: So these also have… you can also generate a report for all those and do the same thing, is that what you’re saying?
503 00:44:10.240 ⇒ 00:44:13.959 Casie Aviles: Yeah, or for Dardash, yeah, we also have this report.
504 00:44:15.150 ⇒ 00:44:20.079 Casie Aviles: But when we built the automation, we were basically just copying the manual.
505 00:44:20.810 ⇒ 00:44:21.720 Casie Aviles: process.
506 00:44:24.890 ⇒ 00:44:25.510 Samuel Roberts: Okay.
507 00:44:26.980 ⇒ 00:44:29.210 Samuel Roberts: What about that, daily sales?
508 00:44:30.120 ⇒ 00:44:30.700 Uttam Kumaran: Yeah.
509 00:44:31.380 ⇒ 00:44:35.929 Casie Aviles: Yes, for this one, yeah, that’s what I was trying to show earlier, and…
510 00:44:36.510 ⇒ 00:44:45.630 Casie Aviles: The only… the thing here is when… for, like, the manual process, what we do is we have to go into the inbox, or Robert’s inbox.
511 00:44:46.060 ⇒ 00:44:48.630 Casie Aviles: And then we have to go to this file.
512 00:44:49.980 ⇒ 00:44:53.319 Casie Aviles: And then… we have to edit this in the browser.
513 00:44:54.020 ⇒ 00:44:54.750 Samuel Roberts: Where…
514 00:44:55.240 ⇒ 00:44:56.620 Casie Aviles: Then we’ll have to get this.
515 00:44:57.600 ⇒ 00:44:58.330 Casie Aviles: Yay.
516 00:44:58.980 ⇒ 00:45:00.819 Samuel Roberts: What is generating this, though?
517 00:45:01.220 ⇒ 00:45:08.229 Casie Aviles: Yeah, that’s… that’s kind of, like… what… We wanted to, identify.
518 00:45:08.780 ⇒ 00:45:09.100 Samuel Roberts: Oh my god.
519 00:45:09.100 ⇒ 00:45:13.380 Casie Aviles: If we could just pull… From whichever source they have.
520 00:45:13.550 ⇒ 00:45:18.419 Casie Aviles: These same numbers, so instead of having to go through the…
521 00:45:18.860 ⇒ 00:45:23.879 Samuel Roberts: the workbooks manually. We just store it somewhere.
522 00:45:25.490 ⇒ 00:45:28.100 Samuel Roberts: Yeah, so some… where… what email is that coming… like, what is that?
523 00:45:29.240 ⇒ 00:45:30.630 Samuel Roberts: Coming from…
524 00:45:31.440 ⇒ 00:45:36.659 Casie Aviles: You mean where it’s getting sent to? It’s ending… yeah, it’s here in Robert’s…
525 00:45:36.920 ⇒ 00:45:37.730 Samuel Roberts: Right, but like.
526 00:45:37.730 ⇒ 00:45:38.480 Casie Aviles: Tonya could be.
527 00:45:38.570 ⇒ 00:45:41.799 Samuel Roberts: Is someone sending this, or is this an automated thing?
528 00:45:42.500 ⇒ 00:45:45.889 Casie Aviles: Oh yeah, it’s being sent by… by Talis.
529 00:45:50.350 ⇒ 00:45:55.520 Samuel Roberts: So is someone filling these out manually still? And then sending these along?
530 00:45:56.650 ⇒ 00:45:58.610 Casie Aviles: That part, I’m not… not sure about.
531 00:45:58.610 ⇒ 00:45:58.990 Samuel Roberts: Okay.
532 00:45:58.990 ⇒ 00:45:59.730 Casie Aviles: per…
533 00:45:59.730 ⇒ 00:46:06.309 Samuel Roberts: So we need to clarify what’s going on with the… the daily sales… Email in general.
534 00:46:07.050 ⇒ 00:46:08.410 Samuel Roberts: To know what’s behind that.
535 00:46:08.860 ⇒ 00:46:09.560 Samuel Roberts: Okay.
536 00:46:16.610 ⇒ 00:46:17.550 Samuel Roberts: Okay.
537 00:46:22.350 ⇒ 00:46:28.600 Samuel Roberts: Alright, so then… and then the info that we’re actually getting from that is… Which parts here?
538 00:46:30.740 ⇒ 00:46:38.769 Casie Aviles: Oh, if I go back to the scorecard, it would be this, the non… under non-core.
539 00:46:39.070 ⇒ 00:46:39.730 Samuel Roberts: Okay.
540 00:46:40.890 ⇒ 00:46:47.839 Casie Aviles: So, we’re getting 2025 revenue and 2024, so current year and last year, and then these ones.
541 00:46:48.350 ⇒ 00:46:56.280 Awaish Kumar: Yeah, the… like, we have some tables in holistics, which might… Aww.
542 00:46:57.230 ⇒ 00:46:59.750 Awaish Kumar: Help us get this, like, revenue data?
543 00:47:03.510 ⇒ 00:47:04.170 Casie Aviles: Okay.
544 00:47:13.210 ⇒ 00:47:20.430 Uttam Kumaran: Yeah, so how does Holistics fit into this? Like, I guess this is also what I’m saying, can we put some people on this? Like, who have we talked to so far?
545 00:47:20.790 ⇒ 00:47:24.150 Uttam Kumaran: Like, I’m trying to just, like, find out who to go ask these questions to.
546 00:47:25.710 ⇒ 00:47:28.200 Uttam Kumaran: Like, Casey, have you talked to anyone on their data side?
547 00:47:29.560 ⇒ 00:47:30.700 Casie Aviles: No, I’m…
548 00:47:31.060 ⇒ 00:47:40.119 Casie Aviles: most… I’m out of… I’m outside of their channel, and this Holistics one kind of showed up late when we are… when we’re already done with
549 00:47:40.270 ⇒ 00:47:44.689 Casie Aviles: All of the… automations, and I think that was…
550 00:47:45.020 ⇒ 00:47:52.999 Casie Aviles: what Robert was asking from Robert Cantor, like, from the team, I think, yeah. I think he was the data guy there.
551 00:47:54.610 ⇒ 00:48:00.259 Uttam Kumaran: So there is… so I guess, like, let’s keep going here. So this daily sales… so there’s…
552 00:48:01.210 ⇒ 00:48:07.750 Uttam Kumaran: There is an attachment that comes in on which you’re looking at… Two rows from here.
553 00:48:10.160 ⇒ 00:48:16.159 Uttam Kumaran: Yeah, like, can we… I guess I just want to, like, put it in a sticky, or put it, like, into one of these.
554 00:48:16.160 ⇒ 00:48:16.860 Samuel Roberts: Yeah.
555 00:48:17.670 ⇒ 00:48:25.549 Uttam Kumaran: Can we just write it in here? Like, I just want to make it super crystal clear, like, what the F is going on, because I’m having trouble even following.
556 00:48:27.350 ⇒ 00:48:31.799 Uttam Kumaran: So yeah, go ahead. If you say it out loud, I’ll write it down.
557 00:48:32.750 ⇒ 00:48:37.850 Casie Aviles: Yeah, so we’re getting the total sales under daily here.
558 00:48:39.920 ⇒ 00:48:42.860 Casie Aviles: I’m sharing it also on my screen also.
559 00:48:43.160 ⇒ 00:48:45.290 Samuel Roberts: So, it’s this one.
560 00:48:46.110 ⇒ 00:48:47.380 Casie Aviles: And then this one.
561 00:48:47.590 ⇒ 00:48:49.750 Casie Aviles: The… this year and last year.
562 00:48:53.070 ⇒ 00:48:53.710 Uttam Kumaran: Okay.
563 00:48:55.060 ⇒ 00:49:00.179 Casie Aviles: So we’re getting that, and… We just have to multiply it.
564 00:49:01.140 ⇒ 00:49:09.339 Casie Aviles: Based on… so, when we’re inputting it here in the daily impact scorecard, we’re multiplying it by a thousand, but…
565 00:49:09.680 ⇒ 00:49:16.310 Casie Aviles: Essentially, that’s just, I guess, something that they do, but… It’s still the same data.
566 00:49:17.300 ⇒ 00:49:21.340 Casie Aviles: And then down here, we also get…
567 00:49:21.820 ⇒ 00:49:25.259 Casie Aviles: All other store sales, this one.
568 00:49:25.830 ⇒ 00:49:31.330 Casie Aviles: Which is also for the current year and last year. These… these two.
569 00:49:31.750 ⇒ 00:49:35.260 Casie Aviles: Specifically, NWS and non-POS.
570 00:49:38.250 ⇒ 00:49:38.910 Uttam Kumaran: Okay.
571 00:49:42.030 ⇒ 00:49:44.919 Uttam Kumaran: So both of those, you’re multiplying by 1,000.
572 00:49:45.730 ⇒ 00:49:47.360 Casie Aviles: Yeah, also this one.
573 00:49:48.570 ⇒ 00:49:50.079 Uttam Kumaran: And then where does that go?
574 00:49:53.840 ⇒ 00:50:00.319 Casie Aviles: Yeah, it would go here to… the same… Yeah, labels here.
575 00:50:00.490 ⇒ 00:50:00.980 Casie Aviles: We’re…
576 00:50:00.980 ⇒ 00:50:04.009 Uttam Kumaran: So can you… can you math that on this diagram?
577 00:50:04.460 ⇒ 00:50:08.170 Uttam Kumaran: Which, like, which report it’s going… like, so this isn’t, like, see, this…
578 00:50:09.120 ⇒ 00:50:13.340 Uttam Kumaran: Okay, so… oh, it’s going to Daily Impact Scorecard, okay.
579 00:50:13.340 ⇒ 00:50:14.200 Casie Aviles: Yes, yes.
580 00:50:18.640 ⇒ 00:50:19.400 Uttam Kumaran: Alright.
581 00:50:20.030 ⇒ 00:50:21.639 Casie Aviles: For example, here.
582 00:50:21.930 ⇒ 00:50:23.750 Casie Aviles: I would paste that here.
583 00:50:24.770 ⇒ 00:50:27.880 Casie Aviles: So I would have to do something like this.
584 00:50:36.720 ⇒ 00:50:38.310 Casie Aviles: And I would paste that.
585 00:50:38.420 ⇒ 00:50:39.510 Casie Aviles: Over here.
586 00:50:40.630 ⇒ 00:50:41.450 Uttam Kumaran: I see.
587 00:50:44.140 ⇒ 00:50:45.500 Uttam Kumaran: So, okay.
588 00:50:48.880 ⇒ 00:50:49.540 Casie Aviles: Excuse me.
589 00:50:51.100 ⇒ 00:50:52.679 Casie Aviles: Yeah, that’s pretty much it.
590 00:50:52.930 ⇒ 00:50:55.650 Uttam Kumaran: And then what’s in holistics that we need to bring in?
591 00:50:56.640 ⇒ 00:50:58.079 Uttam Kumaran: So, like, an open… an open…
592 00:50:58.080 ⇒ 00:50:59.640 Samuel Roberts: What is Holistics? I haven’t heard of that.
593 00:50:59.640 ⇒ 00:51:01.660 Uttam Kumaran: It’s just another BI tool.
594 00:51:01.660 ⇒ 00:51:03.890 Samuel Roberts: It’s another VI tool, okay, cool.
595 00:51:03.890 ⇒ 00:51:12.430 Uttam Kumaran: So one question I have, like, for here, for them, is, like, where are these sales figures coming from?
596 00:51:12.760 ⇒ 00:51:13.510 Uttam Kumaran: Right?
597 00:51:13.510 ⇒ 00:51:14.610 Samuel Roberts: Like, there’s another… yeah.
598 00:51:14.610 ⇒ 00:51:19.070 Uttam Kumaran: So basically, this is, like… so I want to kind of create, like, a couple things. So one is this is, like, a…
599 00:51:19.420 ⇒ 00:51:21.340 Uttam Kumaran: V2 architecture.
600 00:51:22.430 ⇒ 00:51:26.699 Uttam Kumaran: And then, let’s say we were to go ahead and create, like, a…
601 00:51:26.930 ⇒ 00:51:30.569 Uttam Kumaran: a V3 architecture to the right.
602 00:51:30.920 ⇒ 00:51:35.420 Uttam Kumaran: Basically, what we want instead of this
603 00:51:35.590 ⇒ 00:51:41.320 Uttam Kumaran: Is a… literally, like, whatever the source of truth for sales data is, right?
604 00:51:42.100 ⇒ 00:51:46.730 Uttam Kumaran: Like, sales data source… If I can get rid of this…
605 00:51:47.100 ⇒ 00:51:47.740 Samuel Roberts: Yeah.
606 00:51:48.680 ⇒ 00:51:52.989 Uttam Kumaran: I just want to move this out, right? So this is sales data source.
607 00:51:54.030 ⇒ 00:52:00.089 Uttam Kumaran: I don’t want holist… so then we also have a couple things here. We have, like, holistics…
608 00:52:00.420 ⇒ 00:52:05.590 Uttam Kumaran: We have… These are… so we have holistics, we have…
609 00:52:05.750 ⇒ 00:52:13.040 Uttam Kumaran: Look at our studio, and we have Excel Cloud, right, or whatever, Office Cloud.
610 00:52:13.360 ⇒ 00:52:20.979 Awaish Kumar: We have great… Huristic is also connected to some data warehouse, which… we don’t know which one it is.
611 00:52:22.500 ⇒ 00:52:24.719 Uttam Kumaran: Yeah, so that’s what I’ll find out.
612 00:52:28.590 ⇒ 00:52:40.899 Uttam Kumaran: So this one is here, this is here. So, for Google Ads, we want both of these to go through Polytomic, ideally. So, ideally, this is like this.
613 00:52:41.430 ⇒ 00:52:43.050 Uttam Kumaran: This is like this…
614 00:52:43.460 ⇒ 00:52:43.910 Samuel Roberts: Yeah.
615 00:52:43.910 ⇒ 00:52:47.560 Uttam Kumaran: This is like this… Right.
616 00:52:49.410 ⇒ 00:53:00.770 Uttam Kumaran: And then… This is also, like, here… This is also through here.
617 00:53:02.850 ⇒ 00:53:05.179 Uttam Kumaran: Right? This is, like, an ideal world.
618 00:53:11.290 ⇒ 00:53:14.410 Uttam Kumaran: Yeah, like, basically, this is probably, like, what the…
619 00:53:14.930 ⇒ 00:53:18.569 Uttam Kumaran: new world could look like, right? What else am I missing here?
620 00:53:21.690 ⇒ 00:53:22.460 Uttam Kumaran: Oops.
621 00:53:31.190 ⇒ 00:53:34.089 Uttam Kumaran: And then instead, this goes from here…
622 00:53:34.680 ⇒ 00:53:39.550 Uttam Kumaran: There’s, like, some… basically, this is, like… Logic…
623 00:53:56.960 ⇒ 00:54:02.360 Uttam Kumaran: This is where we have something like…
624 00:54:03.750 ⇒ 00:54:06.189 Uttam Kumaran: Like, we would have our dbt layer, right?
625 00:54:16.120 ⇒ 00:54:20.160 Uttam Kumaran: So this is dbt layer… And then we have…
626 00:54:37.540 ⇒ 00:54:38.349 Samuel Roberts: Excuse me.
627 00:54:38.780 ⇒ 00:54:45.589 Samuel Roberts: If we’re talking, like, really ideal, there’s also, like, there is a DoorDash reporting API that is not generally available.
628 00:54:46.140 ⇒ 00:54:47.000 Uttam Kumaran: Oh, really?
629 00:54:47.300 ⇒ 00:54:53.409 Samuel Roberts: I… they could, I don’t know. I don’t know who gets access to that, but you can record.
630 00:54:53.570 ⇒ 00:54:56.399 Samuel Roberts: interest, so I think they’d have to go ahead and do that.
631 00:54:57.010 ⇒ 00:54:58.220 Uttam Kumaran: Oh, really? Okay.
632 00:54:58.610 ⇒ 00:55:01.029 Samuel Roberts: I think there was something else for…
633 00:55:01.220 ⇒ 00:55:01.760 Casie Aviles: Dude.
634 00:55:01.760 ⇒ 00:55:02.720 Samuel Roberts: Yeah…
635 00:55:02.980 ⇒ 00:55:06.799 Casie Aviles: Like, they have to… they have to do something like this to get, like.
636 00:55:06.800 ⇒ 00:55:14.950 Samuel Roberts: Yeah, there was a whole agreement for, yeah, Uber Eats as well. So, like, if we’re talking, like, fully ideal, then this would also be something that would have to happen.
637 00:55:16.240 ⇒ 00:55:18.799 Uttam Kumaran: Yeah, cool, let’s do that. Let’s put that in here.
638 00:55:18.990 ⇒ 00:55:22.009 Samuel Roberts: Yeah. So, can you, can you, can you replace these, Sam, with…
639 00:55:22.010 ⇒ 00:55:24.390 Uttam Kumaran: each of the APIs, and have each of those.
640 00:55:24.390 ⇒ 00:55:24.990 Samuel Roberts: flowed.
641 00:55:24.990 ⇒ 00:55:28.850 Uttam Kumaran: to have each of those flow into Polytomic.
642 00:55:30.730 ⇒ 00:55:34.010 Samuel Roberts: Yeah, so this is… am I on V3? Yeah, very cool.
643 00:55:39.120 ⇒ 00:55:43.789 Samuel Roberts: So, Casey, there’s definitely DoorDash, there was definitely Uber Eats, what about,
644 00:55:44.190 ⇒ 00:55:46.879 Samuel Roberts: Uber Ads. Do you know what that one?
645 00:55:47.530 ⇒ 00:55:49.180 Samuel Roberts: They all have something that was…
646 00:55:50.790 ⇒ 00:55:53.500 Samuel Roberts: Yeah, I don’t know if that API… I don’t know what that API was.
647 00:55:55.260 ⇒ 00:55:56.330 Samuel Roberts: Yeah, the umpire.
648 00:56:00.260 ⇒ 00:56:03.570 Samuel Roberts: I think you can… there’s some reporting metrics here.
649 00:56:10.010 ⇒ 00:56:10.559 Casie Aviles: Look at…
650 00:56:10.570 ⇒ 00:56:13.790 Samuel Roberts: I see reporting metrics for ads.
651 00:56:15.680 ⇒ 00:56:16.600 Samuel Roberts: Dude.
652 00:56:17.820 ⇒ 00:56:19.029 Casie Aviles: You know, this one.
653 00:56:19.720 ⇒ 00:56:20.430 Casie Aviles: But…
654 00:56:21.100 ⇒ 00:56:21.690 Samuel Roberts: Yeah.
655 00:56:24.330 ⇒ 00:56:27.320 Samuel Roberts: Is this actually accessible now, or is this another thing they’d have to…
656 00:56:28.610 ⇒ 00:56:34.010 Casie Aviles: Yeah, I believe it’s the same case with… Uber Eats.
657 00:56:34.550 ⇒ 00:56:35.290 Samuel Roberts: Okay.
658 00:56:40.230 ⇒ 00:56:43.509 Samuel Roberts: Yeah, under development, something could change. Yeah, that’s right.
659 00:56:56.610 ⇒ 00:57:00.590 Samuel Roberts: So we shouldn’t need… or V…
660 00:57:10.050 ⇒ 00:57:13.160 Samuel Roberts: And get rid of both of these, then, I guess?
661 00:57:14.390 ⇒ 00:57:15.180 Casie Aviles: Yes.
662 00:57:15.590 ⇒ 00:57:20.450 Samuel Roberts: And that can just flow… Through Polyatomic, that we would do that.
663 00:57:33.740 ⇒ 00:57:36.049 Samuel Roberts: Okay, only belongings.
664 00:58:19.920 ⇒ 00:58:20.600 Samuel Roberts: I bet.
665 00:59:14.600 ⇒ 00:59:18.909 Mustafa Raja: Can Polytomic also sync with the APIs themselves?
666 00:59:19.700 ⇒ 00:59:21.450 Uttam Kumaran: Yeah, we would ask them to build it.
667 00:59:22.790 ⇒ 00:59:23.460 Mustafa Raja: Okay.
668 00:59:23.460 ⇒ 00:59:24.110 Uttam Kumaran: Yeah.
669 00:59:47.780 ⇒ 00:59:52.139 Uttam Kumaran: Okay, so, like, roughly, this is, like, what it would look like.
670 00:59:53.720 ⇒ 00:59:55.749 Uttam Kumaran: In terms of a true V3?
671 01:00:04.700 ⇒ 01:00:08.229 Samuel Roberts: So we need both Braze things, the API exports and the currents?
672 01:00:10.620 ⇒ 01:00:11.920 Samuel Roberts: Okay, then yeah.
673 01:00:13.510 ⇒ 01:00:15.479 Samuel Roberts: City Mother Gog logic, yeah.
674 01:00:18.000 ⇒ 01:00:21.200 Demilade Agboola: Is there a different liquid export that we can get in the current?
675 01:00:26.410 ⇒ 01:00:27.750 Uttam Kumaran: We have a question for Sam.
676 01:00:28.380 ⇒ 01:00:30.620 Samuel Roberts: What the… I…
677 01:00:31.060 ⇒ 01:00:36.680 Samuel Roberts: I went into the currents and just exported everything. I’m not sure what’s different about the API export, so I thought.
678 01:00:36.680 ⇒ 01:00:42.720 Uttam Kumaran: The API export… the API export is just a raw dump, like, so we can backfill.
679 01:00:43.320 ⇒ 01:00:44.240 Samuel Roberts: Oh, okay.
680 01:00:44.920 ⇒ 01:00:48.259 Samuel Roberts: Okay, yeah, so the currents would just be, like, ongoing, yeah, that’s…
681 01:00:48.590 ⇒ 01:00:53.679 Uttam Kumaran: But this one also… Yeah, so this one would be…
682 01:00:54.170 ⇒ 01:00:57.909 Uttam Kumaran: Well, technically, this is also through S3.
683 01:00:59.180 ⇒ 01:01:02.080 Uttam Kumaran: Yeah, so this is… yeah, this is, like, what we would do.
684 01:01:03.110 ⇒ 01:01:03.730 Samuel Roberts: Yes.
685 01:01:20.350 ⇒ 01:01:22.100 Uttam Kumaran: Okay, how do we feel about this?
686 01:01:27.690 ⇒ 01:01:29.719 Casie Aviles: Yeah, this looks a lot cleaner.
687 01:01:38.660 ⇒ 01:01:43.790 Samuel Roberts: Yeah… Yeah, I mean, ideal… this looks very ideal to me, yeah, so…
688 01:01:44.760 ⇒ 01:01:47.629 Uttam Kumaran: So this is our, like, typical setup, I would say.
689 01:01:48.130 ⇒ 01:01:49.290 Samuel Roberts: That would make sense.
690 01:02:10.100 ⇒ 01:02:11.690 Uttam Kumaran: Yeah, let’s just connect these.
691 01:02:12.270 ⇒ 01:02:17.669 Uttam Kumaran: Mustafa, can you let me clean up these lines? We can just, like, make them all, like, come up to here.
692 01:02:29.730 ⇒ 01:02:31.650 Uttam Kumaran: And let’s shrink this a little bit.
693 01:02:32.560 ⇒ 01:02:33.110 Samuel Roberts: Yeah.
694 01:02:33.990 ⇒ 01:02:35.550 Mustafa Raja: Oh, I’m guessing I knew approved.
695 01:02:35.980 ⇒ 01:02:36.970 Mustafa Raja: integrity.
696 01:02:36.970 ⇒ 01:02:37.444 Casie Aviles: Oh.
697 01:02:39.860 ⇒ 01:02:40.560 Mustafa Raja: Thank you.
698 01:02:59.740 ⇒ 01:03:00.360 Samuel Roberts: Excuse me.
699 01:03:04.250 ⇒ 01:03:04.830 Uttam Kumaran: Okay.
700 01:03:07.980 ⇒ 01:03:08.910 Uttam Kumaran: Okay.
701 01:03:10.310 ⇒ 01:03:13.360 Uttam Kumaran: And then the one thing we’re missing is…
702 01:03:15.880 ⇒ 01:03:18.369 Uttam Kumaran: Oh, okay, that’s what I find for now.
703 01:03:21.020 ⇒ 01:03:25.459 Uttam Kumaran: Okay, so if we were to list out, like, next steps here.
704 01:03:26.020 ⇒ 01:03:28.790 Uttam Kumaran: I need to call them and figure out…
705 01:03:30.630 ⇒ 01:03:34.330 Uttam Kumaran: So, like, ideally, we want to move to V2 first, right?
706 01:03:34.480 ⇒ 01:03:37.139 Uttam Kumaran: So we need to get approval on the warehouse.
707 01:03:38.930 ⇒ 01:03:45.559 Uttam Kumaran: And then, basically, we need to switch anything that’s going to Snowflake to the warehouse.
708 01:03:47.240 ⇒ 01:03:51.179 Uttam Kumaran: We also need to figure out how to get from Mother Duck into these things.
709 01:03:51.830 ⇒ 01:03:53.749 Uttam Kumaran: So that’s still not really clear to me.
710 01:03:54.150 ⇒ 01:03:54.890 Samuel Roberts: Yeah.
711 01:03:58.400 ⇒ 01:03:59.120 Uttam Kumaran: Right.
712 01:04:00.730 ⇒ 01:04:04.060 Uttam Kumaran: Like, how do you get from Mother Duck into these three areas?
713 01:04:15.640 ⇒ 01:04:19.510 Samuel Roberts: Looks like there might be a way to go… to Google…
714 01:04:19.510 ⇒ 01:04:23.990 Demilade Agboola: Are they Google Sheets, or are they, are they, Excel files?
715 01:04:24.300 ⇒ 01:04:26.189 Uttam Kumaran: They’re, like, yeah, I guess, Casey, what are.
716 01:04:26.190 ⇒ 01:04:29.560 Casie Aviles: Their Excel files on SharePoint.
717 01:04:29.750 ⇒ 01:04:30.460 Casie Aviles: So it’s…
718 01:04:37.610 ⇒ 01:04:40.479 Awaish Kumar: Yeah, that’s going to be a man and fill anyway.
719 01:04:41.010 ⇒ 01:04:41.770 Samuel Roberts: Yeah.
720 01:04:54.150 ⇒ 01:04:55.120 Samuel Roberts: Yeah.
721 01:05:00.580 ⇒ 01:05:03.399 Awaish Kumar: That’s why we have, like, intermediate Google Sheets.
722 01:05:03.590 ⇒ 01:05:08.269 Awaish Kumar: from where you can just copy-paste into SharePoint Excel files.
723 01:05:08.680 ⇒ 01:05:13.029 Samuel Roberts: Right, so we probably still have some intermediate Google Sheet from Mother Duck then into that, right?
724 01:05:13.750 ⇒ 01:05:14.620 Awaish Kumar: Yes.
725 01:05:15.240 ⇒ 01:05:15.900 Samuel Roberts: Okay.
726 01:05:18.840 ⇒ 01:05:22.419 Awaish Kumar: And, for the… Like, the first post?
727 01:05:22.420 ⇒ 01:05:22.930 Casie Aviles: Excellent.
728 01:05:22.930 ⇒ 01:05:23.870 Awaish Kumar: Don’t, don’t worry.
729 01:05:23.870 ⇒ 01:05:24.939 Samuel Roberts: staging spreadsheet.
730 01:05:30.170 ⇒ 01:05:33.530 Samuel Roberts: So we’d still have that in V2, right? The… yeah.
731 01:05:37.130 ⇒ 01:05:37.860 Samuel Roberts: Perfect.
732 01:07:06.370 ⇒ 01:07:07.689 Samuel Roberts: Then,
733 01:07:07.880 ⇒ 01:07:14.459 Samuel Roberts: Casey, for V2, would we want to… you were talking about maybe just doing the reports for all of the browser-based stuff?
734 01:07:15.530 ⇒ 01:07:19.720 Casie Aviles: Yeah, I’m thinking of… Do we find that? Is that…
735 01:07:19.720 ⇒ 01:07:21.350 Samuel Roberts: Less brittle, you think?
736 01:07:23.040 ⇒ 01:07:24.309 Samuel Roberts: A less fragile.
737 01:07:26.060 ⇒ 01:07:32.580 Casie Aviles: If, if I can fig… yeah, it should be less if… since we’re just going to generate the reports.
738 01:07:32.580 ⇒ 01:07:33.130 Samuel Roberts: Yeah.
739 01:07:33.130 ⇒ 01:07:37.210 Casie Aviles: going to… You know, look, go through the selectors.
740 01:07:37.760 ⇒ 01:07:39.100 Samuel Roberts: Sure.
741 01:07:39.720 ⇒ 01:07:46.189 Casie Aviles: So I think… what I just need to figure out is how I could reliably authenticate.
742 01:07:47.420 ⇒ 01:07:51.410 Samuel Roberts: Yeah. I’m sure there’s a way to figure that out with browser-based, that’s gotta be a thing they’ve…
743 01:07:52.510 ⇒ 01:07:53.530 Samuel Roberts: If so, yeah.
744 01:07:54.980 ⇒ 01:07:55.610 Casie Aviles: Yes.
745 01:08:29.779 ⇒ 01:08:31.229 Samuel Roberts: Wait, are those not the same?
746 01:08:31.529 ⇒ 01:08:32.879 Samuel Roberts: Process, then, now?
747 01:08:33.189 ⇒ 01:08:38.309 Samuel Roberts: the Uber Eats promos, UberAds, and the DoorDash, would that all be one… like, one, like…
748 01:08:38.509 ⇒ 01:08:42.499 Samuel Roberts: There were two different things, one was scraping, one was generating reports. Would they be the same now?
749 01:08:43.279 ⇒ 01:08:44.539 Casie Aviles: Oh, yeah, yeah, they should be.
750 01:08:44.540 ⇒ 01:08:47.359 Samuel Roberts: With the wait 15 minutes thing, still.
751 01:08:48.970 ⇒ 01:08:50.029 Samuel Roberts: I’m assuming.
752 01:08:50.810 ⇒ 01:08:56.749 Casie Aviles: Hmm… Yeah, I think that we should, like, have, some… Okay.
753 01:08:57.340 ⇒ 01:09:00.199 Casie Aviles: interval. Maybe this could be less novel.
754 01:09:00.740 ⇒ 01:09:04.010 Samuel Roberts: Yeah, we can figure that out, like, based on what it is, but,
755 01:09:06.240 ⇒ 01:09:13.830 Samuel Roberts: Okay, well, that one is now… Ask Weekend… Delete that.
756 01:09:27.340 ⇒ 01:09:29.069 Samuel Roberts: Then we use that. Oops.
757 01:09:37.700 ⇒ 01:09:39.659 Samuel Roberts: So now this line should go into…
758 01:09:42.990 ⇒ 01:09:43.760 Samuel Roberts: Yeah.
759 01:09:46.700 ⇒ 01:09:47.870 Samuel Roberts: And then this…
760 01:09:50.939 ⇒ 01:09:53.129 Samuel Roberts: And then this is still the…
761 01:10:08.420 ⇒ 01:10:14.499 Uttam Kumaran: Okay, so can we do just, like, a broad review? I’ll just… I’m just gonna say, kind of, the summary from my side, so…
762 01:10:15.070 ⇒ 01:10:19.009 Uttam Kumaran: I need to see if we can get the source of some of these items.
763 01:10:21.950 ⇒ 01:10:28.810 Uttam Kumaran: So I guess, can we… can we do… can we do a couple things? Can we build, like, a little bit of, like, a legend, and so I can start to mark…
764 01:10:29.040 ⇒ 01:10:34.609 Uttam Kumaran: things of, like, open items, or, like, we need the source, or things like that. Like, can you guys help me do that?
765 01:10:35.530 ⇒ 01:10:41.459 Uttam Kumaran: I’m happy to go get, like, whatever we need, but I can’t handle… it’s just too much for me to organize.
766 01:10:41.460 ⇒ 01:10:43.270 Samuel Roberts: Yeah.
767 01:10:43.840 ⇒ 01:10:51.379 Uttam Kumaran: So if you can mark clearly, like, which things I need to get us access to, what decisions you guys need me to go get them to make.
768 01:10:51.840 ⇒ 01:10:55.889 Uttam Kumaran: On the infra side, I can go do that.
769 01:10:57.400 ⇒ 01:11:02.590 Samuel Roberts: So… Let me just put some question mark stamps then, is that…
770 01:11:03.240 ⇒ 01:11:06.640 Samuel Roberts: Like… Or do you want me to, like, just make a list?
771 01:11:07.900 ⇒ 01:11:09.610 Uttam Kumaran: I just need anything, dude, but I…
772 01:11:09.610 ⇒ 01:11:10.470 Samuel Roberts: Okay, I can do that.
773 01:11:10.470 ⇒ 01:11:13.429 Uttam Kumaran: Think about where I’m coming from, like, I have to go.
774 01:11:13.430 ⇒ 01:11:14.150 Samuel Roberts: Yeah, yeah.
775 01:11:14.150 ⇒ 01:11:16.469 Uttam Kumaran: 10 or 15 things here, I just need a really…
776 01:11:16.800 ⇒ 01:11:17.310 Samuel Roberts: Okay.
777 01:11:17.310 ⇒ 01:11:24.960 Uttam Kumaran: I want to use this diagram as, like, a guide… guiding force, but I need to just know clearly, like, here are all the things that we need to do.
778 01:11:25.140 ⇒ 01:11:25.650 Uttam Kumaran: T.
779 01:11:25.650 ⇒ 01:11:27.060 Samuel Roberts: Needed for V2.
780 01:11:28.020 ⇒ 01:11:32.589 Samuel Roberts: I’m gonna type it up in a little… Okay, so we need, mother duct approval, I guess?
781 01:11:35.700 ⇒ 01:11:36.560 Samuel Roberts: Right.
782 01:11:36.730 ⇒ 01:11:39.300 Samuel Roberts: We need to know…
783 01:11:42.770 ⇒ 01:11:45.100 Samuel Roberts: Where that dailies… Sorry?
784 01:11:45.100 ⇒ 01:11:45.820 Awaish Kumar: How you doing?
785 01:11:46.150 ⇒ 01:11:47.610 Awaish Kumar: Polydonica.
786 01:11:48.270 ⇒ 01:11:48.990 Samuel Roberts: Thank you.
787 01:11:51.690 ⇒ 01:11:58.670 Samuel Roberts: We wanna know… The piece of the daily sales and holistics question marks?
788 01:11:58.890 ⇒ 01:12:01.020 Awaish Kumar: sales data source.
789 01:12:01.870 ⇒ 01:12:02.940 Samuel Roberts: Yes, thank you.
790 01:12:04.890 ⇒ 01:12:07.110 Samuel Roberts: Sales data source,
791 01:12:08.960 ⇒ 01:12:17.109 Samuel Roberts: I think the other question, I mean, that becomes, you know, maybe they need to apply for these APIs, but we can wait on that even if that’s not V2.
792 01:12:17.210 ⇒ 01:12:24.690 Samuel Roberts: But I don’t know how long that, you know, I don’t know what Uber’s process is like for that, so I don’t know if that’s something that’s worth pushing right now or not to them.
793 01:12:26.270 ⇒ 01:12:26.990 Demilade Agboola: That’ll be walking.
794 01:12:27.010 ⇒ 01:12:27.500 Samuel Roberts: Pushing bike.
795 01:12:27.500 ⇒ 01:12:31.120 Demilade Agboola: I think we should just at least mention it to them, so we can.
796 01:12:31.120 ⇒ 01:12:31.750 Samuel Roberts: Yeah.
797 01:12:31.750 ⇒ 01:12:33.410 Demilade Agboola: Start to get that ball rolling.
798 01:12:33.950 ⇒ 01:12:40.240 Samuel Roberts: DoorDash… slash Uber API application.
799 01:12:44.720 ⇒ 01:12:49.200 Samuel Roberts: raises their Looker… We wanna go…
800 01:12:53.200 ⇒ 01:12:56.219 Awaish Kumar: Are we still looking to go from Looker, or are we going from…
801 01:12:57.760 ⇒ 01:13:01.019 Samuel Roberts: We want to try to get that underlying data source, too, for Facebook and Google.
802 01:13:04.020 ⇒ 01:13:06.539 Demilade Agboola: Oh, we’ll just connect directly to Facebook ads and…
803 01:13:06.810 ⇒ 01:13:08.820 Demilade Agboola: Yeah, okay. So then we need…
804 01:13:09.070 ⇒ 01:13:11.329 Samuel Roberts: So we don’t want this looker report. This is…
805 01:13:12.960 ⇒ 01:13:13.750 Samuel Roberts: Yeah,
806 01:13:18.330 ⇒ 01:13:21.910 Awaish Kumar: For Google Ads, we already know, right? We are bringing it through Polyatomic.
807 01:13:22.130 ⇒ 01:13:23.949 Awaish Kumar: We just need Phase 1.
808 01:13:26.010 ⇒ 01:13:32.480 Samuel Roberts: Oh, so we’re just bypassing whatever they were even doing there, and just going straight… okay, so go ahead. Google Ads we already have, so we just need meta ads.
809 01:13:33.220 ⇒ 01:13:34.459 Samuel Roberts: Do you prefer polyatomic.
810 01:13:39.310 ⇒ 01:13:41.110 Samuel Roberts: Is that correct? Oops.
811 01:13:48.080 ⇒ 01:13:51.140 Samuel Roberts: What are the other questions here?
812 01:13:51.410 ⇒ 01:13:52.769 Samuel Roberts: Esther, the mother duck, is…
813 01:13:58.800 ⇒ 01:14:05.999 Uttam Kumaran: I have to hop, like, really soon, so I… you guys… you guys think you can send it to me over Slack?
814 01:14:08.140 ⇒ 01:14:12.599 Samuel Roberts: Yeah, I mean, I think this is everything I’ve thought of, unless I’m missing anything else here, anyone.
815 01:14:12.750 ⇒ 01:14:15.370 Uttam Kumaran: So, can you… can you go through it? I just wanna see what it is.
816 01:14:15.370 ⇒ 01:14:21.029 Samuel Roberts: Yeah, we need, approval from OtherDuck, approval for Polytomic to get it onto their…
817 01:14:21.460 ⇒ 01:14:27.939 Samuel Roberts: They’re billing their infra. The sales data source for that email that we’re pulling from.
818 01:14:28.160 ⇒ 01:14:33.340 Samuel Roberts: Potentially get them to do the DoorDash Uber APIs applications.
819 01:14:33.530 ⇒ 01:14:38.080 Uttam Kumaran: Okay. And then meta adds access for Polytomic, and then we can bypass that looker report, too.
820 01:14:38.520 ⇒ 01:14:47.650 Uttam Kumaran: And then we should also ask about, meta ads, Google Ads.
821 01:14:48.120 ⇒ 01:14:49.359 Awaish Kumar: We have ordinance.
822 01:14:52.050 ⇒ 01:14:54.510 Mustafa Raja: About the holistic data source.
823 01:14:54.510 ⇒ 01:14:55.670 Samuel Roberts: Oh, holistics?
824 01:14:56.980 ⇒ 01:14:57.639 Awaish Kumar: We have…
825 01:14:57.640 ⇒ 01:14:58.480 Mustafa Raja: Understanding.
826 01:14:58.480 ⇒ 01:15:04.889 Awaish Kumar: We can, like, work on identifying the data warehouse, which is being used
827 01:15:05.310 ⇒ 01:15:07.780 Awaish Kumar: As a input in holistics.
828 01:15:10.210 ⇒ 01:15:12.900 Awaish Kumar: Like, for holistics, there’s a source.
829 01:15:13.240 ⇒ 01:15:15.439 Awaish Kumar: And there’s a warehouse which they are using.
830 01:15:16.640 ⇒ 01:15:17.770 Awaish Kumar: Building holistics, right?
831 01:15:17.850 ⇒ 01:15:18.720 Samuel Roberts: Wow.
832 01:15:18.720 ⇒ 01:15:19.830 Awaish Kumar: Verdos, we don’t know.
833 01:15:21.170 ⇒ 01:15:23.009 Samuel Roberts: What is warehouse for instance? Alright.
834 01:15:25.450 ⇒ 01:15:26.909 Awaish Kumar: Is that everything, then?
835 01:15:27.550 ⇒ 01:15:33.320 Awaish Kumar: Maybe a list of stakeholders, like, maybe we can…
836 01:15:33.450 ⇒ 01:15:36.549 Awaish Kumar: Like, not ask them or identify ourselves, whatever.
837 01:15:36.760 ⇒ 01:15:37.360 Awaish Kumar: Noodle?
838 01:15:37.360 ⇒ 01:15:37.790 Uttam Kumaran: Yeah.
839 01:15:37.790 ⇒ 01:15:43.500 Awaish Kumar: And who’s working on what parts, like, who’s responsible for sales data, for approvals, for…
840 01:15:43.730 ⇒ 01:15:48.070 Awaish Kumar: Dad, DoorDash, Mirage. List of stakeholders.
841 01:15:48.360 ⇒ 01:15:57.459 Uttam Kumaran: Okay, I have to run, but yeah, just send me this list as soon as you’re done in Slack. Okay. But this is great. Yeah, this is exactly what I need to do the next piece.
842 01:15:58.780 ⇒ 01:15:59.540 Uttam Kumaran: Okay.
843 01:16:01.100 ⇒ 01:16:02.350 Samuel Roberts: We’ll finish that up and get that out.
844 01:16:07.520 ⇒ 01:16:09.300 Samuel Roberts: Okay, so.
845 01:16:10.200 ⇒ 01:16:12.640 Awaish Kumar: For each part of our workflow.
846 01:16:39.690 ⇒ 01:16:41.870 Samuel Roberts: These people, I had no idea. Okay, cool.
847 01:16:46.420 ⇒ 01:16:51.390 Casie Aviles: I think… Just probably data in general, but… Bye.
848 01:16:55.470 ⇒ 01:16:57.500 Awaish Kumar: Yeah, we can write down what, like.
849 01:16:57.910 ⇒ 01:17:06.649 Awaish Kumar: who is the CEO, whatever they are, like, the Excel sheets we fill, who is the person who’s looking at that.
850 01:17:09.560 ⇒ 01:17:14.369 Awaish Kumar: Like, who’s the one, like… Sending, like.
851 01:17:14.690 ⇒ 01:17:19.750 Awaish Kumar: In the past, we got some… a lot of requests on that the data is incorrect, or whatever.
852 01:17:19.940 ⇒ 01:17:22.560 Awaish Kumar: So, like, who are those guys, and…
853 01:17:23.290 ⇒ 01:17:25.730 Awaish Kumar: Which were, like, looking at this report.
854 01:17:26.330 ⇒ 01:17:29.610 Awaish Kumar: And escalating the numbers, things like that.
855 01:18:05.980 ⇒ 01:18:09.469 Casie Aviles: I think this is mostly them, I just forgot the other guy’s name.
856 01:18:22.960 ⇒ 01:18:25.960 Casie Aviles: I think this is pretty much done.
857 01:18:26.410 ⇒ 01:18:27.380 Samuel Roberts: Okay. We could…
858 01:18:27.380 ⇒ 01:18:30.939 Casie Aviles: Double-check with Robert if this is out here, but yeah.
859 01:18:32.270 ⇒ 01:18:37.169 Samuel Roberts: Yeah, I mean, if we’re gonna, keith, if you could just post that to the Insomnia channel?
860 01:18:37.680 ⇒ 01:18:45.149 Samuel Roberts: And then maybe just, like, tag… well, definitely tag U-Tam, tag Robert, make sure that those are the right stakeholders, perhaps, if he knows.
861 01:18:46.210 ⇒ 01:18:47.120 Casie Aviles: Okay, sure.
862 01:18:47.960 ⇒ 01:18:50.730 Samuel Roberts: Just to make sure that we’re… Aligned there.
863 01:18:51.980 ⇒ 01:18:56.150 Samuel Roberts: Cool. Anything else we need to finish up here, then?
864 01:18:59.320 ⇒ 01:19:05.660 Samuel Roberts: This felt pretty productive. I definitely have a much better sense of what’s going on here, and how we can get to V3 eventually.
865 01:19:06.800 ⇒ 01:19:09.060 Samuel Roberts: That’s good. Any other thoughts?
866 01:19:14.250 ⇒ 01:19:16.450 Casie Aviles: I think, yeah, this is very helpful.
867 01:19:16.930 ⇒ 01:19:17.530 Samuel Roberts: Yeah.
868 01:19:20.530 ⇒ 01:19:22.099 Samuel Roberts: Great, okay, so,
869 01:19:26.520 ⇒ 01:19:28.730 Samuel Roberts: I think that’s it, then. Alright, cool.
870 01:19:30.870 ⇒ 01:19:34.050 Samuel Roberts: Alright, I think we can hop then.
871 01:19:35.310 ⇒ 01:19:36.469 Samuel Roberts: Thank you all.
872 01:19:36.880 ⇒ 01:19:43.449 Samuel Roberts: This is good, I feel. I… yeah, I really appreciate it. This is… this is wonderful. I have now a much better sense of all this, so I appreciate the time.
873 01:19:43.450 ⇒ 01:19:45.190 Mustafa Raja: Spain. Alright.
874 01:19:45.810 ⇒ 01:19:46.520 Samuel Roberts: By all.
875 01:19:46.760 ⇒ 01:19:47.570 Casie Aviles: Thank you.