Meeting Title: Demilade <> Hannah - U.S. Case Study Date: 2025-09-16 Meeting participants: Hannah Wang, Demilade Agboola
WEBVTT
1 00:00:18.320 ⇒ 00:00:19.309 Demilade Agboola: Hi, Hannah.
2 00:00:19.870 ⇒ 00:00:21.359 Hannah Wang: Hey, how’s it going?
3 00:00:21.940 ⇒ 00:00:23.160 Demilade Agboola: Pretty good, how are you?
4 00:00:23.960 ⇒ 00:00:24.730 Hannah Wang: Ugh.
5 00:00:24.880 ⇒ 00:00:26.690 Hannah Wang: Hanging in there.
6 00:00:26.870 ⇒ 00:00:28.810 Demilade Agboola: Yeah, it’s too cheap.
7 00:00:29.380 ⇒ 00:00:35.690 Hannah Wang: Yeah, it’s only Tuesday, I wish… I wish it was always Friday, but that’s okay.
8 00:00:36.530 ⇒ 00:00:37.980 Demilade Agboola: Yeah.
9 00:00:37.980 ⇒ 00:00:40.929 Hannah Wang: Are you moving to the States anytime soon?
10 00:00:41.730 ⇒ 00:00:44.960 Demilade Agboola: Not… not soon, in, like, the next…
11 00:00:45.060 ⇒ 00:00:47.790 Demilade Agboola: Couple of… maybe, like, 6 months’ time, hopefully.
12 00:00:47.790 ⇒ 00:00:48.910 Hannah Wang: Oh, okay.
13 00:00:48.930 ⇒ 00:00:49.840 Demilade Agboola: Oh, we’ll see.
14 00:00:50.180 ⇒ 00:00:50.630 Hannah Wang: Cool.
15 00:00:50.630 ⇒ 00:00:52.389 Demilade Agboola: things along first, yeah.
16 00:00:52.670 ⇒ 00:00:54.510 Hannah Wang: I see. Got it.
17 00:00:56.270 ⇒ 00:00:58.369 Demilade Agboola: I should be visiting in, like, November.
18 00:00:59.230 ⇒ 00:01:02.110 Hannah Wang: For… because your girlfriend lives here, right?
19 00:01:06.250 ⇒ 00:01:07.260 Hannah Wang: I don’t know.
20 00:01:07.700 ⇒ 00:01:15.140 Demilade Agboola: Yeah, because she lives there. Yeah, so I plan to go to the Dominican Republic with her.
21 00:01:15.260 ⇒ 00:01:16.050 Demilade Agboola: Oh.
22 00:01:16.600 ⇒ 00:01:18.429 Demilade Agboola: But yeah, I have a friend who’s getting married.
23 00:01:18.720 ⇒ 00:01:24.759 Hannah Wang: Oh, cool. I feel like weddings are always a good excuse to… to travel.
24 00:01:25.600 ⇒ 00:01:28.960 Demilade Agboola: I mean, if you’re doing a destination wedding, yes.
25 00:01:28.960 ⇒ 00:01:30.070 Hannah Wang: Oh, yeah.
26 00:01:31.050 ⇒ 00:01:34.180 Hannah Wang: Because I literally had a wedding over the weekend.
27 00:01:34.340 ⇒ 00:01:37.870 Demilade Agboola: My… one of my closest friends got married, I was the best man.
28 00:01:38.390 ⇒ 00:01:49.239 Demilade Agboola: Okay. Yeah, and it was chaotic, but it was fun. I am… so I’m in Nigeria right now, but I leave on Saturday. I’ll be back tomorrow on Saturday.
29 00:01:49.580 ⇒ 00:01:50.070 Hannah Wang: Wow.
30 00:01:50.230 ⇒ 00:01:50.910 Demilade Agboola: Okay.
31 00:01:51.710 ⇒ 00:01:59.170 Hannah Wang: Dang, busy, busy week. I feel like being in the… the wedding party is always so chaotic, because there’s just…
32 00:01:59.730 ⇒ 00:02:07.670 Hannah Wang: Yeah, it’s just chaos the entire day, basically, because you have to, like, get ready and do all this stuff.
33 00:02:09.300 ⇒ 00:02:12.420 Demilade Agboola: I think for me, the most chaotic part is the Nigerian
34 00:02:12.460 ⇒ 00:02:30.080 Demilade Agboola: party part. Like, Nigerians have, like, different… like, the culture is different. So at one point in time, I had to go speak to both the groom’s parents and the bride’s parents, because they felt disrespected, and they wanted to take pictures again with the couple.
35 00:02:30.080 ⇒ 00:02:30.850 Hannah Wang: Whoa.
36 00:02:31.120 ⇒ 00:02:35.559 Demilade Agboola: Then there was… like, it’s just… it’s just very random stuff. It’s…
37 00:02:36.240 ⇒ 00:02:39.149 Demilade Agboola: Then at one point, we forgot the rings at the…
38 00:02:39.150 ⇒ 00:02:40.460 Hannah Wang: Oh, no!
39 00:02:40.460 ⇒ 00:02:58.200 Demilade Agboola: hotel that we spent the night in, so I had to get someone to, like, drive all the way there to get the rings, because we hadn’t told any of the officiating ministers, so we’re basically… the guy that drove there was racing against the clock, just to get back to get the rings in time.
40 00:02:58.680 ⇒ 00:02:59.375 Demilade Agboola: Then…
41 00:03:00.640 ⇒ 00:03:14.620 Demilade Agboola: It was pretty chaotic. Then there was something else, and I’m trying to remember. I know there was something else, like, weird, then at the reception, the guy responsible for bringing the groom’s clothes to… that he was going to change into, they didn’t bring them for some reason.
42 00:03:14.620 ⇒ 00:03:15.740 Hannah Wang: Oh, no.
43 00:03:15.740 ⇒ 00:03:18.979 Demilade Agboola: That’s another one to get too close.
44 00:03:19.290 ⇒ 00:03:30.869 Demilade Agboola: I don’t know, there’s just a bunch of things just happening, right? And I was just making sure that, you know, my friend had a very good day, showed him from all this, like, drama and everything, but…
45 00:03:30.870 ⇒ 00:03:31.380 Hannah Wang: Yeah.
46 00:03:31.470 ⇒ 00:03:40.049 Demilade Agboola: Like, people, people, people that saw me on, like, because it was on a Friday, people that saw me on, like, Friday night were like, I looked dead tired.
47 00:03:41.100 ⇒ 00:03:43.370 Demilade Agboola: They said, like, you look, you’re about to drop.
48 00:03:43.370 ⇒ 00:03:43.970 Hannah Wang: Yes.
49 00:03:43.970 ⇒ 00:03:51.040 Demilade Agboola: I still went for the after-party that we organized in the club, and I didn’t get home until, like, 6. I didn’t sleep until, like, 6.30.
50 00:03:51.350 ⇒ 00:03:52.570 Demilade Agboola: AM.
51 00:03:52.570 ⇒ 00:03:53.709 Hannah Wang: Oh my gosh.
52 00:03:54.590 ⇒ 00:04:03.260 Demilade Agboola: Yeah, it was just… it was just a lot. Saturday, I had a dinner with some of my… because, like, the city I met this guy in, and eventually…
53 00:04:03.400 ⇒ 00:04:18.819 Demilade Agboola: settle down in in Nigeria is where I started my tech journey. So, the hub, we organized the dinner for, like, because I was around, and people, some people were around as well, because they came in for his wedding. So we just organized the dinner, so there was organization around that. It was just… there was just so much going on.
54 00:04:18.829 ⇒ 00:04:20.219 Hannah Wang: Totally, yeah.
55 00:04:20.940 ⇒ 00:04:26.609 Demilade Agboola: I’m not the most social person, so I had literally called… I probably called more times about the past 7 days.
56 00:04:26.860 ⇒ 00:04:30.319 Demilade Agboola: like, number of calls than I have in the past, like, 3 months. Like, I don’t mix…
57 00:04:31.550 ⇒ 00:04:35.890 Demilade Agboola: I was drained, I was tired, I was just like, god damn it, but… Yeah.
58 00:04:35.890 ⇒ 00:04:55.200 Hannah Wang: It’s over now. Yo, that sounds… that is chaotic. I… if… I’m also not the most extroverted, so I feel like in those scenarios, I need to take, like, 2 weeks off of social interaction after such a big day of doing lots of stuff, so I get you.
59 00:04:55.730 ⇒ 00:04:57.780 Hannah Wang: Feeling tired, yeah.
60 00:04:59.820 ⇒ 00:05:02.950 Hannah Wang: Well, hopefully… work…
61 00:05:03.260 ⇒ 00:05:15.919 Hannah Wang: is it… it’s not the same type of tiring. I feel like it’s a different type of tiring, so hopefully you can recover socially, and mentally, and physically, so…
62 00:05:15.920 ⇒ 00:05:16.530 Demilade Agboola: Yes.
63 00:05:16.700 ⇒ 00:05:17.830 Demilade Agboola: That’s the plan.
64 00:05:17.830 ⇒ 00:05:19.889 Hannah Wang: Yep, I get you.
65 00:05:20.000 ⇒ 00:05:26.480 Hannah Wang: Okay, so, probably… this is probably going to be the first of many…
66 00:05:26.480 ⇒ 00:05:42.699 Hannah Wang: times I interview you, maybe in the future, it’ll be, like, second nature enough for you to record looms, so that we don’t have to, like, hop on a call, and we can both save time. But essentially, with case studies.
67 00:05:42.730 ⇒ 00:05:55.929 Hannah Wang: I mean, obviously, the bottleneck is usually, Lutam or Robert, because they’re… they were originally the ones drafting up all the content and the copy for all the case studies, but…
68 00:05:55.950 ⇒ 00:06:13.670 Hannah Wang: I think now I’m gonna go chase down the actual engineer, so that’s why we’re on the call. So it should be pretty straightforward. I… I’m probably going to… I’m gonna ask you to share your screen whenever you start, like, demoing and showing me things, and I’m going to…
69 00:06:13.670 ⇒ 00:06:24.059 Hannah Wang: ask you a series of questions, just so that I can take the transcript and run it through AI and get all the copy needed for the case study, so…
70 00:06:24.140 ⇒ 00:06:26.640 Hannah Wang: Even if the answer
71 00:06:27.110 ⇒ 00:06:45.730 Hannah Wang: even if you’ve already answered it, or even if it was redundant or obvious, I still want you to answer the question so that, I can catch everything in the transcript. So today, I know we’re going to talk about, all things urban STEMS related,
72 00:06:45.740 ⇒ 00:06:50.410 Hannah Wang: Forgive me for not knowing all the technical things, but…
73 00:06:50.490 ⇒ 00:06:58.410 Hannah Wang: I know initially there were two case studies, and then Utam asked if we could combine it into one case study.
74 00:06:58.510 ⇒ 00:07:10.640 Hannah Wang: So I don’t know how you want to, like, walk through that, but basically it’s the looker cleanup, and then ETL ingestion cleanup. So I guess it’s, like, a urban strength cleanup case study.
75 00:07:10.640 ⇒ 00:07:23.309 Hannah Wang: With Looker and the ETL ingestion. So, if there’s anything that you can, like, show me and share screen with, so later I can maybe get a screen grab of it, that would be helpful.
76 00:07:24.430 ⇒ 00:07:31.980 Hannah Wang: So I’m just gonna start with the first set of questions, and then we can go from there. Do you have any questions for me before we start?
77 00:07:32.660 ⇒ 00:07:35.069 Demilade Agboola: No, not yet.
78 00:07:35.250 ⇒ 00:07:40.900 Hannah Wang: Okay, yeah, not yet. Okay, so…
79 00:07:41.070 ⇒ 00:07:51.400 Hannah Wang: For the liquor cleanup, for the whole project for Urban STEM, do you know how long it took, or I guess approximately which quarter?
80 00:07:51.620 ⇒ 00:07:54.060 Hannah Wang: You… we worked on it.
81 00:07:56.110 ⇒ 00:07:59.840 Demilade Agboola: So it took about a month-ish?
82 00:08:00.250 ⇒ 00:08:07.729 Demilade Agboola: Okay. And I think we did it in, like, Q2, this Q3, everything in, like, Q2.
83 00:08:08.090 ⇒ 00:08:08.900 Hannah Wang: Okay.
84 00:08:09.090 ⇒ 00:08:14.450 Hannah Wang: And then who were the team members involved? It was…
85 00:08:14.710 ⇒ 00:08:16.639 Hannah Wang: I don’t know if, like, like.
86 00:08:17.290 ⇒ 00:08:22.960 Hannah Wang: previous people worked on stuff. But yeah, do you know who the team members were?
87 00:08:23.710 ⇒ 00:08:31.709 Demilade Agboola: So it was largely, Kyle driving it, but Utam and I were also, like, supporting and giving advice on how we should go about setting things.
88 00:08:31.840 ⇒ 00:08:32.510 Hannah Wang: Okay.
89 00:08:33.280 ⇒ 00:08:44.360 Hannah Wang: And then do you have, like, a… like, what would you call this project type? Is it just, like, deprecation, or… yeah, I guess, like, what would you call it?
90 00:08:45.500 ⇒ 00:08:51.329 Demilade Agboola: I would say it’s more of a… yeah, cleanup and deprecation.
91 00:08:51.330 ⇒ 00:08:51.990 Hannah Wang: Okay.
92 00:08:53.920 ⇒ 00:09:02.780 Hannah Wang: Okay, cool. So, I’m gonna move into more context-related questions. So, I’m just trying to understand, the environment
93 00:09:02.950 ⇒ 00:09:14.779 Hannah Wang: the working environment of urban setup and infrastructure and all that stuff before you implemented the cleanup. So yeah, what was the working environment like before
94 00:09:15.000 ⇒ 00:09:23.640 Hannah Wang: you started the cleanup? Like, was everything all over the place? Was it messy? Was it super complex? Like, just describe to me what it was like before.
95 00:09:25.150 ⇒ 00:09:30.649 Demilade Agboola: So it was just super messy. So basically, what had happened was…
96 00:09:30.870 ⇒ 00:09:36.680 Demilade Agboola: As years had gone by, people were building out different, like, looks.
97 00:09:36.910 ⇒ 00:09:41.260 Demilade Agboola: and explorers and dashboards in Looker.
98 00:09:41.410 ⇒ 00:09:46.780 Demilade Agboola: And no one was really, like, maintaining it, or pruning it, or just trying to get things back in order.
99 00:09:47.370 ⇒ 00:09:56.090 Demilade Agboola: So sometimes you’d have dashboards that I had not been used in a very long time, you would have things that were just not, you know.
100 00:09:56.380 ⇒ 00:10:06.719 Demilade Agboola: useful anymore, or one-time… things that were meant to be one-time reports that still, you know, existed. So obviously, you know, it was hard to…
101 00:10:07.410 ⇒ 00:10:09.040 Demilade Agboola: maintain this.
102 00:10:09.290 ⇒ 00:10:15.859 Demilade Agboola: And things were just growing upwards and outwards, so just being able to curtail it, being able to get things back into a state where
103 00:10:16.050 ⇒ 00:10:17.920 Demilade Agboola: Everything that is in there.
104 00:10:18.780 ⇒ 00:10:30.760 Demilade Agboola: is useful, and serves a purpose. And they had, like, well, 800, over 800, like, data objects in terms… in Looker, right? Or, data items in Looker.
105 00:10:31.140 ⇒ 00:10:40.320 Demilade Agboola: And that wasn’t, like, if you have over 800, you can’t always find what’s useful. Some things are using old sources. It’s just not helpful to have that much.
106 00:10:40.450 ⇒ 00:10:43.059 Demilade Agboola: So, the idea was we needed to clean it up.
107 00:10:43.130 ⇒ 00:11:02.320 Demilade Agboola: And in some cases, some things were created by people who were not at the company anymore, so obviously, if the logic had changed, if things had… were not being maintained properly, it’s just bad data. And so, it was important for us to be able to get to a point where we could clean up as much as possible.
108 00:11:03.460 ⇒ 00:11:07.149 Hannah Wang: Okay, and do you know if the Urban STEM
109 00:11:07.270 ⇒ 00:11:17.759 Hannah Wang: like, previously tried to, like, clean up everything, or was it kind of their first time, doing it? Like, cleaning it up and stuff?
110 00:11:18.450 ⇒ 00:11:29.560 Demilade Agboola: I’m not sure if they have tried to, but if they have tried to, and this was the state after, that’s quite disturbing, to be honest. But, I’m not sure, I’m not sure if they had.
111 00:11:29.560 ⇒ 00:11:29.900 Hannah Wang: Okay.
112 00:11:29.900 ⇒ 00:11:30.330 Demilade Agboola: partner.
113 00:11:30.540 ⇒ 00:11:31.579 Demilade Agboola: No worries.
114 00:11:31.770 ⇒ 00:11:39.159 Hannah Wang: Okay, so kind of moving on to the challenges that the client was facing. So…
115 00:11:39.160 ⇒ 00:11:52.699 Hannah Wang: I guess due to the nature of messy data and just it being everywhere, do you know what types of problems that urban stems, what they were experiencing? I know they’re, like, a flower.
116 00:11:52.740 ⇒ 00:12:06.810 Hannah Wang: shop flower company, so did it… did it, like, mess with the orders, or did it, like, slow things down, shipping, all that stuff? Like, yeah, tell me, like, what types of problems they’re running into because of messy data.
117 00:12:07.830 ⇒ 00:12:14.790 Demilade Agboola: The issues that they had were not necessarily reflected in terms of
118 00:12:16.280 ⇒ 00:12:22.880 Demilade Agboola: In terms of… so they have issues that… with their data generally, yes, but in terms of, like, the cleanup and deprecation part.
119 00:12:23.860 ⇒ 00:12:28.450 Demilade Agboola: What ended up happening is… Is that compounding?
120 00:12:28.670 ⇒ 00:12:31.890 Demilade Agboola: The ability to find useful data.
121 00:12:32.200 ⇒ 00:12:37.419 Hannah Wang: Because there’s so many things going on, so many people are responsible for so many things.
122 00:12:37.420 ⇒ 00:12:47.539 Demilade Agboola: it’s sort of really hard to pinpoint the most accurate data. So, obviously, you’re using… you’re using time to be able to figure out what the data, the most useful data is.
123 00:12:47.890 ⇒ 00:12:50.750 Demilade Agboola: Unless someone points you in the right direction, and…
124 00:12:50.990 ⇒ 00:12:55.200 Demilade Agboola: that… that also takes time, so that’s, like, 9 hours. It’s a function of…
125 00:12:55.510 ⇒ 00:13:04.059 Demilade Agboola: Because of the complexity, unnecessary complexity of everything, it’s not the easiest to figure things out, and that also, like, played a role.
126 00:13:05.280 ⇒ 00:13:11.820 Hannah Wang: And do you know how, like, what exactly, like, how that impacts those decisions?
127 00:13:12.280 ⇒ 00:13:18.430 Hannah Wang: Like, having… not having access to the data, like, quickly, and it being all over the place.
128 00:13:20.080 ⇒ 00:13:25.660 Demilade Agboola: Yeah, I mean, it basically meant that, like, some of the… the analysis that they would need to do
129 00:13:25.890 ⇒ 00:13:29.000 Demilade Agboola: warrants… I bet.
130 00:13:29.390 ⇒ 00:13:48.649 Demilade Agboola: easy to find, or most useful. And the analysis varies from a lot of things. We’re talking things like inventory, we’re talking things like, risk orders, audits, we’re talking things like, repeat customers. Like, they have a wide variety. Again, when you have over 800 plus, like, data objects.
131 00:13:48.700 ⇒ 00:13:55.409 Demilade Agboola: In Looker, yeah, there’s a wide variety of things. Some of these are also not useful anymore.
132 00:13:55.530 ⇒ 00:14:02.009 Demilade Agboola: So again, there was always the risk of people utilizing old data or old logic.
133 00:14:02.160 ⇒ 00:14:08.229 Demilade Agboola: To… to make… Current reports, or current,
134 00:14:08.380 ⇒ 00:14:19.590 Demilade Agboola: current decision. So, being able to figure that part out, and being able to help them reduce it to a point where it’s manageable, everything’s easy to access, and easy to utilize.
135 00:14:20.280 ⇒ 00:14:20.919 Hannah Wang: Got it.
136 00:14:21.040 ⇒ 00:14:25.109 Hannah Wang: Okay, cool. So, moving on to the solution.
137 00:14:25.110 ⇒ 00:14:41.470 Hannah Wang: So yeah, you can feel free to dive into all the technical nitty-gritty and throw out jargon. Even if I don’t understand it, it’ll be good to put it into the case study, so don’t worry about, like, if I understand it or not. So yeah, feel free to just…
138 00:14:41.730 ⇒ 00:14:43.140 Hannah Wang: Yeah, tell me.
139 00:14:43.480 ⇒ 00:14:47.990 Hannah Wang: How you cleaned everything up, and what the solution involved.
140 00:14:49.600 ⇒ 00:14:53.969 Demilade Agboola: Okay, so what we did, we had a two-step process. I’m kind of sharing my screen now.
141 00:14:54.160 ⇒ 00:14:54.810 Hannah Wang: Yes.
142 00:14:55.130 ⇒ 00:15:01.719 Demilade Agboola: Alright, so we have the two-step process. The first step was gathering every single, like, dashboard and data artifact.
143 00:15:02.030 ⇒ 00:15:03.700 Demilade Agboola: from Looker.
144 00:15:03.840 ⇒ 00:15:07.589 Demilade Agboola: We put it in here, so that’s how we know we have 837.
145 00:15:08.100 ⇒ 00:15:08.970 Demilade Agboola: Products.
146 00:15:09.810 ⇒ 00:15:14.500 Demilade Agboola: And then we were able to run the metadata analysis on it.
147 00:15:14.650 ⇒ 00:15:17.749 Demilade Agboola: So we queried for the metadata, so things like who…
148 00:15:18.650 ⇒ 00:15:22.210 Demilade Agboola: The username that created it, so that was the first step.
149 00:15:22.630 ⇒ 00:15:27.800 Demilade Agboola: The next thing we did was… when was it? Who was it? The coordinated last?
150 00:15:28.740 ⇒ 00:15:33.340 Demilade Agboola: So, again, that’s another thing we did. And then the last thing we did was, like.
151 00:15:33.680 ⇒ 00:15:43.760 Demilade Agboola: when last was he accessed? So, again, we’re trying to see how useful this is to the company, because obviously something that’s not been accessed since, for instance, 2020.
152 00:15:44.370 ⇒ 00:15:48.210 Demilade Agboola: Isn’t something that is the most important to see 2023.
153 00:15:48.660 ⇒ 00:15:53.129 Demilade Agboola: And then view count, so how often is it viewed?
154 00:15:53.430 ⇒ 00:15:55.390 Demilade Agboola: By whoever in the company.
155 00:15:55.750 ⇒ 00:15:58.020 Demilade Agboola: So again, that’s another thing we did.
156 00:15:59.080 ⇒ 00:16:05.670 Demilade Agboola: And so, we also then added, based off the models that were feeding it in DBT,
157 00:16:05.990 ⇒ 00:16:08.059 Demilade Agboola: How accurate are they?
158 00:16:08.240 ⇒ 00:16:14.439 Demilade Agboola: So, like, how accurate… how we found the models to be within DBT?
159 00:16:14.610 ⇒ 00:16:29.550 Demilade Agboola: If we see that the, you know, maybe tests are failing with the model, or, like, the numbers are not the most accurate numbers, therefore, this leans to that dashboard not being the most accurate, and we also would not want to, like, pass that number on to the customers, or to the stakeholders, right?
160 00:16:29.650 ⇒ 00:16:33.210 Demilade Agboola: So, being able to, like, say, okay.
161 00:16:34.770 ⇒ 00:16:36.930 Demilade Agboola: Is it being used a lot? Yes.
162 00:16:37.120 ⇒ 00:16:39.300 Demilade Agboola: But is it accurate? No.
163 00:16:39.460 ⇒ 00:16:41.899 Demilade Agboola: Or is it being used a lot? No.
164 00:16:42.180 ⇒ 00:16:58.420 Demilade Agboola: And is it accurate? No, then we will deprecate. So we finally used those, like, those logic together, created the deprecation layer, and then we also passed it on to the clients as well, that, hey, we’re going to deprecate these tables, or sorry, these looks, these dashboards.
165 00:16:58.780 ⇒ 00:17:07.680 Demilade Agboola: what are your… do you have any counter or pushback against it? And so this is where we have the outputs, the inputs here as well.
166 00:17:08.510 ⇒ 00:17:12.999 Demilade Agboola: And so we were able to factor everything together and come up with the final deprecation layer.
167 00:17:13.490 ⇒ 00:17:16.510 Demilade Agboola: We’re like, okay, these are what we’re gonna deprecate.
168 00:17:16.910 ⇒ 00:17:20.309 Demilade Agboola: And we finally deprecated these ones.
169 00:17:21.740 ⇒ 00:17:27.670 Demilade Agboola: So there’s a link to the different, like, looks, dashboards, and, you know, other artifacts.
170 00:17:27.880 ⇒ 00:17:33.410 Demilade Agboola: So in case you wanted to go quickly and look at it and see what exactly was going on here, you could.
171 00:17:33.950 ⇒ 00:17:38.910 Demilade Agboola: But part of what I also said about, like, people who
172 00:17:39.460 ⇒ 00:17:47.189 Demilade Agboola: who are utilizing it. So, for instance, this person here, he hasn’t… he has… he’s left the company as of the beginning of this year.
173 00:17:47.390 ⇒ 00:17:50.930 Demilade Agboola: And I believe the person who created it also is a company.
174 00:17:51.060 ⇒ 00:18:03.939 Demilade Agboola: And there are a number of people like that. So ultimately, you had, like, logic being built out, the people had left, they had not been utilized since 2020, and so it’s… in that case, you would want to get rid of it, because
175 00:18:04.190 ⇒ 00:18:06.659 Demilade Agboola: Yeah, who’s really using it?
176 00:18:07.830 ⇒ 00:18:08.440 Hannah Wang: Yeah.
177 00:18:09.080 ⇒ 00:18:09.750 Demilade Agboola: Yeah.
178 00:18:12.540 ⇒ 00:18:21.159 Hannah Wang: to the right a little bit, that column that says DBT… I don’t know if you…
179 00:18:21.490 ⇒ 00:18:25.119 Hannah Wang: Yeah, what’s the DBT accuracy?
180 00:18:26.780 ⇒ 00:18:33.929 Demilade Agboola: Yeah, so… Their infrastructure requires dbt to transform the data.
181 00:18:34.870 ⇒ 00:18:38.330 Demilade Agboola: And then… the Looker… LookML.
182 00:18:39.720 ⇒ 00:18:44.239 Demilade Agboola: is built on top of dbt, and that is shown within Looker.
183 00:18:45.380 ⇒ 00:18:52.899 Demilade Agboola: So, if dbt isn’t handling data properly, or if there are question marks about the models that, you know, dbt
184 00:18:53.370 ⇒ 00:19:03.579 Demilade Agboola: the question marks about the dbt models. Well, unfortunately, that just means that the data that the LookerML and Looker is built on top are inaccurate models.
185 00:19:04.180 ⇒ 00:19:07.939 Demilade Agboola: Or models that are not giving out the information that the… that…
186 00:19:08.200 ⇒ 00:19:10.179 Demilade Agboola: In the form that they need.
187 00:19:10.750 ⇒ 00:19:11.749 Demilade Agboola: Right. Got it.
188 00:19:11.920 ⇒ 00:19:14.920 Demilade Agboola: So, if the data is bad.
189 00:19:15.170 ⇒ 00:19:27.049 Demilade Agboola: then the model, whether it’s being utilized or not, is questionable. Like, the output is questionable, the dashboard is being utilized is questionable. So we don’t also want… because even as much as we’re trying to, like, clean off
190 00:19:27.150 ⇒ 00:19:35.680 Demilade Agboola: bloatware from the perspective of utilization, or, when last it was accessed. We also have to think about it from the perspective of
191 00:19:35.860 ⇒ 00:19:40.659 Demilade Agboola: Is the data good as well? So there are 3 main things. Is the data good?
192 00:19:40.890 ⇒ 00:19:42.460 Demilade Agboola: When last was he accessed?
193 00:19:42.850 ⇒ 00:19:44.370 Demilade Agboola: How often is it viewed?
194 00:19:44.650 ⇒ 00:19:48.169 Demilade Agboola: If it’s not viewed often, then it’s not really that important.
195 00:19:48.310 ⇒ 00:19:51.570 Demilade Agboola: Especially if it has not been accessed in a very long time.
196 00:19:51.870 ⇒ 00:19:59.620 Demilade Agboola: But also, we also wanted to ensure that even if it was being accessed recently, or it was accessed, you know, frequently.
197 00:20:00.200 ⇒ 00:20:02.789 Demilade Agboola: It still has to be good data.
198 00:20:05.670 ⇒ 00:20:06.359 Hannah Wang: Got it.
199 00:20:06.960 ⇒ 00:20:07.390 Demilade Agboola: Yeah.
200 00:20:07.390 ⇒ 00:20:12.180 Hannah Wang: And then for all the 800-something tables, like.
201 00:20:12.660 ⇒ 00:20:26.789 Hannah Wang: How did you get all the names, and how did you, like, import it? Or not, I guess not import it, but yeah, did you just, like, go… did they send you that list, or did you have to go, like…
202 00:20:27.120 ⇒ 00:20:30.759 Demilade Agboola: Yeah, there was a script that was created to extract this data.
203 00:20:31.070 ⇒ 00:20:35.099 Demilade Agboola: So you run against the local instance, local user.
204 00:20:35.300 ⇒ 00:20:42.710 Demilade Agboola: And was able to extract the metadata from it as well. I see. It came as a CSV, and the CSV was uploaded.
205 00:20:43.250 ⇒ 00:20:46.299 Demilade Agboola: So these were what we were able to get out of it.
206 00:20:46.300 ⇒ 00:20:47.760 Hannah Wang: Gotcha, okay.
207 00:20:48.560 ⇒ 00:20:50.680 Hannah Wang: Okay, cool.
208 00:20:51.250 ⇒ 00:20:59.650 Hannah Wang: So, as a result of cleaning this up, do you know how many dashboards were stale and removed, and how many are still, yeah, used?
209 00:21:00.580 ⇒ 00:21:02.920 Demilade Agboola: So we could quickly look at this.
210 00:21:29.740 ⇒ 00:21:30.650 Demilade Agboola: Oh, boy.
211 00:21:34.860 ⇒ 00:21:37.620 Demilade Agboola: So, I’m sorry, we use this as well.
212 00:21:56.260 ⇒ 00:22:01.030 Demilade Agboola: So, 740 were what we deprecated, or were set up for deprecation.
213 00:22:02.490 ⇒ 00:22:03.380 Hannah Wang: Oh, wow.
214 00:22:03.840 ⇒ 00:22:06.500 Demilade Agboola: Yeah, so… Crazy.
215 00:22:07.090 ⇒ 00:22:10.010 Demilade Agboola: Yeah, so that’s when, you know, there was a lot of blood work.
216 00:22:10.480 ⇒ 00:22:11.310 Hannah Wang: I see.
217 00:22:11.760 ⇒ 00:22:18.590 Hannah Wang: Okay, so 740 were deprecated, so the remainder is still, obviously, up and running.
218 00:22:19.150 ⇒ 00:22:19.940 Demilade Agboola: Yes, sir.
219 00:22:20.650 ⇒ 00:22:25.799 Demilade Agboola: So, instead of 836, 740 means, like, 98 are left.
220 00:22:27.340 ⇒ 00:22:29.109 Demilade Agboola: Or 94, actually.
221 00:22:31.350 ⇒ 00:22:37.370 Hannah Wang: Cool. Any other, like, metrics?
222 00:22:37.670 ⇒ 00:22:38.670 Hannah Wang: that…
223 00:22:38.930 ⇒ 00:22:52.489 Hannah Wang: happened as a result of this. Like, for case studies, I always try to put numbers in, because I feel like numbers always aren’t catchy, and people like numbers. So, I know that, I can put the percentage of
224 00:22:52.710 ⇒ 00:23:01.750 Hannah Wang: Dashboards that were deprecated, but were there any other numbers as a result of, this cleanup?
225 00:23:04.860 ⇒ 00:23:06.070 Hannah Wang: If not, it’s okay.
226 00:23:07.090 ⇒ 00:23:11.169 Demilade Agboola: No, not really. I think that’s the major number. Okay.
227 00:23:11.610 ⇒ 00:23:18.580 Demilade Agboola: I think, ultimately, there might be some business context in being able to quantify, like, the man hours saved.
228 00:23:19.440 ⇒ 00:23:25.569 Demilade Agboola: Or, like, you know… Yeah, the man hours saved by just having unnecessary, like, things removed.
229 00:23:25.820 ⇒ 00:23:33.789 Demilade Agboola: And after, when we’re done, we’re able to, like, just try and organize things in an easier-to-find way, because now, instead of having 800+,
230 00:23:33.960 ⇒ 00:23:37.690 Demilade Agboola: It’s actually, you know, organize 94 dashboards.
231 00:23:39.450 ⇒ 00:23:43.050 Hannah Wang: Yeah, that makes sense, but you don’t happen to have that number, or…
232 00:23:43.170 ⇒ 00:23:47.500 Hannah Wang: I… yeah, it’s hard to know how many hours were saved, I guess.
233 00:23:47.970 ⇒ 00:23:48.780 Demilade Agboola: Yeah.
234 00:23:49.400 ⇒ 00:23:58.920 Hannah Wang: Okay, is there any feedback that we got from the Urban STEM team, that you can remember about
235 00:23:59.500 ⇒ 00:24:01.690 Hannah Wang: This deprecation work.
236 00:24:03.170 ⇒ 00:24:05.770 Demilade Agboola: No, nothing specifically.
237 00:24:07.090 ⇒ 00:24:08.900 Hannah Wang: positive feedback, though?
238 00:24:09.800 ⇒ 00:24:21.730 Demilade Agboola: I mean, yeah, I mean, the stakeholders… one of the stakeholders, Zach, wanted us to even cut even more, even more. He wanted the numbers to be as low as possible, and just get rid of any blow. I think, again.
239 00:24:22.460 ⇒ 00:24:32.430 Demilade Agboola: it was just… it had reached a point where I guess they were, like, really frustrated and tired with all the data that existed, and wanted, like, as deep cuts as possible, to be honest.
240 00:24:32.720 ⇒ 00:24:39.320 Demilade Agboola: So yeah, I think you had just reached the point where The… the things being…
241 00:24:39.640 ⇒ 00:24:41.819 Demilade Agboola: The things that were still left.
242 00:24:42.080 ⇒ 00:24:45.100 Demilade Agboola: They had to be really, really essential, basically.
243 00:24:45.650 ⇒ 00:24:46.210 Hannah Wang: Yeah.
244 00:24:47.030 ⇒ 00:24:48.069 Hannah Wang: That makes sense.
245 00:24:48.500 ⇒ 00:24:52.529 Hannah Wang: Let’s see…
246 00:24:53.180 ⇒ 00:25:01.180 Hannah Wang: Oh, another thing I wanted to ask was about the tools that you use. So, obviously, Looker was involved.
247 00:25:01.420 ⇒ 00:25:06.810 Hannah Wang: Dbt, I’m assuming, is involved. Was there any other…
248 00:25:07.350 ⇒ 00:25:11.659 Hannah Wang: Tool, or, like, tech stack that you can…
249 00:25:12.020 ⇒ 00:25:16.699 Hannah Wang: remember off the top of your head, just so I can put the logos on the case study.
250 00:25:19.130 ⇒ 00:25:26.180 Demilade Agboola: And… just also… Us also, like, going through…
251 00:25:28.340 ⇒ 00:25:30.819 Demilade Agboola: I’m just also going to, like, Redshift.
252 00:25:31.110 ⇒ 00:25:32.320 Hannah Wang: Okay. I’ve seen…
253 00:25:32.320 ⇒ 00:25:39.060 Demilade Agboola: Yeah, what exactly… I’m trying to see if I can find the query that Kyle made.
254 00:26:32.840 ⇒ 00:26:35.740 Demilade Agboola: Sorry, struggling to find the query.
255 00:26:35.900 ⇒ 00:26:36.830 Hannah Wang: Right, okay.
256 00:26:46.680 ⇒ 00:26:48.449 Demilade Agboola: I’ll try and locate it and send it to you.
257 00:26:48.860 ⇒ 00:26:49.560 Hannah Wang: Okay.
258 00:26:50.230 ⇒ 00:26:57.520 Hannah Wang: Okay, cool. So Redshift, DPT, Looker… Okay, cool.
259 00:26:57.930 ⇒ 00:27:06.610 Hannah Wang: Oh… I’m trying to see if there are any other questions I need to get from you.
260 00:27:07.030 ⇒ 00:27:09.529 Hannah Wang: Mah, let’s see…
261 00:27:23.480 ⇒ 00:27:27.050 Hannah Wang: I think that’s… I think that’s it.
262 00:27:27.280 ⇒ 00:27:34.090 Hannah Wang: If there’s anything else I need from you, I’ll just DM you, but I think this was good. A good start.
263 00:27:34.690 ⇒ 00:27:40.969 Hannah Wang: Yeah, you can send me that query, whenever you get a chance, and then I’ll probably…
264 00:27:41.080 ⇒ 00:27:48.410 Hannah Wang: Design the case study, and then have you check it, just so that the facts… are correct.
265 00:27:48.690 ⇒ 00:27:49.260 Hannah Wang: I don’t know.
266 00:27:49.680 ⇒ 00:27:50.760 Hannah Wang: They’ve got them down.
267 00:27:51.290 ⇒ 00:27:54.050 Hannah Wang: Alright, thank you, Demolade, I appreciate your time.
268 00:27:54.790 ⇒ 00:27:56.040 Demilade Agboola: Okay, sounds good.
269 00:27:56.190 ⇒ 00:27:57.670 Hannah Wang: Alright, talk soon.
270 00:27:58.180 ⇒ 00:27:59.709 Demilade Agboola: Alright, talk soon. Bye.