Meeting Title: Magic Spoon — Brainforge sync Date: 2026-01-15 Meeting participants: Demilade Agboola, Cinnamon Toast, Ashwini Sharma, Justin Tabarini, Michael Thorson


WEBVTT

1 00:00:19.000 00:00:23.789 Cinnamon Toast: Beef. Very beef heavy. You’re gone, you should get it on.

2 00:00:24.400 00:00:28.010 Cinnamon Toast: I got my, I got my Thursday curry, so good.

3 00:00:29.560 00:00:30.630 Cinnamon Toast: Hey, Demon!

4 00:00:34.330 00:00:35.589 Cinnamon Toast: We can’t hear you.

5 00:00:36.720 00:00:37.930 Demilade Agboola: How’s everyone doing?

6 00:00:40.200 00:00:45.459 Demilade Agboola: Sorry, so I have a microphone, like, an actual microphone, so you have to switch it on, so I have.

7 00:00:45.460 00:00:45.980 Cinnamon Toast: Hmm.

8 00:00:45.980 00:00:49.140 Demilade Agboola: Mute myself and turn it on, so that was… that was the…

9 00:00:51.630 00:00:52.340 Cinnamon Toast: Oh, good.

10 00:00:52.850 00:00:54.350 Demilade Agboola: That’s good to hear, that’s good to hear.

11 00:00:55.200 00:00:57.599 Demilade Agboola: Are we ready to start?

12 00:00:57.790 00:00:58.969 Demilade Agboola: Are we waiting for anyone?

13 00:01:00.570 00:01:01.530 Cinnamon Toast: I think we’re ready!

14 00:01:01.530 00:01:02.600 Demilade Agboola: Alright, let’s go.

15 00:01:10.330 00:01:11.190 Cinnamon Toast: Oh, goodness.

16 00:01:12.320 00:01:20.340 Demilade Agboola: Okay, so, just, before we get into it, this was more of a slower week compared to, like, the previous weeks we’ve had,

17 00:01:20.400 00:01:33.650 Demilade Agboola: the blockers we’ve mentioned had, like, you know, restricted how much we could get done this week. That being said, we still tried to make the best of, like, what we had, and move things as quickly as we could with what we had.

18 00:01:34.330 00:01:42.199 Demilade Agboola: So again, just same old reminder, this is our end goal, so we all remember what the…

19 00:01:42.200 00:01:44.569 Cinnamon Toast: final statement needs to be like.

20 00:01:46.200 00:02:00.009 Demilade Agboola: And so, yes, we will go into that. So, in terms of our Spain’s API update, I will hand over to Awesh, but, like, in terms of what we’ve done, we’ve been able to have our local, like, data

21 00:02:01.280 00:02:10.379 Demilade Agboola: in, like, not Wish, Ashwini’s system, and Ashwini will be looking at comparing it to, like, the Redshift,

22 00:02:11.390 00:02:14.530 Demilade Agboola: this redshift table that Michael sent over to him.

23 00:02:16.000 00:02:18.359 Demilade Agboola: So wait, Mashrini, can you please?

24 00:02:19.110 00:02:24.099 Ashwini Sharma: Yeah, so I’m still waiting for the table that, Michael said would be ready,

25 00:02:24.290 00:02:28.510 Ashwini Sharma: It’s not yet ready, right, Michael? I haven’t checked it over here, but .

26 00:02:29.430 00:02:35.349 Speaker 1 (Cinnamon Toast): I have an update on that. Yeah, it looks like it timed out last night, just like it was…

27 00:02:36.135 00:02:37.844 Cinnamon Toast: Yeah, it was just too big of a file.

28 00:02:37.850 00:02:45.560 Speaker 1 (Cinnamon Toast): to upload directly. So, right after this meeting, I have some time cut out to just, like, get that file into the warehouse.

29 00:02:47.650 00:02:57.629 Speaker 1 (Cinnamon Toast): Yeah, that’s… that’s the update there. So sorry about that. I thought it would have run overnight, but I think my laptop, like, locally just, like, timed out and restarted.

30 00:02:57.950 00:03:05.189 Ashwini Sharma: Okay, no, no issues, yeah. So, the other update from my side is, like, I had already loaded,

31 00:03:05.580 00:03:08.789 Ashwini Sharma: SPIN’s data for up to December 28th.

32 00:03:09.020 00:03:11.920 Ashwini Sharma: 28th of December 2025, that’s the latest

33 00:03:11.960 00:03:23.910 Ashwini Sharma: date for which we had data, and then what I did is I created another table based on top of it, which aggregates the data for different timeframes for every week of data that we have.

34 00:03:23.930 00:03:32.830 Ashwini Sharma: So, what that means is, like, we go back up to 4 weeks from the latest week, the week before that, the week before that, and so on.

35 00:03:32.920 00:03:42.029 Ashwini Sharma: We go back 4 weeks, we go back 12 weeks, we go back 24, 52 weeks, and then we can see those, metrics.

36 00:03:42.120 00:03:56.110 Ashwini Sharma: associated with each week, right? And then you can run your analysis on top of that. I’ll share the table name. The dbt model is also ready with me. I’ll share that as a PR with you.

37 00:03:56.320 00:03:58.389 Ashwini Sharma: If you want, you can take a look at it.

38 00:03:58.710 00:04:13.259 Ashwini Sharma: The other PR is also almost ready, I just want to add a few documentation on that, and then I’ll share that as well, which is the pipeline-related PR, which loads, and the way that I have created it is it only runs on

39 00:04:13.320 00:04:19.560 Ashwini Sharma: data release date. So, if you could tell me, like, at what time of the day the data is released?

40 00:04:19.769 00:04:24.870 Ashwini Sharma: on those dates, I could alter the… the frequency,

41 00:04:25.480 00:04:28.439 Ashwini Sharma: I mean, I mean, the Quran schedule accordingly, so that

42 00:04:28.540 00:04:31.069 Ashwini Sharma: whenever it runs it, it has, I mean.

43 00:04:31.610 00:04:34.749 Ashwini Sharma: It ensures that data is already there, right?

44 00:04:35.500 00:04:52.339 Speaker 1 (Cinnamon Toast): Yeah, I think just, like, like, there’s no, like, day, like, day of rush to pull that data. I mean, currently it takes our user, like, 4 days to get all that data out, so I would say, like, just to ensure we get a catch is, like, running that overnight. Like, the day of the release, running it after hours.

45 00:04:52.460 00:04:56.800 Speaker 1 (Cinnamon Toast): And then the next morning, the data should be in my house. That’s, I think…

46 00:04:57.150 00:04:58.330 Speaker 1 (Cinnamon Toast): The goal for us.

47 00:04:58.630 00:05:02.420 Ashwini Sharma: Okay, cool. I’ll modify the logic accordingly in that case.

48 00:05:03.130 00:05:07.330 Speaker 1 (Cinnamon Toast): Just to make sure we get a catch. For… for the aggregate metrics.

49 00:05:07.600 00:05:16.680 Speaker 2 (Cinnamon Toast): Like, a big question I have is, like, there’s, like, TDP, ACV, which are, like, percentages, like, how did you aggregate those measures?

50 00:05:17.210 00:05:24.760 Ashwini Sharma: Probably I didn’t pull the percentages. I had a bunch of attributes in the Excel sheet that Michael had shared earlier.

51 00:05:24.900 00:05:32.080 Ashwini Sharma: So I took only those attributes. But if that percentages are something that could be calculated, I guess.

52 00:05:33.610 00:05:38.220 Speaker 2 (Cinnamon Toast): Yeah, I think more calling out, we might need to set up a meeting with, like, Sweeney, Heather.

53 00:05:38.740 00:05:41.439 Speaker 2 (Cinnamon Toast): And that’s to, like… Yeah. …take…

54 00:05:41.560 00:05:51.440 Speaker 2 (Cinnamon Toast): figure out, can… do we… like, it’d be good to know, like, right now we’re pulling detailed-level data, and we plan on aggregating it, but if Heather tells us, like.

55 00:05:51.570 00:05:52.960 Speaker 2 (Cinnamon Toast): Oh, no.

56 00:05:53.130 00:05:58.560 Speaker 2 (Cinnamon Toast): ACV cannot be aggregated by doing some sort of weighted average.

57 00:05:59.240 00:06:08.400 Speaker 2 (Cinnamon Toast): it can only be pulled at the brand level, or pulled at the, like, individual level. That’s something we should probably know, because it’ll, like, change the way we pull data, where we’ll have to pull

58 00:06:08.810 00:06:11.419 Speaker 2 (Cinnamon Toast): unit and diarrhea data.

59 00:06:11.800 00:06:18.369 Speaker 2 (Cinnamon Toast): So I think we might need to set up a meeting and just go through each metric a bit. Have item-level, store-level data.

60 00:06:18.500 00:06:21.860 Speaker 2 (Cinnamon Toast): Can I sump this? Revenue? Yes.

61 00:06:22.105 00:06:24.365 Speaker 1 (Cinnamon Toast): Or, like, how… yeah.

62 00:06:24.365 00:06:43.965 Speaker 2 (Cinnamon Toast): Because I think the answer that’s going to happen is ACV, I don’t know if this is a… maybe I’m saying it wrong, is a percent of stores that this product is in. And so when you pull item-level data, it’s always going to be 100, because you’re never going to have the line of zero.

63 00:06:44.315 00:06:48.354 Speaker 2 (Cinnamon Toast): And so it probably won’t work on a data pool like that.

64 00:06:48.365 00:07:05.584 Speaker 2 (Cinnamon Toast): So it can just be good to go through these things, like, with Heather, live, just maybe on, like, pulling up the data, just have just a working session, and just… Yeah. …figure out, can we sum these terms, can we not? Can we average? TDP is a very important one that I want to figure out. I have no idea if you can average it.

65 00:07:05.585 00:07:13.924 Speaker 2 (Cinnamon Toast): Yeah. It’s, like, items per store, per something. It’s, like, weird. Yeah. So I think, like, we need someone who knows what they’re talking about to… Yeah.

66 00:07:13.930 00:07:18.400 Speaker 1 (Cinnamon Toast): with Heather, fly through those things as well, and maybe we come together and, like, develop a plan.

67 00:07:18.400 00:07:29.339 Demilade Agboola: Would it be possible to get a list of these terms and how they’re defined prior to the meeting with Heather? So maybe we can get it, say, tomorrow, and then maybe we have a meeting with Heather on, like, Tuesday, and so we can…

68 00:07:29.740 00:07:39.170 Demilade Agboola: We have some either… some ideas on how we might go about it, as well as, like, what the definitions of those,

69 00:07:39.850 00:07:41.649 Demilade Agboola: Like ACV, yeah.

70 00:07:42.250 00:07:59.100 Speaker 1 (Cinnamon Toast): I have… we have a documentation from Spins on the actual orders and, like, the logic for calculations, yes. So, I’ll add that. We have this shared, like, Spins API document that’s, like, we’re kind of using for the pipeline. I’ll add it to that, and then have a conversation in there.

71 00:07:59.670 00:08:00.260 Demilade Agboola: Okay.

72 00:08:00.470 00:08:01.000 Cinnamon Toast: Shut up.

73 00:08:01.000 00:08:06.140 Demilade Agboola: That’ll be great, so at least, prior to the meeting, we have the… like, we can do that early next week, so Tuesday works.

74 00:08:06.260 00:08:10.189 Demilade Agboola: Once we can get some time, and we can just, like.

75 00:08:10.600 00:08:18.190 Demilade Agboola: Hop into the meeting, we have some context on what data’s available, what the metrics, like, what the definitions of each of these terms are.

76 00:08:18.340 00:08:27.530 Demilade Agboola: And while Heather has been able to, like, guide us into how we should go about it, we’re also aware of what data is available and how that would fit into the definition that we need.

77 00:08:27.690 00:08:28.260 Cinnamon Toast: Yep.

78 00:08:28.880 00:08:30.530 Demilade Agboola: Or the…

79 00:08:31.080 00:08:38.870 Speaker 3 (Cinnamon Toast): Relative date comparisons, are those increments that Ashrini mentioned, the 4 weeks, 12 weeks, 26 weeks, 52 weeks, are those…

80 00:08:39.280 00:08:44.049 Speaker 3 (Cinnamon Toast): The right relative timeframes, and do we need all.

81 00:08:44.051 00:08:44.901 Cinnamon Toast: of those.

82 00:08:45.750 00:08:49.949 Speaker 1 (Cinnamon Toast): We’re… the plan is to pull everything out at a one week.

83 00:08:50.710 00:08:55.040 Speaker 1 (Cinnamon Toast): at, like, the lowest level, one week, and then we’re gonna aggregate on our side, and just…

84 00:08:55.045 00:08:57.384 Speaker 3 (Cinnamon Toast): And we build the… okay, gotcha.

85 00:08:57.390 00:09:02.199 Speaker 1 (Cinnamon Toast): Cool. So yeah, so the trick is, like, kind of aggregating out those different fields.

86 00:09:02.290 00:09:08.770 Speaker 1 (Cinnamon Toast): But yeah, I think to your point, like, maybe action item there is honestly, like, an audit of both what’s

87 00:09:08.800 00:09:26.179 Speaker 1 (Cinnamon Toast): currently in warehouse from, like, the pipeline you’ve built so far. Just, like, triple checking, these are the right fields, do we want to expand on those fields, do we, like, do we hit everything, or do we miss them? JT and I can take that together. Number two is identifying, like, of those fields, which ones do we foresee risk, or…

88 00:09:26.180 00:09:29.080 Speaker 1 (Cinnamon Toast): Unknowns for aggregation.

89 00:09:29.080 00:09:38.010 Speaker 2 (Cinnamon Toast): Yeah, so maybe we set up 30 minutes tomorrow. We’ll share the documentation with you, and we’ll try to meet with Heather next week, where it’s like…

90 00:09:38.400 00:09:47.919 Speaker 2 (Cinnamon Toast): we’re asking Heather, what do we do with these four fields? So we know, like, these are uncertain fields. So just calling out, yeah, the… we might need to…

91 00:09:48.100 00:09:53.349 Speaker 2 (Cinnamon Toast): We’re doing granular pulls. I have a fear we’re gonna have to do a brand-level pull as well.

92 00:09:53.760 00:10:00.659 Demilade Agboola: Okay, alright. Sure. Sounds good. So, yeah, we look forward to the outcome of tomorrow’s call.

93 00:10:00.990 00:10:06.979 Demilade Agboola: And yeah, we look forward to also meeting her early next week, so that’ll play a role into moving forward with this.

94 00:10:07.600 00:10:08.310 Cinnamon Toast: Awful.

95 00:10:08.570 00:10:09.810 Demilade Agboola: Alright, sounds good.

96 00:10:10.620 00:10:19.059 Demilade Agboola: In terms of, like, the long-running models, so what I’m currently trying to do is replace, like, the logic within the tables with

97 00:10:19.320 00:10:21.680 Demilade Agboola: Optimize logic.

98 00:10:23.490 00:10:35.779 Demilade Agboola: That being said, because I’m doing it directly in Redshift, because I don’t have, like, dbt write access, so I’m having to, like, manually compile the code, because dbt compile just allows you to compile.

99 00:10:35.940 00:10:39.650 Demilade Agboola: According to, you know, sickle-ready form.

100 00:10:39.950 00:10:41.950 Demilade Agboola: And you can just paste in, like, redshift.

101 00:10:42.310 00:10:46.360 Demilade Agboola: I have to manually do that, and also, because of how, like.

102 00:10:46.910 00:10:51.840 Demilade Agboola: I’m just writing a query against Redshift. I don’t necessarily… I can’t test, like.

103 00:10:52.010 00:11:03.760 Demilade Agboola: increments, like, the incremental function and how to build incrementally on, to optimize that. So it does make it a bit slower as… and not efficient to do it this way.

104 00:11:03.910 00:11:11.889 Demilade Agboola: That being said, I have been able to try to see what can be done in terms of, like, maybe moving CTEs around.

105 00:11:12.010 00:11:19.010 Demilade Agboola: Or, like, moving, like, the filtering logic higher up in different parts of it. I would share some of that, like,

106 00:11:19.510 00:11:20.480 Demilade Agboola: Later.

107 00:11:20.780 00:11:24.800 Demilade Agboola: once I’m, like, once I’m fully put my thoughts together around that.

108 00:11:25.060 00:11:28.510 Demilade Agboola: But yeah, so that’s kind of what I’m doing in that regard.

109 00:11:31.670 00:11:32.330 Demilade Agboola: Okay, so before.

110 00:11:32.330 00:11:36.030 Speaker 3 (Cinnamon Toast): And then we just found out, just in regards to the access, we.

111 00:11:36.030 00:11:36.780 Speaker 4 (Cinnamon Toast): do not have.

112 00:11:36.780 00:11:41.279 Speaker 3 (Cinnamon Toast): dbt admin access yet, we need someone to come back from vacation for that, but.

113 00:11:41.280 00:11:51.160 Speaker 4 (Cinnamon Toast): We, our accounts should have admin access for both Prefect and AWS. I don’t know if you’ve been able to log into those pieces yet, but… I just validated that.

114 00:11:51.160 00:11:56.490 Speaker 1 (Cinnamon Toast): validated that after this meeting. Okay. Should be good to go there. And for the dbt write access.

115 00:11:56.770 00:11:57.250 Demilade Agboola: Yeah.

116 00:11:58.990 00:12:05.669 Speaker 1 (Cinnamon Toast): You should… I think you should be able to with IT provisions, right? Like… Create a branch? Can you not? Yeah, can you branch?

117 00:12:05.675 00:12:06.544 Speaker 2 (Cinnamon Toast): Yeah, sure.

118 00:12:06.550 00:12:13.719 Speaker 1 (Cinnamon Toast): We can go to the billing, so I feel like that was part of the… Yeah, can you just walk us through that? I want to make sure we set you up.

119 00:12:14.660 00:12:16.940 Demilade Agboola: Give me one second… And…

120 00:12:17.490 00:12:19.339 Speaker 5 (Cinnamon Toast): sort of thing where I can use to hear yourself.

121 00:12:19.340 00:12:20.229 Demilade Agboola: So, the way.

122 00:12:20.230 00:12:21.870 Speaker 5 (Cinnamon Toast): I think billing was the top one.

123 00:12:22.160 00:12:24.660 Demilade Agboola: It works is,

124 00:12:28.130 00:12:33.089 Demilade Agboola: with… when you have right access, there’s some… you’ll see a plot… a…

125 00:12:33.420 00:12:41.379 Demilade Agboola: a tab called Studio, like, so basically you can click on that, and that allows you to write, like, literally write code and push through dbt Cloud.

126 00:12:41.380 00:12:42.210 Speaker 5 (Cinnamon Toast): June.

127 00:12:43.670 00:12:47.379 Speaker 2 (Cinnamon Toast): Oh, I guess… do you not have dbt installed locally on your computer?

128 00:12:47.380 00:12:56.629 Demilade Agboola: So, I do, I can also do it locally, but the problem will be, like, testing. I still need to be able to test against the environment.

129 00:12:56.940 00:13:00.579 Demilade Agboola: And push it, and then go, like, okay, so this is it.

130 00:13:00.810 00:13:07.510 Demilade Agboola: So, to be able to test things like incremental runs, I need to be able to run it on my local branch.

131 00:13:07.720 00:13:11.960 Demilade Agboola: And I can’t do a local branch run without, like, dbt.

132 00:13:12.880 00:13:18.150 Demilade Agboola: like, without… yeah, like, I can’t do a local branch run without either having dbt Cloud.

133 00:13:18.350 00:13:29.260 Demilade Agboola: on here, where I then run, like, do a dbt run all through, or on my local device, like on my VS Code, where I then say, you know, dbt run, select this node.

134 00:13:29.420 00:13:34.600 Demilade Agboola: or, you know, select, DB to run this, and I can see the incremental runs.

135 00:13:34.790 00:13:41.490 Demilade Agboola: So that’s why I need the dbt write access, or the… because it… I can see the code, I can do stuff around the code.

136 00:13:41.710 00:13:46.290 Demilade Agboola: But, like, dbt-specific functions do not work, which is kind of the issue.

137 00:13:48.150 00:13:51.390 Speaker 5 (Cinnamon Toast): how you guys have worked with Orchard in the past, because I know you’ve done some DVDs.

138 00:13:51.390 00:13:57.620 Speaker 1 (Cinnamon Toast): Like, and which dbt functions, I guess, are you, like…

139 00:13:57.940 00:14:11.279 Demilade Agboola: Yeah, it would just be, like, the dbt build, so I can see the incremental increase, it’ll be, like, things, like, around that, not being too crazy. It’s just, I just need to be able to use, like, dbt-specific functions, but that requires, like.

140 00:14:11.880 00:14:16.859 Demilade Agboola: dbt to… I need that right access to be able to do it.

141 00:14:17.610 00:14:19.180 Cinnamon Toast: Really diagnostic, so…

142 00:14:19.650 00:14:20.899 Speaker 1 (Cinnamon Toast): Oh, back then.

143 00:14:21.220 00:14:35.400 Speaker 1 (Cinnamon Toast): So this is, like, specifically, like, building the project, which you need dbt, right, like, CloudX as far as, like, how you actually run, like, diagnostics on the full dbt project, not just, like, oh, did I successfully run this?

144 00:14:36.140 00:14:36.780 Demilade Agboola: Yeah, so…

145 00:14:36.780 00:14:37.440 Speaker 1 (Cinnamon Toast): died.

146 00:14:37.750 00:14:39.480 Demilade Agboola: Yeah, it’ll be, like…

147 00:14:40.360 00:14:46.900 Demilade Agboola: dbt currently metarizes things as tables, for instance, right? If I change it to, say, an incremental model.

148 00:14:47.380 00:14:50.140 Demilade Agboola: One, I would like to test.

149 00:14:50.310 00:14:58.909 Demilade Agboola: how long the incremental updates now take, so I can say, okay, it used to take… it used to be a 40-minute run every day, now it’s,

150 00:14:59.060 00:15:05.029 Demilade Agboola: After the initial refresh, now it’s a 3-minute run, or 4-minute run every day, for instance.

151 00:15:05.210 00:15:09.159 Demilade Agboola: That testing, that, like, dbt build.

152 00:15:10.500 00:15:15.680 Demilade Agboola: I would need to have, like, local tests on it to be able to be sure, like, what those numbers are.

153 00:15:15.790 00:15:31.699 Demilade Agboola: And, like, ultimately, yes, I can send a PR and just change, like, incremental runs, but without necessarily verifying that these things are running locally, I’m potentially just pushing, like, really bad code or, like, any issues into production, so I obviously wouldn’t want to do that.

154 00:15:31.700 00:15:44.479 Demilade Agboola: So I want to be able to test locally, see the results, be sure everything works, be sure there are no, like, weird edge cases, and then I can… I would then feel confident in, like, sending, like, a PR in.

155 00:15:45.040 00:15:46.910 Cinnamon Toast: So this is dbt Insights.

156 00:15:48.160 00:15:54.739 Speaker 2 (Cinnamon Toast): That’s what you need? Well, I guess one question that would help, I think, actually give me an understanding is, like, let’s say you wanted to build

157 00:15:54.840 00:16:08.620 Speaker 2 (Cinnamon Toast): model that you just wrote, you can still use dbt run, you just can’t comp it apples to apples with the existing runtimes. Like, on your local device, you should be able to use, say, dbt run, and you should be able to run anything.

158 00:16:08.880 00:16:15.330 Demilade Agboola: No, I can’t do a dbt run because I don’t have access to the dbt, like, I don’t have credentials to trigger a dbt run.

159 00:16:16.510 00:16:17.100 Cinnamon Toast: That’s funny.

160 00:16:17.100 00:16:18.900 Demilade Agboola: to trigger a dbt run.

161 00:16:18.900 00:16:23.959 Ashwini Sharma: No, Demi, we can do that, we can do that. I mean, we can run locally dbt.

162 00:16:24.960 00:16:27.960 Ashwini Sharma: Okay. And it’s going to write in the redshift.

163 00:16:28.730 00:16:29.290 Speaker 2 (Cinnamon Toast): Yeah.

164 00:16:31.020 00:16:33.300 Speaker 5 (Cinnamon Toast): Are you looking for the insights?

165 00:16:34.470 00:16:35.399 Cinnamon Toast: No, no.

166 00:16:35.400 00:16:40.780 Demilade Agboola: It’s not… it’s not inside, so catalog, the catalog is not working. Like, that… that’s still there.

167 00:16:41.230 00:16:46.300 Speaker 2 (Cinnamon Toast): Yeah, although we might… If you have dbt locally and the code, you should be able to run.

168 00:16:46.780 00:16:49.030 Demilade Agboola: And warehouse permissions. Yeah, and…

169 00:16:49.030 00:17:02.120 Speaker 1 (Cinnamon Toast): to your point, I think just to walk you through of what our workflow of, like, code development has been, is… yeah, you can see, like, these are the runs, or running, like, business… like, these kind of, like, large projects.

170 00:17:02.380 00:17:17.449 Speaker 1 (Cinnamon Toast): Our flow is, like, usually, like, Orchard is managing, like, the dbt repo. When JTC or I, like, want to write dbt code, we’ll run it locally to ensure the table built. We don’t do a lot of, like, performance

171 00:17:17.550 00:17:26.830 Speaker 1 (Cinnamon Toast): like, metrics, we just kind of, like, make sure, oh, did this table build, we QA it, and then we, like, submit a PR, merge with Maine once it’s approved.

172 00:17:26.829 00:17:38.519 Speaker 1 (Cinnamon Toast): we’re not doing, like, the in-cloud performance and, like, model versus model analysis, and I think… is that where you’re trying to go to, where you’re saying, like, I can run the model locally, but it’s not apples to apples?

173 00:17:38.550 00:17:44.609 Speaker 1 (Cinnamon Toast): Versus what’s, like, the hours that the, like, cloud is reporting, which is, like, 4 hours per run, or whatever it is.

174 00:17:45.140 00:17:51.070 Demilade Agboola: Yeah, so, like, doing that won’t necessarily… like, if I run it locally,

175 00:17:51.250 00:18:03.289 Demilade Agboola: I’ll try what… I’ll try asking this method and see, like, if that works, and I’ll give you feedback. But basically, what I’ve been doing so far, I have not been necessarily being able to run, like, dbt locally against…

176 00:18:03.640 00:18:06.640 Demilade Agboola: The instance.

177 00:18:06.800 00:18:19.949 Demilade Agboola: So, without being able to do that, that’s where my issues have been, because then I’m having to, like, manually, like, do what dbt does behind the scenes to be able to compare. So I’ll check…

178 00:18:19.950 00:18:36.789 Speaker 2 (Cinnamon Toast): Yeah, I think you should check what Ashini does. It will still be annoying where, like, you won’t be able to run it versus the prod, but you could still run, like, just, like, it’d be a lot easier where you could run your local model, revert changes, run that model, like, you could just run those two locally, and then compare your local times.

179 00:18:36.790 00:18:38.439 Demilade Agboola: Okay, sure, I’ll do that.

180 00:18:38.440 00:18:39.469 Speaker 2 (Cinnamon Toast): It’ll be simpler.

181 00:18:40.210 00:18:43.719 Demilade Agboola: Yeah, sure, I’ll do that, get back to all of us on that.

182 00:18:43.720 00:18:48.499 Ashwini Sharma: And is the source data the same? Because, like, the dbt cloud

183 00:18:48.990 00:18:55.980 Ashwini Sharma: Probably is running in a production environment, and the warehouse access that you have given is to a dev warehouse.

184 00:18:56.270 00:18:57.810 Ashwini Sharma: I just want to make sure that.

185 00:18:57.810 00:19:04.900 Speaker 2 (Cinnamon Toast): It’s both… it’s only one warehouse. It’s just, there’s a schema for our production. We… we just.

186 00:19:04.900 00:19:05.270 Ashwini Sharma: -Oh.

187 00:19:05.270 00:19:10.109 Speaker 2 (Cinnamon Toast): model dot, and that’s production. It’s a production schema, not a warehouse. So you have access to the exact same things.

188 00:19:10.320 00:19:10.980 Ashwini Sharma: Oh, okay, okay.

189 00:19:10.980 00:19:11.540 Demilade Agboola: Okay, alright.

190 00:19:11.540 00:19:11.880 Ashwini Sharma: Right now.

191 00:19:12.160 00:19:12.730 Demilade Agboola: Oh, no.

192 00:19:12.730 00:19:31.010 Speaker 1 (Cinnamon Toast): All that’s configured in the profiles.yml file. I have it shared in the drive. Maybe we didn’t close that loop earlier on, but you can use, like, configure your dev schema in profiles.yml, and then you should be able to run and test and dump to a, just a… A dev schema. A dev schema.

193 00:19:31.010 00:19:38.129 Demilade Agboola: Okay, gotcha, gotcha. So I’ll look at that, and try and test that. Once I… once that’s up and running, yeah, I will let you know.

194 00:19:38.300 00:19:43.309 Demilade Agboola: I would then do, like, the dbt optimizations against the, data itself.

195 00:19:44.190 00:19:54.140 Speaker 1 (Cinnamon Toast): Cool. Yeah, let me know if you want to walk through it after this meeting or something, too. Just happy to help where we can. I know it’s a little confusing, it’s not like…

196 00:19:54.480 00:19:55.940 Speaker 1 (Cinnamon Toast): Perfect. Dave.

197 00:19:56.700 00:19:57.690 Speaker 1 (Cinnamon Toast): That’s…

198 00:19:58.260 00:19:59.390 Demilade Agboola: That’s fair.

199 00:19:59.910 00:20:06.040 Demilade Agboola: So I think, like, basically for us, the next steps would just be, I mean, we’re halfway through January, time’s flying.

200 00:20:06.560 00:20:10.180 Demilade Agboola: So it’s just, like, scoping for February in terms of, like, partner transition.

201 00:20:10.310 00:20:13.389 Demilade Agboola: So what, what does that look like? What do we need?

202 00:20:13.690 00:20:17.639 Demilade Agboola: And how can we, like, make that as smooth as possible?

203 00:20:18.030 00:20:24.920 Demilade Agboola: And so that would be the first step, and then just generally, like, you know, how, like, getting the data mapped.

204 00:20:25.610 00:20:26.290 Demilade Agboola: kicking off.

205 00:20:26.290 00:20:26.949 Cinnamon Toast: Yeah, actually.

206 00:20:27.830 00:20:31.300 Speaker 2 (Cinnamon Toast): I’m gonna share my screen, if that’s cool. I can actually talk to the second point.

207 00:20:31.870 00:20:33.050 Demilade Agboola: That’s all good.

208 00:20:37.710 00:20:38.680 Speaker 2 (Cinnamon Toast): So…

209 00:20:38.820 00:20:45.030 Speaker 2 (Cinnamon Toast): Right now, I… let me… let me hide these two. These are my working notes. You guys don’t need to know this.

210 00:20:45.330 00:20:46.120 Speaker 2 (Cinnamon Toast): Hi.

211 00:20:46.280 00:20:50.629 Speaker 2 (Cinnamon Toast): Yeah, thank you. Okay, so this is going to be,

212 00:20:51.070 00:20:58.509 Speaker 2 (Cinnamon Toast): the final document, which… this is a G-sheet, which will kind of tell you how to build the final MNN data mart.

213 00:20:58.730 00:21:01.909 Speaker 2 (Cinnamon Toast): Again, the data will be aggregated at a weekly level.

214 00:21:02.180 00:21:08.380 Speaker 2 (Cinnamon Toast): This is all the spend revenue, and so we have the different metrics called out here.

215 00:21:08.870 00:21:19.440 Speaker 2 (Cinnamon Toast): I plan on having this… this column, E, will be the data marked, and so you can see for the ones that are already in the data warehouse, I have the marked specifically you’re gonna pull from.

216 00:21:19.730 00:21:24.330 Speaker 2 (Cinnamon Toast): And then in the data notes, I have the filters you would use.

217 00:21:24.730 00:21:28.780 Speaker 2 (Cinnamon Toast): Or, like, things that I think you would need to know about the table in order to pull from it.

218 00:21:30.320 00:21:38.099 Speaker 2 (Cinnamon Toast): And so, like, here’s examples, like, for… to get Instacart, search Costco, which is gonna be one column, we want to look at the past 3 years.

219 00:21:38.210 00:21:49.449 Speaker 2 (Cinnamon Toast): of this data, you’re probably gonna have to write a CTE that you pull from this… this mark, where you filter for the retailers Instacart, and the campaign retailer is not Instacart… or no, this one is Instacart Costco, sorry.

220 00:21:50.200 00:21:52.650 Speaker 2 (Cinnamon Toast): Since we’re doing it at Costco, and so you would, like.

221 00:21:53.040 00:21:59.560 Speaker 2 (Cinnamon Toast): This would be kind of, like, the path I would deliver you, so that you guys can then go build us the final MMM Mart.

222 00:22:00.420 00:22:05.100 Speaker 2 (Cinnamon Toast): Is this enough information that you feel like you guys would be able to take this and go?

223 00:22:06.560 00:22:14.969 Demilade Agboola: Yes, I think so. It looks pretty detailed, especially because for each perspective column on metric, we’re able to see the definition.

224 00:22:15.670 00:22:16.240 Cinnamon Toast: Yeah.

225 00:22:16.630 00:22:22.990 Ashwini Sharma: Is this MMM Mart 1 table that you’re expecting, or is it… A set of tables containing…

226 00:22:22.990 00:22:27.769 Speaker 2 (Cinnamon Toast): It’s going to be… One table, one column is going to be weak.

227 00:22:28.000 00:22:44.620 Speaker 2 (Cinnamon Toast): which is going to be used, the same definition as we define week in Spins data, is the week we’re going to use here. So that’s going to be the definition of what week it is. Is it, like, a Monday to Monday week, or Monday to Sunday week, whatever it is. Whatever spins does, that’s the level of definition.

228 00:22:44.890 00:22:48.030 Speaker 2 (Cinnamon Toast): And then now… then the weak numbers for the past 3 years will be…

229 00:22:48.260 00:22:53.049 Speaker 2 (Cinnamon Toast): The… that one column, then each of these rows will be a new column.

230 00:22:53.750 00:22:56.539 Speaker 2 (Cinnamon Toast): So, we’ll have one row that will be TDP.

231 00:22:56.580 00:23:15.550 Speaker 2 (Cinnamon Toast): which is a SPINS data set, so we’re, like, Ashwini’s working on getting us to there, and then we’ll eventually have a mart, which the Milate will then pull from that mart into this final MMM Mart. The goal of the MMM Mart is going to be, this is the output, which we package together, drop over to a partner.

232 00:23:16.510 00:23:19.779 Speaker 2 (Cinnamon Toast): So each of these rows will be one column.

233 00:23:20.930 00:23:24.509 Speaker 2 (Cinnamon Toast): Running over the exact same date range, aggregated by week.

234 00:23:25.600 00:23:27.490 Demilade Agboola: Okay. That is…

235 00:23:28.070 00:23:32.609 Demilade Agboola: And for the things that we need to set up, because I can see some things are not currently being tracked.

236 00:23:32.820 00:23:35.450 Demilade Agboola: How, like, are we responsible for setting that up?

237 00:23:36.290 00:23:39.900 Speaker 2 (Cinnamon Toast): So, you guys are not… you guys are responsible for the spin section.

238 00:23:40.050 00:23:40.530 Demilade Agboola: Okay.

239 00:23:40.530 00:23:59.550 Speaker 2 (Cinnamon Toast): we’re pulling together everything else. So likely… so, like, this is an example, programmatic OLV, where we might have 1, 2, or 3 marts, not really marts, they’re just gonna be static tables, which we upload CSVs in the warehouse, and you’re gonna aggregate them, pull the columns, and then put them in this final mart.

240 00:24:00.050 00:24:03.219 Speaker 2 (Cinnamon Toast): So, everything will exist in the warehouse.

241 00:24:04.110 00:24:18.850 Speaker 2 (Cinnamon Toast): Like, the red ones are basically go-gets on my end, so we’re not done getting everything. I wanted to share the format, and the rest of this stuff is, like, I think by, I want to share the format to get feedback. Next Tuesday, we’re having a meeting to finalize these.

242 00:24:19.330 00:24:25.480 Speaker 2 (Cinnamon Toast): And at that point, I can Slack you guys and kind of release you, like, hey, everything that’s not red, you guys can go after.

243 00:24:26.050 00:24:32.599 Demilade Agboola: Alright, sounds good. Also, just so we’re clear and just on the same page, the level of granularity is the Spins API weekly level, basically.

244 00:24:32.920 00:24:40.140 Speaker 2 (Cinnamon Toast): Yes, we basically… we need to match spins. And so then, I’m gonna do another thing that I haven’t added in the data notes here.

245 00:24:40.290 00:24:43.810 Speaker 2 (Cinnamon Toast): It’s gonna be a lot simpler. This is the retailer revenue that we’re gonna add up.

246 00:24:44.000 00:24:51.000 Speaker 2 (Cinnamon Toast): One column is going to be retailer revenue, and it’s going to include all of these different retailers.

247 00:24:51.620 00:24:58.109 Speaker 2 (Cinnamon Toast): that I’ve, like, said yes to. They’re basically all Spins data, so I was waiting to see that data first.

248 00:24:58.370 00:25:01.360 Demilade Agboola: Because I’m making assumptions on what exists right now.

249 00:25:01.540 00:25:09.819 Speaker 2 (Cinnamon Toast): So once I, like, get a good idea of what spins data exists, I’ll have the mart that you need to pull from to calculate one revenue number.

250 00:25:10.680 00:25:16.729 Demilade Agboola: Okay. And also, part of what I also wanted to ask is, are there any, like, granularity mismatches? Like.

251 00:25:17.350 00:25:19.690 Demilade Agboola: Are we already getting, like, weekly…

252 00:25:19.850 00:25:23.539 Demilade Agboola: data that might not necessarily match the weekly data from Spins.

253 00:25:25.990 00:25:27.989 Demilade Agboola: Or do we have the flexibility to…

254 00:25:27.990 00:25:28.780 Speaker 2 (Cinnamon Toast): both,

255 00:25:28.780 00:25:30.410 Cinnamon Toast: Everything here is daily.

256 00:25:30.410 00:25:32.059 Demilade Agboola: Okay, that’s great then.

257 00:25:34.660 00:25:37.420 Speaker 2 (Cinnamon Toast): For the on-site, off-site, we don’t know, like…

258 00:25:37.530 00:25:40.899 Speaker 2 (Cinnamon Toast): Yeah, we have, like, some questions on, like, these three.

259 00:25:41.010 00:25:44.699 Speaker 2 (Cinnamon Toast): fields, but everything else is daily. Like, a majority of it’s daily.

260 00:25:44.800 00:25:45.490 Demilade Agboola: Alright.

261 00:25:45.590 00:25:52.779 Speaker 2 (Cinnamon Toast): you’re gonna have to aggregate to the correct… like, I don’t know what week spins defines as, but match that, you’re good. Everything else is daily.

262 00:25:53.500 00:25:55.869 Demilade Agboola: Okay, I think Ashwin has something to say.

263 00:25:56.070 00:26:03.440 Ashwini Sharma: Yeah, Spins had some data where, I think some brand, brand was null.

264 00:26:04.020 00:26:08.340 Ashwini Sharma: So, not sure if that is going to create any problems.

265 00:26:09.040 00:26:13.070 Cinnamon Toast: or… Which… So.

266 00:26:13.070 00:26:13.860 Ashwini Sharma: Spinsa’s…

267 00:26:13.860 00:26:14.670 Cinnamon Toast: sleeping problem.

268 00:26:15.110 00:26:24.950 Ashwini Sharma: I’m not able to recall, right? If you, if you, let me quickly run a query against that table. I’ve shared the table names, right? If you just look into the raw table.

269 00:26:25.100 00:26:27.260 Ashwini Sharma: You might see some of those.

270 00:26:29.880 00:26:36.009 Speaker 1 (Cinnamon Toast): Brand might return null for some subcategory roll-ups, so if you’re looking

271 00:26:36.470 00:26:51.650 Speaker 1 (Cinnamon Toast): like, it all, like, I think what the query will kick out is, like, oh, subcategory of cereal, all brand roll up, and brand will be null when it’s… it should be, like, a very high dollar amount, because it’s all brands that sold in that subcategory.

272 00:26:51.650 00:26:58.200 Speaker 1 (Cinnamon Toast): It’s, like, probably multiple levels of aggregation in the same table, so we can… we can handle that in model.

273 00:26:58.470 00:26:59.949 Speaker 1 (Cinnamon Toast): We’re just succeed in.

274 00:26:59.950 00:27:00.950 Cinnamon Toast: Yeah, exactly.

275 00:27:02.180 00:27:04.939 Cinnamon Toast: Like, split that out to a different part, basically.

276 00:27:05.560 00:27:06.430 Cinnamon Toast: Pitable.

277 00:27:07.000 00:27:07.590 Ashwini Sharma: Okay.

278 00:27:07.960 00:27:12.730 Speaker 1 (Cinnamon Toast): Yeah, if you… yeah, if you have a query, though, where you’re seeing that, like, let’s…

279 00:27:13.300 00:27:19.439 Speaker 1 (Cinnamon Toast): Yeah, let’s just, like, comment that query. Yeah, exactly. Anytime you see something like that, we’re, like, happy to take a look.

280 00:27:20.170 00:27:23.649 Ashwini Sharma: Sure, I’ll, I’ll send, send the details.

281 00:27:28.340 00:27:30.399 Speaker 2 (Cinnamon Toast): So, I can actually probably… that’s…

282 00:27:30.400 00:27:30.930 Ashwini Sharma: Yeah.

283 00:27:30.930 00:27:34.920 Speaker 2 (Cinnamon Toast): Marketing Insights one week will tell me all the brands that exist, or all the…

284 00:27:34.920 00:27:35.760 Ashwini Sharma: Right.

285 00:27:35.760 00:27:37.749 Speaker 6 (Cinnamon Toast): Retailers that I guess, correct?

286 00:27:37.750 00:27:39.290 Ashwini Sharma: I think he has the query here.

287 00:27:39.290 00:27:43.179 Speaker 6 (Cinnamon Toast): So, retail regions, a little bit lower.

288 00:27:43.630 00:27:44.830 Speaker 6 (Cinnamon Toast): Okay.

289 00:27:44.990 00:27:46.589 Speaker 6 (Cinnamon Toast): More detailed is better. Okay.

290 00:27:47.400 00:27:49.179 Ashwini Sharma: Yeah, dropping it in the Slack.

291 00:27:49.770 00:27:50.510 Cinnamon Toast: Thanks.

292 00:27:50.810 00:27:51.910 Cinnamon Toast: Take a look at that.

293 00:27:52.850 00:27:56.390 Cinnamon Toast: Okay, that’s cool, though.

294 00:27:56.920 00:28:07.559 Speaker 2 (Cinnamon Toast): So anyway, but summary on that, the Gsheet, I know it’s not fully filled out, in terms of, like, we’re missing pieces and we’re gathering those, but that structure works for you, correct?

295 00:28:07.680 00:28:11.879 Speaker 2 (Cinnamon Toast): Whereas, I’ve got the mark… my notes on, kind of, what filters to use.

296 00:28:12.250 00:28:16.950 Speaker 2 (Cinnamon Toast): And you’ll kind of figure out, like, okay, the date, just Slack additional questions, but…

297 00:28:17.260 00:28:20.739 Speaker 2 (Cinnamon Toast): It should be pretty simple. Probably a bunch of CTs.

298 00:28:20.990 00:28:28.889 Demilade Agboola: Yeah, I think that puts us in a good position to be able to attack the data mart and build that out, because at least gives us direction of the source.

299 00:28:29.000 00:28:34.439 Demilade Agboola: The granularity and the definition of the metric, so I think… I think we’re in a good spot with that.

300 00:28:37.060 00:28:53.990 Speaker 1 (Cinnamon Toast): So that’ll be… that’ll be the North Star for the MMM, like, data products, and I think for the pipeline, we’re obviously going to pull in a lot more fields and, like, levels of aggregation, so we’ll continue to work, Ashwini, in that kind of more detailed filtering doc, that we’ve been working in.

301 00:28:54.390 00:28:55.060 Speaker 1 (Cinnamon Toast): Yeah.

302 00:28:55.065 00:28:58.534 Speaker 2 (Cinnamon Toast): I’m only pulling a small… I’m pulling, like, brand-level metrics.

303 00:28:58.745 00:29:03.704 Speaker 2 (Cinnamon Toast): for a 5-week for 4 specific SPINS metrics.

304 00:29:03.845 00:29:12.034 Speaker 2 (Cinnamon Toast): And then revenue. So it’s like, I’ve got 5 metrics I’m playing with, and Michael’s gonna be pulling a lot more. So, I’m kind of a side product of what you guys are working on.

305 00:29:14.200 00:29:21.880 Demilade Agboola: Okay, sounds good. Does… just general questions, does anyone have any, like, feedback on…

306 00:29:22.010 00:29:24.319 Demilade Agboola: This week, and just, like, getting stuff done.

307 00:29:27.430 00:29:44.040 Speaker 1 (Cinnamon Toast): Good progress. Like, honestly, like, thanks for being dynamic and kind of moving between some priorities as we get more details. I think, yeah, we’re learning a lot internally about, like, what the endpoint target is, and, like, what’s available versus what’s not, so thanks for the flexibility.

308 00:29:45.005 00:29:49.154 Speaker 2 (Cinnamon Toast): And keep us updated, like, Slack us about if you’re able to run dbt locally.

309 00:29:49.160 00:29:49.690 Demilade Agboola: Okay, well…

310 00:29:49.690 00:29:53.420 Speaker 2 (Cinnamon Toast): Because that’s definitely a huge barrier. Didn’t realize you weren’t able to do it locally at all.

311 00:29:55.830 00:29:58.500 Speaker 2 (Cinnamon Toast): Will do, will do. It’s a barrier. Yeah, and then I think.

312 00:29:58.505 00:30:14.844 Speaker 4 (Cinnamon Toast): I think for, I know just the direction you guys are waiting on from our end on the partner transition is going to be a lot of these, access and knowledge transfer pieces that we’re… we’ll be working to… to make sure you guys have everything you need, and that we can help provide the context there.

313 00:30:15.140 00:30:24.679 Demilade Agboola: Okay, sounds good. We’ll be looking forward to that. Like I said, we’re already halfway through January, so these last two weeks will be very important, just to, like, be ready for February.

314 00:30:24.850 00:30:26.950 Demilade Agboola: So yeah, we’ll be looking forward to that.

315 00:30:27.160 00:30:37.860 Demilade Agboola: Also just… this is just a general question, do you find the daily updates, like, helpful? Just to, like, see, get an idea of what’s going on, and potentially how you can jump in?

316 00:30:41.110 00:30:44.450 Speaker 3 (Cinnamon Toast): I’ll probably defer to you guys as I feel like they’ve been more MMM-based, but…

317 00:30:45.790 00:30:56.950 Speaker 1 (Cinnamon Toast): Yeah, I think at this stage, like, more information is better. Would love to know, kind of like, it doesn’t need to be, like, formal in any capacity, it could just be, like, a bullet point, like, hey, this is what I’m focusing on today, this is…

318 00:30:57.020 00:31:10.139 Speaker 1 (Cinnamon Toast): confusing, and this is where I have questions. Happy to keep it fairly casual, because when it comes to, like, the dbt stuff, it’s like, oh, we can set up a 10-minute call and figure out how to get around that, or, like, find a workaround. Yeah.

319 00:31:10.145 00:31:21.035 Speaker 2 (Cinnamon Toast): more focused on blockers than updates, I would say. Unless there are questions. Yeah, questions. Blockers, questions. Yeah. Yeah. But, like, then just…

320 00:31:21.265 00:31:24.174 Speaker 2 (Cinnamon Toast): generally what you did, I think it’s more…

321 00:31:24.465 00:31:28.255 Speaker 2 (Cinnamon Toast): We’ll get that at more of a, like, a weekly level, that’s fine.

322 00:31:28.260 00:31:29.310 Demilade Agboola: Okay.

323 00:31:29.310 00:31:30.940 Speaker 2 (Cinnamon Toast): What can we help on a day-to-day?

324 00:31:30.940 00:31:36.740 Demilade Agboola: And if… and if there are no, like, blockers or questions, you’re fine with no updates for that day, and just…

325 00:31:37.960 00:31:40.949 Speaker 4 (Cinnamon Toast): Yeah. Yeah, I think if we’re… if we have questions, we’ll reach out to…

326 00:31:40.950 00:31:42.000 Cinnamon Toast: Yeah.

327 00:31:42.490 00:31:47.590 Demilade Agboola: Alright, sounds good, sounds good. I just don’t want it to be too spammy, if there’s stuff that you’ll… you care to.

328 00:31:47.590 00:31:48.549 Cinnamon Toast: Yeah, no worries.

329 00:31:48.550 00:31:51.970 Speaker 4 (Cinnamon Toast): Alright, sounds good. Appreciate the thoughtfulness there.

330 00:31:52.300 00:31:59.479 Demilade Agboola: Okay then, so we have a meeting scheduled for next week, I believe, so we will definitely have more updates.

331 00:31:59.830 00:32:03.500 Demilade Agboola: for next… I think it’s next week, Friday, so we should have more updates in that case.

332 00:32:04.160 00:32:13.159 Demilade Agboola: We look forward to, like, calls with Heather, and just being able to push, setting up dbt locally for me, and just also,

333 00:32:13.270 00:32:21.680 Demilade Agboola: being able to push some of these, like, plans for next month, and just getting, like, things rolling. Very excited about this project, and just being able to deliver, so…

334 00:32:22.550 00:32:24.250 Cinnamon Toast: Same. Thanks, appreciate it. Thank you both.

335 00:32:24.250 00:32:26.940 Demilade Agboola: Alright then. Thank you, have a good week.

336 00:32:28.360 00:32:28.950 Ashwini Sharma: Thank you.