Meeting Title: Javy Deck Review Date: 2024-12-16 Meeting participants: Nicolas Sucari, Uttam Kumaran, Payas Parab, Robert Tseng


WEBVTT

1 00:03:08.960 00:03:10.050 Nicolas Sucari: Hey? What’s up?

2 00:03:17.690 00:03:18.989 Uttam Kumaran: Hey? Can you hear me?

3 00:03:21.370 00:03:22.080 Nicolas Sucari: Yes.

4 00:03:22.580 00:03:25.800 Uttam Kumaran: Hey? Finally just getting like a break for lunch.

5 00:03:27.000 00:03:27.970 Nicolas Sucari: All night.

6 00:03:28.650 00:03:29.556 Uttam Kumaran: I made

7 00:03:30.310 00:03:34.300 Uttam Kumaran: I made steak over the weekend like skirt steak.

8 00:03:36.000 00:03:41.340 Nicolas Sucari: Oh, yeah, we we call it in Argentina. It’s called Entrana.

9 00:03:41.970 00:03:46.280 Uttam Kumaran: And Tanya. Yeah, I made a bunch of for lunch

10 00:03:47.050 00:03:51.780 Uttam Kumaran: and dinner, and I’ve been eating tons of it so good.

11 00:03:51.780 00:03:52.400 Nicolas Sucari: That’s see.

12 00:03:52.400 00:03:59.510 Uttam Kumaran: Like marinated, and like juice, garlic, parsley, cumin, like all sorts of

13 00:04:00.040 00:04:07.120 Uttam Kumaran: all sorts of stuff marinade for like 2 days. Well, I forgot about it, so marinated for like 2 days.

14 00:04:10.740 00:04:11.160 Nicolas Sucari: Yeah.

15 00:04:11.160 00:04:16.190 Payas Parab: I haven’t had lunch, guys I’m about to. I’m about to have lunch now. I’m like, Oh, man.

16 00:04:16.470 00:04:21.639 Uttam Kumaran: I mean your your lunch is on time. It’s like 2 30. I was like I was about to eat my own hand.

17 00:04:26.950 00:04:29.080 Payas Parab: Oh, how are you guys doing.

18 00:04:29.590 00:04:31.100 Uttam Kumaran: Good dude. How are you?

19 00:04:31.411 00:04:38.569 Payas Parab: Good. Just tired. Had a weekend trip, so I got back at like midnight yesterday and had to like just get started ripping.

20 00:04:38.570 00:04:39.590 Uttam Kumaran: Check again.

21 00:04:40.210 00:04:45.770 Payas Parab: No, not Vegas this time is Chicago. Friends annual like we have like a

22 00:04:45.930 00:04:48.140 Payas Parab: per half year friend’s trip.

23 00:04:49.490 00:04:51.550 Payas Parab: And these are the ones we actually run that

24 00:04:51.970 00:04:58.192 Payas Parab: that data consultancy. So it like doubles, as our like annual planning meeting, and our like friend Annual Meetup.

25 00:04:58.520 00:04:59.573 Uttam Kumaran: Nice nice.

26 00:05:00.100 00:05:00.560 Payas Parab: Yeah.

27 00:05:06.390 00:05:10.350 Payas Parab: what about you guys? Get up to anything? Fun? Relax. What’s.

28 00:05:10.870 00:05:12.449 Uttam Kumaran: Yeah, big relax.

29 00:05:12.880 00:05:13.350 Payas Parab: Nice.

30 00:05:13.350 00:05:15.239 Uttam Kumaran: We literally didn’t do anything.

31 00:05:15.790 00:05:19.219 Uttam Kumaran: Sunday. I worked, I guess. Friday.

32 00:05:19.810 00:05:26.370 Uttam Kumaran: yeah. Friday. Just cooked. So that’s it. Friday. And then Saturday. Put some Christmas lights up.

33 00:05:26.620 00:05:27.110 Payas Parab: I love it.

34 00:05:27.110 00:05:27.490 Uttam Kumaran: Nice.

35 00:05:27.490 00:05:28.299 Nicolas Sucari: Alright! It’s.

36 00:05:28.520 00:05:32.120 Uttam Kumaran: I didn’t finish it, which means it’s at risk of not happening.

37 00:05:33.110 00:05:35.619 Uttam Kumaran: I need to need to go to target and get.

38 00:05:35.750 00:05:40.360 Uttam Kumaran: I. I need some a little bit more chandelier lights that I’m putting up. I’ll finish that up.

39 00:05:40.360 00:05:40.930 Payas Parab: Nice.

40 00:05:41.830 00:05:46.580 Uttam Kumaran: And then, just like went out for drinks with my girlfriend. That’s it.

41 00:05:46.880 00:05:47.810 Payas Parab: Nice.

42 00:05:48.550 00:05:52.459 Nicolas Sucari: Nice. I spent so many, so much time in the pool yesterday.

43 00:05:52.700 00:05:53.457 Uttam Kumaran: At the pool.

44 00:05:54.020 00:05:56.410 Nicolas Sucari: Yeah, it was nice.

45 00:05:56.690 00:06:01.470 Uttam Kumaran: Nice you have like a building pool, or where.

46 00:06:02.640 00:06:21.149 Nicolas Sucari: I have like a building pool. Yeah, my building has like a gym pool. Kind of all of those amenities but yesterday my girlfriend’s parents. They have a house like half half an hour away from here, and they have, like their own pool and that kind of stuff. So it was nice.

47 00:06:22.570 00:06:23.540 Uttam Kumaran: Super nice.

48 00:06:24.500 00:06:32.380 Payas Parab: That’s cool. You actually use that. I feel like I like still to this date, have never used my apartments pool, but I was like when I signed the lease. I was like, Oh, yeah, I have a pool

49 00:06:33.290 00:06:35.369 Payas Parab: like absolutely needed that.

50 00:06:35.780 00:06:36.280 Uttam Kumaran: 1st year out.

51 00:06:37.510 00:06:40.319 Uttam Kumaran: But yeah, now, I don’t have a pool, but right.

52 00:06:40.320 00:06:42.779 Uttam Kumaran: but it’s whatever I don’t know.

53 00:06:43.680 00:06:49.830 Nicolas Sucari: I don’t use the pool, but I use the gym, and and I have like a sauna and all of that, but I.

54 00:06:50.174 00:06:50.520 Payas Parab: Nice.

55 00:06:50.520 00:06:52.469 Nicolas Sucari: Anything like that. Yeah, it’s really nice.

56 00:06:52.470 00:06:53.230 Payas Parab: Energy.

57 00:06:54.310 00:06:58.160 Nicolas Sucari: Like. There’s no way like I don’t know when to use it, you know it’s like

58 00:06:59.280 00:07:02.089 Nicolas Sucari: a girl. Use the gym, and that’s it.

59 00:07:02.830 00:07:11.339 Payas Parab: Need to. You need to take one of these calls from the Sauna. I think it could be a great power move. It’s just like, Hey guys, I’m in the Sauna, but wanted to give some updates here.

60 00:07:11.600 00:07:15.980 Uttam Kumaran: You know, I’ve done that not like a client call, but cause I I just get calls like.

61 00:07:16.390 00:07:23.929 Uttam Kumaran: every week I get a fucking phone call about something. So I’m like, wherever I am like, I’m in the Sauna. I’m like.

62 00:07:24.920 00:07:28.466 Nicolas Sucari: Okay, this will be just take the call. Fine.

63 00:07:29.400 00:07:37.029 Uttam Kumaran: It’s not a video call, though I’m the Sauna I go to is like a 2 person Sauna. At this gym I go to. So there’s usually nobody in there. It’s small.

64 00:07:39.500 00:08:04.560 Payas Parab: Joining. I think he had sent some stuff about the deck. I’d like updated my comments. I had a question for him what he was referring to there. But in general, I think we’re kind of also like after last week, we’re like just pencils down until we get the renewal for the most part right like, I believe there’s some tickets and then some dashboarding after migrate from amplitude. But Robert was kind of like, pencils down until we get the renewal. I don’t know, Tom. If that was like something you guys had chatted about.

65 00:08:05.440 00:08:07.260 Uttam Kumaran: No, I think that’s the last thing I heard, too.

66 00:08:07.260 00:08:09.319 Payas Parab: Okay, sweet. He’s joining awesome.

67 00:08:24.640 00:08:26.380 Robert Tseng: Hello! Hello! Can you hear me?

68 00:08:26.610 00:08:27.060 Payas Parab: Yep.

69 00:08:28.430 00:08:29.090 Robert Tseng: Okay.

70 00:08:32.600 00:08:35.967 Robert Tseng: alright. Let’s do this.

71 00:08:38.559 00:08:42.519 Payas Parab: Some very yellow sweatshirt right here, very.

72 00:08:42.520 00:08:42.959 Robert Tseng: And he hasn’t.

73 00:08:42.960 00:08:49.150 Payas Parab: Nice. It’s like a nice yellow, but it’s like very yellow, you know. It’s like nice, but it’s very, very yellow.

74 00:08:49.400 00:08:49.840 Uttam Kumaran: No.

75 00:08:49.840 00:08:50.260 Robert Tseng: That’s it.

76 00:08:50.260 00:08:51.590 Uttam Kumaran: I like that color. It’s nice.

77 00:08:51.590 00:08:52.400 Robert Tseng: So

78 00:08:53.390 00:08:59.349 Robert Tseng: you guys have probably seen me wear this too much. But this is rainy day weather. I don’t really want to wear anything else.

79 00:09:00.580 00:09:04.790 Uttam Kumaran: It’s the closest thing to sunshine we get when the weather isn’t bad. Here.

80 00:09:08.420 00:09:15.610 Robert Tseng: Okay, so I’m assuming that you guys got a chance to look at the deck.

81 00:09:15.910 00:09:16.730 Robert Tseng: Okay?

82 00:09:17.840 00:09:18.979 Robert Tseng: And then,

83 00:09:19.850 00:09:27.070 Robert Tseng: yeah, I was thinking that we could just run through it really like on on this call. And then figure out what else we need before we send it out to them on today.

84 00:09:29.270 00:09:31.419 Robert Tseng: Okay. So here we are.

85 00:09:32.050 00:09:39.893 Robert Tseng: So you know, slight, subtle rebrand right here and then,

86 00:09:40.520 00:09:43.059 Robert Tseng: yeah, I guess need some

87 00:09:43.430 00:09:48.549 Robert Tseng: thoughts here. Oh, I didn’t tag people in these sites. Okay? Well, maybe.

88 00:09:48.690 00:10:05.130 Robert Tseng: Yeah. I guess where I was going with, this was basically like 3 slides kind of covering each of the different like wins that we made hopefully. This is like great work that could end up being turned into our case study as well, so I don’t know if these numbers are exactly right, but I guess maybe I’ll 1st

89 00:10:05.630 00:10:13.130 Robert Tseng: kind of call out for pious like, yeah, I think we looked at a a spreadsheet that you put together, and you pulled like 30 day numbers.

90 00:10:13.130 00:10:13.580 Payas Parab: Yes.

91 00:10:13.863 00:10:17.260 Robert Tseng: Just to show that. So if we could kind of just somehow.

92 00:10:17.420 00:10:19.709 Robert Tseng: I don’t know. Get like a cleaned up table.

93 00:10:19.710 00:10:20.899 Payas Parab: Yeah, like, a nice. Okay.

94 00:10:20.900 00:10:21.560 Robert Tseng: Here.

95 00:10:21.560 00:10:21.880 Payas Parab: Sure.

96 00:10:22.141 00:10:27.590 Robert Tseng: You don’t have to use this format. But if we can, yeah, maybe even like mention, like.

97 00:10:28.280 00:10:34.730 Robert Tseng: Con, kind of calling out like, which of these things? Yeah, like, maybe it’s like,

98 00:10:36.630 00:10:39.270 Payas Parab: I mean, I think I think the categories here right? It’s like

99 00:10:39.420 00:10:43.499 Payas Parab: it’s like 1. 1 of them is like tied to like mutability. The second is like.

100 00:10:44.300 00:10:46.800 Payas Parab: Yeah, it’s like mutability. It’s

101 00:10:47.870 00:10:54.280 Payas Parab: I mean, I think it’s just reliance on events firing right like it’s like the lack of reliance on events firing.

102 00:10:54.470 00:10:56.460 Robert Tseng: Yeah, mutability reliability.

103 00:10:56.460 00:11:03.580 Payas Parab: Deduplication, I think, is another one, too, like there’s a lot of work that occurs now with yeah, deduplication.

104 00:11:04.550 00:11:06.500 Robert Tseng: Yeah, so

105 00:11:09.790 00:11:11.629 Robert Tseng: or, yeah, I don’t know. You can kind of structure that.

106 00:11:11.630 00:11:15.020 Payas Parab: Yeah, something like that. Yeah, it’s like, these are the 3 main benefits. Yeah.

107 00:11:15.320 00:11:16.380 Robert Tseng: Yeah. So if you can.

108 00:11:16.380 00:11:17.789 Payas Parab: You guys tag me there. So I know.

109 00:11:17.790 00:11:18.410 Robert Tseng: Yeah.

110 00:11:18.560 00:11:19.119 Payas Parab: Because I was just going.

111 00:11:19.120 00:11:19.890 Robert Tseng: Sure.

112 00:11:19.890 00:11:20.820 Payas Parab: Thank you.

113 00:11:21.540 00:11:22.450 Robert Tseng: Okay. Great.

114 00:11:22.450 00:11:23.010 Payas Parab: Yeah.

115 00:11:23.300 00:11:28.980 Robert Tseng: This one. Maybe this is more of like a Nico kind of one. But

116 00:11:29.180 00:11:34.850 Robert Tseng: maybe, you know, we replace all Aman’s like custom Cloud fair shit so we can like

117 00:11:35.270 00:11:43.813 Robert Tseng: I don’t know something, some flow showing how we like moved his calculations to the data warehouse I was thinking of.

118 00:11:44.630 00:11:45.070 Nicolas Sucari: It’s right.

119 00:11:45.070 00:11:49.799 Robert Tseng: To to show that you have any questions on that Nico.

120 00:11:50.020 00:11:51.680 Robert Tseng: or is that, does that make sense.

121 00:11:53.200 00:12:00.640 Nicolas Sucari: Yeah, I think it makes sense. I mean, we can share a little bit on how we are doing to ingest a lot of the data, share some screenshots and then

122 00:12:01.052 00:12:05.779 Nicolas Sucari: do the modeling until we get to the dashboard. And yeah, we can paste some screenshots there. Yeah.

123 00:12:05.780 00:12:13.579 Robert Tseng: Yeah, I don’t need you to write any essay just like some screenshots with arrows is good, like, I’m sure he kind of knows this already, but it’s just for.

124 00:12:14.220 00:12:15.709 Robert Tseng: but I don’t know if it ends up.

125 00:12:15.930 00:12:16.380 Nicolas Sucari: Yeah.

126 00:12:16.655 00:12:23.260 Robert Tseng: Maybe I’m gonna end up turning this into something more beautiful. But but yeah, I think just for the sake of communicating that

127 00:12:23.510 00:12:25.089 Robert Tseng: that was one of the wins

128 00:12:25.580 00:12:30.427 Robert Tseng: cost some cost assumptions. Maybe this is back to bias again as well.

129 00:12:31.460 00:12:36.840 Robert Tseng: because, you know, we have, like the we have one sheet with the inputs, and then

130 00:12:37.460 00:12:43.160 Robert Tseng: I don’t. I guess I didn’t actually click in to see what Jonathan had edited. But I don’t know if there’s

131 00:12:43.440 00:12:53.660 Robert Tseng: a way for us to visualize like, how easy that was to upload because now, from like one source, they can like update all of the cost. Assumptions right? So.

132 00:12:53.660 00:12:54.280 Payas Parab: Yeah.

133 00:12:54.280 00:12:56.510 Robert Tseng: Was kind of like what I was thinking. We could feature.

134 00:12:56.510 00:12:59.970 Payas Parab: I mean, we have the detailed diagram, but that one’s probably a little too

135 00:13:00.140 00:13:04.780 Payas Parab: detailed. I threw that on one of the slides. You tagged me, and I just threw that in the appendix with a link.

136 00:13:04.780 00:13:05.640 Robert Tseng: Yeah. Great.

137 00:13:05.640 00:13:08.449 Payas Parab: But like I was like, let’s just like, you know, like, here’s what

138 00:13:08.947 00:13:11.370 Payas Parab: I think it’s slide 23

139 00:13:12.450 00:13:14.849 Payas Parab: in the main. I just dumped it in the appendix. But this is the like.

140 00:13:14.850 00:13:15.390 Robert Tseng: Yep.

141 00:13:15.540 00:13:20.779 Payas Parab: The new. And and it might be a little technical. So maybe there’s like a simplified version of this like.

142 00:13:20.780 00:13:27.899 Robert Tseng: Yeah, I I want. I want Justin to see this. So to think about like the the 1st few slides like I want.

143 00:13:28.240 00:13:33.070 Robert Tseng: He has no idea what we’re doing. I wanted to be able to just see, like some something about like what we did.

144 00:13:33.300 00:13:33.770 Payas Parab: Sure.

145 00:13:33.770 00:13:42.919 Robert Tseng: Yeah, so yeah, I would say, probably kind of something simpler than slide 23. This is still a great diagram. And we yeah, definitely block.

146 00:13:42.920 00:13:49.519 Payas Parab: Yeah, the audience, for that was like Ryan and Nico, just to speed up that process. So they’re not spinning their wheels. But yeah, I can make a simplified version of that.

147 00:13:49.860 00:13:50.500 Robert Tseng: Cool

148 00:13:51.043 00:14:02.799 Robert Tseng: on the recap. This is just the notion. Page. So, yeah, that’s that. I kind of just relabeled some of these slides. These are our month. One kind of when we 1st started what? We what we knocked out.

149 00:14:03.030 00:14:10.496 Robert Tseng: Q. 3. Kind of summary. What we knocked out. Q. 4. Okay, I see you already. Kind of finish this

150 00:14:11.430 00:14:12.700 Robert Tseng: or did you? I’m not sure.

151 00:14:12.700 00:14:20.270 Payas Parab: I did. I did. Yeah, I I just use the words like, recently unblocked, because I’m like, literally the day before the contract ended, they essentially updated it right? So like.

152 00:14:20.557 00:14:21.992 Robert Tseng: Let’s just call it done.

153 00:14:23.790 00:14:24.359 Robert Tseng: Okay?

154 00:14:24.480 00:14:34.200 Robert Tseng: Yeah. But okay. So blocked by cogs reporting, and

155 00:14:37.230 00:14:45.200 Robert Tseng: okay, I think that’s pretty clear. They’ll they’ll understand that we didn’t finish that. But overall, I mean, I think this is done too. We we finish that.

156 00:14:47.220 00:14:48.250 Robert Tseng: Great.

157 00:14:48.490 00:14:55.849 Robert Tseng: Okay? Then I think this is really kind of the cell for like, why extend with us? I I think, just for everyone’s context.

158 00:14:56.140 00:15:07.519 Robert Tseng: yeah, Aman wants to do a month to month, but I’m trying to push him to a 3 month extension. And I think it’ll involve like these different objectives, so maybe we could spend a bit of time just

159 00:15:07.700 00:15:13.889 Robert Tseng: talking through this. This was just the staff Aug request that he had. He wanted a full time

160 00:15:14.210 00:15:22.123 Robert Tseng: analysts. And so, yeah, I guess he’s going with the one that we are kind of. We recommended.

161 00:15:23.340 00:15:30.719 Robert Tseng: yeah, I think you guys don’t have to worry about that. I that was that was on me. Who kind of working, working with him on this?

162 00:15:31.860 00:15:37.980 Robert Tseng: I don’t really think, as far as like, what, how this would actually impact our workflow.

163 00:15:38.700 00:15:45.568 Robert Tseng: This guy is just somebody who Jared or Justin would call to like add filters to amplitude reports.

164 00:15:46.060 00:15:49.430 Robert Tseng: and I guess there’s a bit of training on our end that.

165 00:15:49.580 00:16:03.920 Robert Tseng: and I don’t want any of us to spend our time on it. I think Jarell would be the one to do it to teach Matthew how to go into amplitude and change filters. But yeah, it’s just somebody that they can who’s available on call to like, be able to make these small changes.

166 00:16:04.040 00:16:06.030 Robert Tseng: but I would like to

167 00:16:06.220 00:16:14.599 Robert Tseng: like oversee his workload, so I can understand, like, what kinds of reporting requests is he getting like, how frequently is he getting them?

168 00:16:15.049 00:16:23.419 Robert Tseng: These are the same questions I was hoping for us to be able to answer, and they’re in Q. 4. But you know, if it takes like having like a

169 00:16:23.530 00:16:24.740 Robert Tseng: full time

170 00:16:24.860 00:16:31.491 Robert Tseng: guy to to figure that out, then that’s fine like it. It. This this is just something that

171 00:16:31.960 00:16:38.540 Robert Tseng: that that they that they they really want for some reason. So yeah.

172 00:16:38.540 00:16:42.200 Uttam Kumaran: Are they working with them? It’s just whether month to month, or 3 months.

173 00:16:42.480 00:16:46.790 Uttam Kumaran: Or is that like? Is is that not clear? Yet?

174 00:16:47.040 00:16:48.130 Robert Tseng: Oh, with him!

175 00:16:48.130 00:16:50.850 Uttam Kumaran: No, no with with with the Javi.

176 00:16:51.729 00:17:07.110 Robert Tseng: Well, I’m gonna put this forward. I’m gonna put forward a 3 month extension proposal. I know that I’m on once month to month. He’s verbalized that before, but I want him. I want him to just take take this so I don’t know, do do you? What do you? What do you think.

177 00:17:07.270 00:17:15.377 Uttam Kumaran: No, I was saying like, if you then is, would Matthew just be working directly with us, basically like through our process, or

178 00:17:15.730 00:17:18.370 Robert Tseng: Yeah. So Matthew would be.

179 00:17:18.690 00:17:20.700 Uttam Kumaran: I would prefer that. So that

180 00:17:20.890 00:17:22.700 Uttam Kumaran: basically we start to just watch.

181 00:17:23.010 00:17:24.000 Robert Tseng: Yeah, yeah, he would be.

182 00:17:24.000 00:17:31.310 Uttam Kumaran: I mean, I would say it’s maybe instead of through brain Forge, you could just say, like alongside, or basically like, he just goes through the same Pm process.

183 00:17:31.790 00:17:37.350 Uttam Kumaran: It’s just gonna be on his own. And it’s gonna screw us. And it’s gonna screw him basically.

184 00:17:37.350 00:17:43.299 Robert Tseng: Yeah, I mean, legally speaking, the drills staffing Org Fusion. Node

185 00:17:43.610 00:17:48.170 Robert Tseng: doesn’t want like I guess they will contract them him to.

186 00:17:48.170 00:17:48.739 Uttam Kumaran: You are.

187 00:17:48.740 00:17:54.310 Robert Tseng: To us directly, and then Aman wanted to say.

188 00:17:55.131 00:18:14.380 Robert Tseng: within like, maybe after the 1st month like that job, you would contract with him directly, which is whatever like I think we we’re gonna bake in some margin. I mean, he gets paid. He’ll his. His rate is like 13 an hour. So we’ll probably bake in some margin and probably bump him up to like 20 an hour or something, and that’ll be ours for the 1st month.

189 00:18:14.530 00:18:15.060 Uttam Kumaran: Okay.

190 00:18:15.060 00:18:16.950 Robert Tseng: And then, if

191 00:18:17.630 00:18:26.729 Robert Tseng: I don’t know if Joby loves him and wants to like work, kind of extend him and offer full time directly, like, whatever like, that’s that’s between them. So

192 00:18:27.136 00:18:36.200 Robert Tseng: that’s kind of what I had, what I what I had. At least, yeah. So just at least for the 1st month he would be part of our our team, I suppose.

193 00:18:37.540 00:18:38.210 Uttam Kumaran: Okay.

194 00:18:39.600 00:18:52.549 Robert Tseng: Okay, cool. So, and then on the Etl side, I know that. You know, Nico, you have the most context on kind of where we’re at with the 5 trend affordable. We already did some sort of discovery on like

195 00:18:52.860 00:18:58.513 Robert Tseng: I mean, I’m on is just asking like, Can you get all the connectors in there like? Do we know how much it would cost?

196 00:18:58.740 00:18:59.390 Nicolas Sucari: Yeah, no. No.

197 00:18:59.390 00:19:00.520 Robert Tseng: A couple of notes

198 00:19:00.750 00:19:07.320 Robert Tseng: there, I mean, maybe you could even reuse some of the stuff that was already mentioned in one of this messages.

199 00:19:08.910 00:19:26.700 Nicolas Sucari: Yes, I think. Yes, but I would need to like dig a little bit deeper into if we have, like all of the possibilities that 5 train is giving us in portable. But yeah, I mean in the pricing is 14. I think it’s 500 1,500 per 10 connectors. So that’s like.

200 00:19:26.880 00:19:47.839 Nicolas Sucari: that’s a fixed price for portable. If you wanna add that there in the notes. That’s fine, that that’s the scale plan. So that’s okay. But yeah, we need to dig deeper into if they have all of the connectors, if the connectors are bringing the same data. And we started doing that. But I don’t know if if we need to keep doing that now, or like, just wait until the renewal.

201 00:19:49.390 00:19:53.430 Uttam Kumaran: Yeah. I mean, I I agree, like, I don’t know. I’m not a hundred percent that

202 00:19:53.780 00:19:55.790 Uttam Kumaran: it’s gonna be a replacement. Yet.

203 00:19:56.000 00:19:57.699 Uttam Kumaran: We’re probably like 60%.

204 00:19:57.950 00:19:58.690 Nicolas Sucari: Yeah, exactly.

205 00:19:58.690 00:20:06.290 Nicolas Sucari: I don’t wanna say like, we’re gonna replace everything because I’m not sure if portable can do everything that 5 can do. But

206 00:20:06.490 00:20:09.210 Nicolas Sucari: yeah, I don’t wanna make sure. I wanna make sure that before.

207 00:20:09.210 00:20:16.829 Uttam Kumaran: Purposes of this will say, like, I mean, goal is to cut Bill by 50%.

208 00:20:23.870 00:20:26.729 Nicolas Sucari: Yeah, maybe maybe the objective is like.

209 00:20:27.010 00:20:32.899 Nicolas Sucari: Etm immigration from 5 tran or to other option. Yeah, not to portable directly.

210 00:20:32.900 00:20:34.450 Robert Tseng: Yeah, I won’t mention.

211 00:20:34.450 00:20:35.439 Uttam Kumaran: No cost or whatever.

212 00:20:35.440 00:20:36.710 Nicolas Sucari: Yeah, exactly.

213 00:20:37.190 00:20:39.700 Uttam Kumaran: And this is the real kicker.

214 00:20:39.700 00:20:40.200 Nicolas Sucari: Yeah.

215 00:20:48.720 00:20:50.240 Robert Tseng: Great. I think that’s fine.

216 00:20:54.020 00:20:57.036 Robert Tseng: What was I thinking here?

217 00:20:58.130 00:21:02.530 Robert Tseng: yeah, I guess developing key communications are pointing across the.

218 00:21:02.530 00:21:06.269 Uttam Kumaran: Well, just around a certain like business unit or.

219 00:21:07.130 00:21:10.399 Robert Tseng: Well, I think this is just like.

220 00:21:10.910 00:21:21.740 Robert Tseng: what does general maintenance like entail? I don’t know if we had anything specific. I know we have the performance marketing stuff that is like something that has been mentioned multiple times.

221 00:21:23.158 00:21:26.050 Robert Tseng: But you know, like, although

222 00:21:26.260 00:21:41.460 Robert Tseng: even the financial reporting and the and the order level, reporting that we’ve done like. There’s more that we could keep doing with that. So I they didn’t give us any specific request on how to take that step further. But I imagine that’s just something that we want to

223 00:21:42.390 00:21:44.830 Robert Tseng: call out that we could support.

224 00:21:55.630 00:21:56.440 Robert Tseng: Okay.

225 00:22:01.960 00:22:02.355 Uttam Kumaran: Okay.

226 00:22:04.050 00:22:04.720 Robert Tseng: Great

227 00:22:05.681 00:22:11.520 Robert Tseng: I left a couple of questions here. So this is kind of like where we spent a good chunk of our time. Last time we met with them on

228 00:22:13.230 00:22:29.420 Robert Tseng: basically, he showed us like an amplitude. Hey? They’re able to, you know, by bringing an email and matching their own guid that, like amplitude does some identity switching and they act. They can identify visitors that didn’t even necessarily purchase

229 00:22:31.150 00:22:40.180 Robert Tseng: I I think something that we said we would investigate was, we would see like, what share of in in visitors can they actually identify right now?

230 00:22:40.610 00:22:46.168 Robert Tseng: Maybe that I don’t know, I guess, since we didn’t do it. Maybe this is a bigger list than what we we thought.

231 00:22:46.690 00:22:57.690 Robert Tseng: I want to like. I guess there’s a there’s a bit of ambiguity here on like.

232 00:22:57.900 00:23:14.880 Robert Tseng: what of the market of the performance? Marketing capabilities? Should we move into Snowflake, I guess, and like, what’s our role in that? So. I don’t know. I guess maybe if we could kind of just align on that, I can edit the notes here to try to steer us in that direction.

233 00:23:18.360 00:23:19.330 Uttam Kumaran: Yeah.

234 00:23:25.310 00:23:28.690 Uttam Kumaran: Amplitude just does amplitude just does this with a pixel right now.

235 00:23:29.430 00:23:32.760 Robert Tseng: Yeah, it’s just a pixel. I guess.

236 00:23:33.400 00:23:38.199 Robert Tseng: because Pius correct me if I’m wrong. But we like they are able to look at.

237 00:23:38.770 00:23:42.649 Robert Tseng: There was like. So we we were talking internally, like.

238 00:23:42.910 00:23:47.640 Robert Tseng: Okay, it sounds like he wants to have like, I don’t know, like landing.

239 00:23:48.470 00:23:53.360 Robert Tseng: He wants to see a user that came in on a particular landing page.

240 00:23:53.700 00:23:54.680 Robert Tseng: And like.

241 00:23:55.160 00:24:07.240 Robert Tseng: yeah, this landing page, what? Which landing page? They they come in on like impact their their Ltv. Or their retention, or whatever it’s like this really big screen.

242 00:24:07.240 00:24:07.810 Uttam Kumaran: Yeah, yeah.

243 00:24:07.810 00:24:11.170 Robert Tseng: Like it goes beyond like campaign level.

244 00:24:11.749 00:24:19.050 Robert Tseng: Like, it’s not just channels and and campaigns that like attribution that they’re looking at. But yeah.

245 00:24:20.100 00:24:22.106 Uttam Kumaran: Yeah, I mean, basically what this is is like.

246 00:24:23.300 00:24:31.379 Uttam Kumaran: you need to just map their actions. And then, you know which customers took which actions, and then, of course, you know which one spent and which one didn’t.

247 00:24:31.900 00:24:34.709 Uttam Kumaran: I mean, this is something that we could totally do in Snowflake

248 00:24:35.438 00:24:39.109 Uttam Kumaran: in terms of the actual like identity resolution

249 00:24:39.670 00:24:44.200 Uttam Kumaran: like, I don’t know. That’s more about like using it. I got

250 00:24:45.380 00:24:49.190 Uttam Kumaran: that’s not something that’s stitching. I don’t know if it gets better

251 00:24:49.380 00:25:03.640 Uttam Kumaran: in Snowflake. But really, the really, the point is that like, if they’re having a hard time really mapping, saying, I want to look at 2 events, and how it maps to people spending these 3 events for this one event. That’s pretty easy to do in Snowflake.

252 00:25:07.530 00:25:12.140 Payas Parab: Like tracking that event then like. So you’d you’d bring in the events into Snowflake.

253 00:25:12.832 00:25:14.590 Uttam Kumaran: We have a sales data. Yeah.

254 00:25:15.070 00:25:22.009 Payas Parab: We have the sales data, right? Which yeah, we know is like probably to be more reliable like than an order event. Right? It’s like.

255 00:25:22.420 00:25:23.210 Payas Parab: but I don’t.

256 00:25:23.210 00:25:27.509 Uttam Kumaran: We need to use the order like if we can. If if that person is

257 00:25:28.260 00:25:35.730 Uttam Kumaran: like, amplitude should give us the stitched event, activity, history for that customer, and then when they click on order it.

258 00:25:35.840 00:25:38.059 Uttam Kumaran: we probably have their email right.

259 00:25:38.240 00:25:44.170 Payas Parab: How easy is that to pull out of there? Because I I think it’s like somewhat restricted right? And like how much you can pull out of like.

260 00:25:48.270 00:25:50.220 Uttam Kumaran: What do you mean? Pull up! Pull! What? Out of what?

261 00:25:50.220 00:25:53.480 Payas Parab: Like you’d have to move that into Snowflake right? You’d have to move.

262 00:25:53.480 00:25:53.910 Uttam Kumaran: Hmm.

263 00:25:53.910 00:25:59.929 Payas Parab: Like, Hey, this user? Id, we have this email, and I have it stitched to order in order in my snow.

264 00:25:59.930 00:26:02.250 Uttam Kumaran: They will do that like in Ga.

265 00:26:03.380 00:26:06.360 Uttam Kumaran: they’ll give you the user Id, if they can resolve it.

266 00:26:06.520 00:26:09.490 Uttam Kumaran: Similarly, in other tools that I’ve used.

267 00:26:09.840 00:26:10.270 Payas Parab: Okay.

268 00:26:10.270 00:26:12.069 Uttam Kumaran: They’ll put that into Snowflake.

269 00:26:14.780 00:26:19.339 Payas Parab: I’m I’m I’m just worried about with amplitude, because with the warehouse native stuff, too, it’s like

270 00:26:19.570 00:26:27.639 Payas Parab: it was just like they were. Just so it was just like, I mean again, like I’m on steak was like, it’s like underwhelming. And I’m like, I agree with you, dude like we tried it. But it’s like

271 00:26:28.390 00:26:34.070 Payas Parab: we realistically can can’t do shit in the amplitude like snowflake integration right now.

272 00:26:34.400 00:26:38.460 Payas Parab: like at least when it comes to like moving the snowflake data into there and then.

273 00:26:38.710 00:26:46.329 Payas Parab: like, you know, the warehouse native solution. So I’m worried that what you’re describing right is like moving the events out of amplitude into Snowflake.

274 00:26:47.040 00:26:50.780 Payas Parab: and then being able to like, do analysis with like our other tables.

275 00:26:50.880 00:26:54.859 Payas Parab: I worry that that. Have you done that before, for, sure or like I worry, we.

276 00:26:54.860 00:26:56.360 Uttam Kumaran: Odd with amplitude. No.

277 00:26:56.360 00:27:05.740 Payas Parab: That’s what I’m I’m worried. We get down that path like right, Robert, like we kind of did that with the other thing where it’s like we started down that path. Being like this is theoretically possible. And then, like.

278 00:27:05.870 00:27:19.499 Payas Parab: there’s a bunch of like weird blockers from amplitude on it. And I worry with the other thing, it’d be similar, because you’re sort of removing the data from their platform right? They just become like an SDK that collects data, not like a visualization tool.

279 00:27:19.810 00:27:24.020 Payas Parab: So I worry. There’d be some like blockers. There, that’s just my my concern.

280 00:27:27.730 00:27:30.699 Robert Tseng: Yeah. So let me just call out a couple of things. So

281 00:27:30.910 00:27:37.950 Robert Tseng: I just like pulled up some random user and and amplitude to see kind of like how the event stream is currently logged right now.

282 00:27:38.280 00:27:57.779 Robert Tseng: So, yeah, these are all amplitude events up to the to this point, like view, landing page view store product email collected. Okay, attendance. We have a different source here. We have like they’re adding the cart initiating checkout. Even this is not a I don’t think this is a shopify event. I think this is still a.

283 00:27:58.270 00:28:01.050 Robert Tseng: This is still a custom like amplitude event.

284 00:28:01.822 00:28:06.807 Robert Tseng: I guess this is one of the best one, because he doesn’t look like he actually made the order.

285 00:28:07.210 00:28:08.960 Robert Tseng: But yeah, I think it’s like.

286 00:28:10.020 00:28:17.700 Robert Tseng: okay, like, can we? Actually, well, we’re not exporting this entire stream into like into snowflake. Because.

287 00:28:17.910 00:28:25.550 Robert Tseng: yeah, theoretically, if we’re already hooking up to the 3rd party sources directly this attentive data we will have in there in Snowflake.

288 00:28:26.440 00:28:30.730 Robert Tseng: Like shopify order data we already have. But these, like other.

289 00:28:30.730 00:28:32.629 Uttam Kumaran: Yeah, the clicks. You don’t have the clicks.

290 00:28:32.630 00:28:34.338 Robert Tseng: We don’t have the clicks.

291 00:28:35.090 00:29:00.579 Robert Tseng: yeah. So if we like, isolate. Okay, all the custom click events that are in amplitude. We push those into Snowflake, as, like another events screen. That we can then merge with other events that we collected from other sources do is is the is the approach that we end up building like one massive event stream in Snowflake. And then we just do all the reporting off of that or like.

292 00:29:00.690 00:29:05.540 Robert Tseng: or why, I guess that’s kind of where we that’s where we feel like we need to figure that out.

293 00:29:06.660 00:29:11.359 Uttam Kumaran: Yeah. So when you when you sync amplitude to 5, 10, you’re gonna get all the events.

294 00:29:12.125 00:29:25.900 Uttam Kumaran: Whether or not whether or not we move, reporting all the way to there it. It may not be relevant, like we could just isolate the events they care about, which is like, okay. They want to look at customers who

295 00:29:26.110 00:29:35.789 Uttam Kumaran: they want to look, basically look at all of the clicked on or purchased events, or people who bought, and they want to trace that back with? Did they click on one button or the other?

296 00:29:36.340 00:29:47.880 Uttam Kumaran: You know, or like, for all the people that click on on one button versus the other. How many people ended up buying? You don’t need like everything there, you just amplitude, I think, will sync

297 00:29:48.440 00:29:53.650 Uttam Kumaran: all of the events, I’m pretty sure, but

298 00:29:53.830 00:30:07.359 Uttam Kumaran: like we could, you don’t have to like move all reporting to there you could. That would just solve that one problem like I looked at the 5 train connector. And this is probably how portable has a setup, too, is they just create one fad events table, basically with everything in it.

299 00:30:08.140 00:30:11.899 Uttam Kumaran: and then we just isolate the event in particular that they care about.

300 00:30:12.130 00:30:19.980 Uttam Kumaran: and then basically try to trace it back to the purchase event. If you could do that in amplitude, then I would just solve that problem here.

301 00:30:22.740 00:30:25.779 Uttam Kumaran: But this is where I don’t know whether the shopify data is in here.

302 00:30:28.980 00:30:34.469 Robert Tseng: So, yeah, I mean, I think it is a shit load of data like 7 days, 471.

303 00:30:34.470 00:30:40.230 Payas Parab: What’s the value in doing like? I just think it’s like the way I looked at it. And like we kind of talked about, this right is like.

304 00:30:40.380 00:30:59.560 Payas Parab: maybe you do just have both tools, and there’s just like one is your pre purchase journey, and the other is your post purchase journey right like. It doesn’t make sense to like migrate all the amplitude events that they’ve already stitched together for us, and have, like uis, that are built like that, the issue becomes in my mind with like shopify and the other types of data.

305 00:30:59.630 00:31:12.450 Payas Parab: they, they’re still events based right? Their entire format is events based. But for certain types of analysis, events don’t make sense right? Like I like, I think, like repeat order customers like we see that.

306 00:31:12.610 00:31:19.189 Payas Parab: like the Snowflake data and whatever shopify pulled in captured it way. Better so like I wanna make sure like

307 00:31:19.340 00:31:42.990 Payas Parab: like we, I would ask at the end of this is like, why did we do all that right like? Why did we move everything day like from amplitude into snowflake? Then we’re like storing down 2 sets of data. They’re paying for a bunch of snowflake data, warehousing and amplitude which part of the main part of the cost is data warehousing for them. So we’re just like duplicating that work when it doesn’t need to be. And

308 00:31:43.270 00:32:07.210 Payas Parab: I think amplitude. I just think it’s like we just have to get that idea. And in my mind. It’s like, Get that idea clear in his head of like that pre-purchase and post purchase journey, because in real business applications I don’t see that coming up that often. That’s like, oh, like customers that once like clicked on this listicle, are more likely to complain like that feels like an insane stretch to me, but maybe it maybe it is relevant. I don’t know.

309 00:32:12.560 00:32:20.441 Uttam Kumaran: Yeah, I mean, look, I think it’s just like if you if we can’t answer, I mean, you’re going through it. If we can’t answer this question here, then, we only have one option. Anyways.

310 00:32:20.890 00:32:24.530 Uttam Kumaran: if you can answer the question in amplitude, then I would just answer in amplitude.

311 00:32:26.030 00:32:38.330 Uttam Kumaran: But there’s, I mean, there’s obviously reasons to use Snowflake for a host of other stuff. But if you can basically put both events here, both the purchase event and another event. And then basically look at like.

312 00:32:39.350 00:32:44.769 Uttam Kumaran: hey, does does people clicking here versus another button purchase higher or lower?

313 00:32:45.750 00:32:50.239 Uttam Kumaran: I think if you have the purchase events in here. Then you can probably do it in here.

314 00:32:55.825 00:33:01.439 Uttam Kumaran: Can you isolate people that press one button people that press another button, and then the metric is like

315 00:33:01.820 00:33:04.139 Uttam Kumaran: account purchases, basically, or something.

316 00:33:07.740 00:33:09.850 Robert Tseng: Yeah, yeah, you can. I mean I. So

317 00:33:10.040 00:33:14.920 Robert Tseng: ye yes, you can. You can. You can do that and you can do that even with

318 00:33:15.240 00:33:19.020 Robert Tseng: out looking like like right now.

319 00:33:19.180 00:33:24.330 Robert Tseng: I I, this, this chart basically just shows, what share of users have

320 00:33:24.650 00:33:40.030 Robert Tseng: have like not been identified. And it. It looks like it’s 80. So to me, what this is telling me is 80% of the user data that they’re pulling into here is useless because it doesn’t end up tying to like an actual order. Yes, you can capture some of the behaviors of the users.

321 00:33:40.481 00:33:53.800 Robert Tseng: So you can know, like for this anonymous user. Like, yeah, we see this event stream. If he like, saw a landing pages for product. But he hasn’t done any like action that allows identity stitching to happen. But.

322 00:33:53.800 00:33:56.899 Uttam Kumaran: It says there’s only 20% of people buy like.

323 00:33:58.170 00:34:02.080 Uttam Kumaran: or is this is this literally like the identity stitching is not happening.

324 00:34:02.744 00:34:08.859 Robert Tseng: Well, yeah, I think that’s the question. I don’t know for sure, like, how much of this are. Yeah, we we don’t know is.

325 00:34:09.030 00:34:11.620 Robert Tseng: is this 80%? Yeah. Well, I don’t. I don’t know.

326 00:34:11.620 00:34:12.290 Uttam Kumaran: I mean, yeah.

327 00:34:12.290 00:34:12.880 Robert Tseng: Yeah.

328 00:34:13.504 00:34:18.440 Robert Tseng: I I mean, I would assume that most of them are people that don’t make a purchase that would make sense to me.

329 00:34:19.877 00:34:22.169 Uttam Kumaran: Do shit with that, anyway. So like.

330 00:34:22.179 00:34:36.129 Robert Tseng: Yeah. So if we were to move this into Snowflake, would, how would this number change, I guess, is kind of like a way that we can try to size. Size this for for a month, right at like enterprise level. No one’s like moving anonymous data into amplitude like this.

331 00:34:36.429 00:34:42.426 Robert Tseng: like Doordash was using amplitude for a while, but obviously, like most of the users, are not identified.

332 00:34:42.760 00:34:53.349 Uttam Kumaran: So most of the events are wasted. The only thing you do is just bring the events like, if you can’t, Etl, just the events you need then, in the model. We only isolate the events that matter.

333 00:34:53.650 00:35:00.600 Uttam Kumaran: You know you’re not bringing every single page view and stuff like that. The only reason I see this important is like, look, let’s say they have, like

334 00:35:01.060 00:35:06.009 Uttam Kumaran: 5 or 10 key events that they want to track if they correlate to purchases.

335 00:35:06.260 00:35:11.699 Uttam Kumaran: and they want to do that alongside other data, and they want to have that in real, then it makes sense.

336 00:35:11.960 00:35:24.859 Uttam Kumaran: Otherwise, if this is a analysis, I think it’s probably solvable here by just looking at of the folks that we can identify. Here is the relation of one event to the purchase event. That’s it, right.

337 00:35:28.110 00:35:41.929 Robert Tseng: Yeah. So if I were to do that in amplitude, I would probably, you know, like, make sure that they are identified. And I would turn. This is like the cohort of users I would use, and then I would start to just like, do I would just I would just plot like.

338 00:35:42.180 00:35:48.380 Robert Tseng: I don’t know different. Yeah, you would try to find you would do a correlation analysis to try to find like which of these, like actual

339 00:35:49.382 00:35:51.820 Robert Tseng: pre-sale events correlates.

340 00:35:52.050 00:35:57.079 Robert Tseng: But you’re trying to find like, what’s that ideal journey that users take before they get to purchase? Yeah.

341 00:35:57.080 00:36:05.550 Uttam Kumaran: That’s something like, I don’t. I just don’t know if that’s easy to do in amplitude, I would probably just say, like, Get all the data and then do that in excel.

342 00:36:08.220 00:36:18.699 Uttam Kumaran: cause it’s not you could do this in Snowflake like you can write this query. I don’t know whether, like you could write this query in amplitude to like, tell you this combination of events

343 00:36:20.570 00:36:26.169 Uttam Kumaran: not only is happens, but it also is the highest like correlation to a purchase, right.

344 00:36:27.830 00:36:38.299 Robert Tseng: Yeah, I mean, they have, like a few built in like the of these these charts. They can get you there. But it’s it’s very yeah. It’s it’s not gonna be as granular as what you could do in excel or writing.

345 00:36:38.300 00:36:41.799 Uttam Kumaran: Like, I think, Price, if you just get the

346 00:36:41.980 00:36:48.320 Uttam Kumaran: if you just get the events that matter in the purchase events, then I think you could easily, pretty much just do this in excel.

347 00:36:50.920 00:36:53.619 Uttam Kumaran: but they may not be able to check this all the time.

348 00:36:56.660 00:37:16.520 Robert Tseng: Okay. So I mean, I guess, like, kind of the the line that we’re drawing the sand is like, are we going to push them towards move everything into Snowflake. You don’t actually need amplitude. Anything that you need for amplitude. We could build it in Snowflake, snowflake, Meta base, or whatever or like. Are we going to keep? Tell them, keep amplitude, and this is the narrowly

349 00:37:16.750 00:37:25.110 Robert Tseng: like scoped way of how you should use amplitude. Everything else should not be an amplitude like I feel like we need to give them some direction like that.

350 00:37:25.110 00:37:26.849 Uttam Kumaran: That’s how I feel. Look so if

351 00:37:27.080 00:37:38.630 Uttam Kumaran: amplitude has to exist to do the event collection and everything that’s related to site metrics. And like, yeah, pre purchase. But then I think everything should also get pushed.

352 00:37:38.870 00:37:42.990 Uttam Kumaran: and we should do the whole like holistic analysis in Snowflake.

353 00:37:43.290 00:37:45.780 Uttam Kumaran: That’s this is that’s the common pattern.

354 00:37:47.160 00:37:47.970 Uttam Kumaran: Like

355 00:37:48.240 00:37:53.170 Uttam Kumaran: you can brand. We can’t do. We can’t link shopify stuff here. We can’t link gorgeous stuff here.

356 00:37:53.791 00:37:56.079 Uttam Kumaran: That stuff is impossible really easily.

357 00:37:56.558 00:38:01.750 Uttam Kumaran: Doesn’t mean amplitude goes away. It’s just yeah. I would say that that’s the narrow scope is like.

358 00:38:01.960 00:38:09.020 Uttam Kumaran: basically quick analysis where you just want to consider web data could do it here. Otherwise it’s the same data that we’ll get in Snowflake.

359 00:38:10.210 00:38:15.150 Payas Parab: There’s there’s also like on the marketing use case. We kind of have just been handing like

360 00:38:15.460 00:38:26.940 Payas Parab: prepackaged things like we actually should. If we better understand that use case right? Like this concept of like, okay, these are listicles. And we want to track. That’s what it’s doing right like. Then, like Guam said is like, maybe if we do have a data migration, it’s like

361 00:38:27.180 00:38:33.879 Payas Parab: we have, like very specific events. It’s like this listicle. And it’s like an attribute in a snowflake table. That’s like view.

362 00:38:33.880 00:38:35.859 Uttam Kumaran: Person watched person saw a listicle.

363 00:38:35.860 00:38:43.410 Payas Parab: Yes, let’s put them in. That is like the best way for their future. To be able to do that. Analysis is like we have a pipeline that’s like

364 00:38:43.950 00:38:53.189 Payas Parab: has viewed Listicle and that’s like a Boolean right in like the order. Right? Is like originating listed. Not a Boolean sorry like that’s a field.

365 00:38:53.190 00:39:04.210 Uttam Kumaran: Exactly it. So Tip, like another way of saying, this is like we had clients who were like came from this channel or engaged with a blog before. And you can just basically have a bunch of Boolean flags.

366 00:39:04.640 00:39:07.150 Uttam Kumaran: And then that table is easily created where it’s like.

367 00:39:07.150 00:39:08.010 Payas Parab: That’s yeah.

368 00:39:08.910 00:39:13.299 Uttam Kumaran: Yeah, that’s something it’s like you. It’s easy, very possible, to do in Snowflake. And like

369 00:39:16.640 00:39:22.750 Uttam Kumaran: I would just do that in Snowflake, and then pies for context. Snowflake is, it’s like very cheap cost them like a hundred bucks a month. So

370 00:39:23.070 00:39:31.520 Uttam Kumaran: the storage isn’t gonna really change much. It’s really like. How often they run models is really the cost. The storage stuff is pretty cheap.

371 00:39:31.520 00:39:32.740 Payas Parab: Got it. Okay.

372 00:39:35.340 00:39:36.110 Robert Tseng: Okay.

373 00:39:39.130 00:39:40.060 Robert Tseng: Yeah.

374 00:39:40.720 00:39:48.689 Robert Tseng: I alright. So I think I think sounds like we’re in agreement that we want to push them towards

375 00:39:49.290 00:39:50.130 Robert Tseng: like.

376 00:39:52.470 00:40:03.779 Robert Tseng: yes, define. This is the narrowly scoped usage of amplitude, the rest of the data. Let’s push it into into snowflake. I mean, I imagine, Utam, what you’re describing is that like user

377 00:40:03.970 00:40:23.119 Robert Tseng: user table, with all these custom attributes like, that’s kind of what amplitude tries to do with. It’s like concept of profiles like you have right now, this is messy as hell, because it has like 60 like properties right now. But this is basically what they’re trying to do like they have every user that’s tied to the campaign. The contract that they came in on.

378 00:40:23.410 00:40:23.770 Uttam Kumaran: Give you?

379 00:40:24.570 00:40:27.900 Uttam Kumaran: Can you create a custom attribute like, can you create a combination

380 00:40:29.820 00:40:36.700 Uttam Kumaran: on your own like you could say, this medium, this source and click this button, mark them as like X.

381 00:40:37.540 00:40:42.839 Robert Tseng: Yeah, I. So you’re trying to do like like properties that are calcul that are like aggregation.

382 00:40:42.840 00:40:43.220 Uttam Kumaran: Don’t worry about.

383 00:40:43.220 00:40:49.140 Robert Tseng: Properties. Yeah, you can’t. You can’t do that. Yeah, these are all custom fields that are assigned through his. I don’t know. Like whatever.

384 00:40:49.140 00:41:16.900 Uttam Kumaran: Yeah, so these are just descriptors. I think the thing that he would they would want is like, Can we start to one? I understand? There’s 2 questions. One, can we understand what people, what drives people to purchase like, what? What process through the site? The second thing is to affect that we need to track that long term. Right? So then, for like cool, we want to send more traffic to those places. We need to measure that like combination of factors as a flag

385 00:41:17.160 00:41:20.149 Uttam Kumaran: that, like we have to do in Snowflake, as like a.

386 00:41:20.600 00:41:24.360 Uttam Kumaran: you know, like watch the blog, or read the the thing, or.

387 00:41:25.930 00:41:33.560 Robert Tseng: Yeah, cause. Right now, you would just basically go into report and amplitude, and you would go select those attributes. And you would like.

388 00:41:33.560 00:41:34.040 Uttam Kumaran: And then.

389 00:41:34.040 00:41:35.790 Robert Tseng: Like create that? Yeah.

390 00:41:36.030 00:41:44.820 Uttam Kumaran: But that would never stick. So 1st thing is finding out like what it is, what is the combination? The second thing is that measuring and growing

391 00:41:44.930 00:41:46.470 Uttam Kumaran: that right.

392 00:41:46.470 00:41:47.010 Robert Tseng: Yeah.

393 00:41:47.010 00:41:49.629 Uttam Kumaran: I think that’s maybe a good way of articulating to them is like.

394 00:41:49.630 00:41:52.340 Robert Tseng: Yeah, there’s no way of surfacing. What are those combinations.

395 00:41:52.340 00:41:54.939 Uttam Kumaran: No, you can’t. This is a that’s a job to do. Yeah, yeah.

396 00:41:54.940 00:42:05.399 Robert Tseng: Yeah, it’s a very like needle in the haystack approach. Yes, you’re collecting all this stuff. But like you have to manually create reports, and that’s what the full time analyst would be doing. You’d literally just be creating.

397 00:42:05.630 00:42:10.920 Robert Tseng: trying to create a bunch of cohorts to find what the highest value one is because it won’t surface that to you.

398 00:42:11.770 00:42:14.239 Uttam Kumaran: It’s gonna be very hard to do that here. And

399 00:42:14.540 00:42:22.029 Uttam Kumaran: and it’s going to be just something. If you just know how to do the correlation analysis with factors. You could do that in Snowflake pretty easily.

400 00:42:22.320 00:42:23.050 Robert Tseng: Yeah.

401 00:42:23.650 00:42:27.030 Uttam Kumaran: Like find the propensity based on the event for a conversion.

402 00:42:27.430 00:42:34.860 Uttam Kumaran: So that’s what I would tell them, I said. Amplitude just has the descriptions of everything. But if one, we wanna actually do that analysis like

403 00:42:35.200 00:42:44.760 Uttam Kumaran: it’s gonna be way easier to do it in sequel, or that person will have to export from amplitude anyways. And the second thing is, once you have that combination, we want to track towards it. Basically.

404 00:42:45.140 00:42:50.809 Robert Tseng: Yep, okay, alright. I think I know how to finish this and bring it home. So.

405 00:42:51.760 00:42:55.949 Uttam Kumaran: Do you want to check out the if we’re done you want? I can show you the figma thing really quick.

406 00:42:56.170 00:42:57.570 Robert Tseng: Oh, yeah, yeah, I’m done.

407 00:43:01.630 00:43:08.399 Nicolas Sucari: I’m I’m gonna work on on those screenshots, Robert, and I’m gonna send it to you. You can replace the one from that slide. Okay.

408 00:43:08.400 00:43:09.789 Robert Tseng: Yeah. Thanks. Nico.

409 00:43:17.950 00:43:20.539 Uttam Kumaran: Okay, if you’re in sales assets here.

410 00:43:23.790 00:43:28.170 Uttam Kumaran: Basically, what you can do is if you just select like

411 00:43:30.390 00:43:32.079 Uttam Kumaran: like, we just have everything in this one.

412 00:43:32.080 00:43:34.880 Robert Tseng: Oh, okay, it’s all in the same area. Got it? Okay? So.

413 00:43:34.880 00:43:38.490 Uttam Kumaran: Yeah, you can literally just like, take this.

414 00:43:38.750 00:43:41.150 Uttam Kumaran: copy it, paste it, move it.

415 00:43:41.150 00:43:41.740 Robert Tseng: Okay.

416 00:43:41.970 00:43:50.610 Uttam Kumaran: And then can you just check? Can you try to just edit one thing? It’s it may give you an issue with fonts. And then I’m just gonna send you the right fonts.

417 00:43:51.220 00:43:53.070 Uttam Kumaran: It’ll it’ll take like one second.

418 00:43:53.620 00:43:55.300 Robert Tseng: If I if I edit, and something’s off.

419 00:43:55.300 00:43:58.910 Uttam Kumaran: See if you can go into here and like edit up to the text.

420 00:43:59.490 00:44:00.310 Robert Tseng: Okay.

421 00:44:01.280 00:44:04.449 Robert Tseng: Let me do that. I have it open right now. Let me just do it.

422 00:44:05.920 00:44:08.280 Robert Tseng: Do do.

423 00:44:22.300 00:44:26.219 Uttam Kumaran: Oh, do you have? Wait! Hold on! Is it letting you edit or not?

424 00:44:27.610 00:44:30.189 Robert Tseng: I’m not sure I just. I just moved to the right.

425 00:44:30.560 00:44:31.760 Robert Tseng: No, I can’t.

426 00:44:31.760 00:44:33.140 Uttam Kumaran: Okay, one, second.

427 00:44:33.440 00:44:36.630 Robert Tseng: Did I? Did. I just do that, or did you do that? I’m not sure.

428 00:44:36.630 00:44:38.720 Uttam Kumaran: I was like I just left. I can do anything I want.

429 00:44:38.720 00:44:43.120 Robert Tseng: Oh, okay, I like created a frame, but I don’t really know what that means.

430 00:44:44.230 00:44:46.130 Uttam Kumaran: Okay, well, you do have.

431 00:44:46.520 00:44:47.260 Nicolas Sucari: Try like.

432 00:44:47.260 00:44:48.019 Robert Tseng: Oh, shoot! Please.

433 00:44:48.020 00:44:52.539 Nicolas Sucari: Clicking on the text like you need to like click until you get

434 00:44:52.540 00:44:54.380 Nicolas Sucari: it really is the next spot.

435 00:44:55.420 00:44:59.580 Robert Tseng: Yeah, every time I click it, it just opens another brand new box.

436 00:44:59.900 00:45:01.839 Uttam Kumaran: Keep clicking into it until it goes, until, like.

437 00:45:01.840 00:45:02.610 Nicolas Sucari: Clicking. Yeah.

438 00:45:02.610 00:45:05.049 Uttam Kumaran: It’s like deeper and deeper. It’s just layers.

439 00:45:05.308 00:45:07.120 Nicolas Sucari: If you see on the left there.

440 00:45:07.120 00:45:09.630 Uttam Kumaran: So like like click onto the text.

441 00:45:10.934 00:45:13.980 Robert Tseng: Okay, it’s telling me you’re missing font. Is that what you’re talking about?

442 00:45:13.980 00:45:15.959 Uttam Kumaran: Yeah. What fun are you missing.

443 00:45:15.960 00:45:18.060 Robert Tseng: Helvetica. New.

444 00:45:19.110 00:45:21.012 Uttam Kumaran: One sec. I’m gonna send you the phones.

445 00:45:21.250 00:45:31.505 Robert Tseng: Okay, okay, that’s why I was not like getting the clicking thing. Okay, yeah, I just

446 00:45:32.510 00:45:34.980 Uttam Kumaran: I just tagged you in slack

447 00:45:35.210 00:45:45.850 Uttam Kumaran: where the font is. I need you to. I need you to one download the font and then just double click on it, and Apple will tell you you want to install this and just

448 00:45:46.420 00:45:47.309 Uttam Kumaran: install it.

449 00:45:48.230 00:45:50.040 Robert Tseng: Hey? Alright! Alright!

450 00:45:50.990 00:45:54.390 Robert Tseng: I just think I’m I think I’m doing it.

451 00:45:55.400 00:45:58.139 Robert Tseng: Install. Sigma, is it okay?

452 00:45:58.140 00:46:01.189 Uttam Kumaran: And then it. I just need you to install this font installer.

453 00:46:05.620 00:46:13.850 Uttam Kumaran: If you go to the figma downloads, there’s 2 links. So there’s the actual font. You can download that double, click. It, install it onto your machine.

454 00:46:14.920 00:46:21.730 Uttam Kumaran: Do is just install the font, install our agent or font installer.

455 00:46:22.090 00:46:25.440 Robert Tseng: Yeah, I have the fun. Do I have to get the map, the desktop app to you? Then.

456 00:46:27.040 00:46:29.269 Uttam Kumaran: No. Do you already have the desktop app.

457 00:46:29.270 00:46:30.050 Robert Tseng: No, I don’t.

458 00:46:30.310 00:46:33.250 Uttam Kumaran: Oh, then, just install the installer. Don’t worry. You don’t need the desktop.

459 00:46:33.720 00:46:38.340 Robert Tseng: Yeah, I have the Mac OS installer for the font installer, which I did that already.

460 00:46:38.340 00:46:42.620 Uttam Kumaran: Okay, cool. And then, did you already install the font onto your machine?

461 00:46:43.320 00:46:45.929 Robert Tseng: I don’t think he linked me the font. I think you just.

462 00:46:45.930 00:46:47.389 Uttam Kumaran: Threads, too, that it’s it’s.

463 00:46:47.903 00:46:48.930 Robert Tseng: Okay. Okay.

464 00:46:48.930 00:46:51.989 Uttam Kumaran: And message. Aesthetica New.

465 00:46:52.910 00:46:54.350 Robert Tseng: Yeah, I got it.

466 00:46:54.350 00:46:56.670 Uttam Kumaran: Download that, install it, and then refresh.

467 00:46:56.800 00:46:59.829 Uttam Kumaran: Think much should shouldn’t give you an issue.

468 00:47:00.260 00:47:01.000 Robert Tseng: Okay?

469 00:47:03.970 00:47:05.540 Robert Tseng: Retch.

470 00:47:13.480 00:47:14.400 Uttam Kumaran: What did it say?

471 00:47:15.730 00:47:18.910 Robert Tseng: No, I don’t think it refreshed, because it still says I’m missing it.

472 00:47:21.170 00:47:25.040 Uttam Kumaran: Did you? Oh, you installed the font on your machine.

473 00:47:25.700 00:47:26.300 Robert Tseng: Yep.

474 00:47:26.790 00:47:31.629 Robert Tseng: And I I yeah, the figment Asian installer. I kind of loaded that already.

475 00:47:32.415 00:47:35.949 Robert Tseng: I don’t know if I have to open something to load in the font, but.

476 00:47:38.790 00:47:42.299 Uttam Kumaran: Can you close your browser and reopen it.

477 00:47:42.830 00:47:44.800 Robert Tseng: Yeah, I think I think I’ll do that.

478 00:47:47.050 00:47:48.849 Robert Tseng: There’s like no refresh.

479 00:47:49.610 00:47:50.200 Uttam Kumaran: It’s

480 00:47:51.960 00:47:58.000 Uttam Kumaran: yeah. It’s just the designers all know how to do these things. And I had to learn all this shit.

481 00:47:58.580 00:48:00.999 Robert Tseng: Okay, I just refreshed somehow. So.

482 00:48:01.000 00:48:03.559 Uttam Kumaran: Now just double, click into like any

483 00:48:06.290 00:48:09.910 Uttam Kumaran: any text. Just keep clicking until you get to the text.

484 00:48:10.080 00:48:11.589 Robert Tseng: Oh, yeah, I got it. Okay.

485 00:48:11.590 00:48:13.210 Uttam Kumaran: Okay. Great. Perfect.

486 00:48:13.487 00:48:19.309 Robert Tseng: But it looks like there’s a couple of other texts that I can’t. Oh, no, they’re all they’re all the same.

487 00:48:19.310 00:48:21.351 Uttam Kumaran: Should be good with everything.

488 00:48:21.900 00:48:30.560 Uttam Kumaran: yeah, we have a whole like design, brand book and everything. But yeah, hopefully, do you need more than one page. If if so, you can just copy the

489 00:48:31.140 00:48:31.940 Uttam Kumaran: the.

490 00:48:31.940 00:48:36.890 Robert Tseng: Yeah, I I might copy it. Okay, let me just finish this, and then I’ll send it over.

491 00:48:36.890 00:48:37.225 Uttam Kumaran: Okay.

492 00:48:38.160 00:48:41.559 Robert Tseng: Yeah, then oh, well.

493 00:48:45.470 00:48:48.250 Robert Tseng: I kinda screwed up whatever. Oh.

494 00:48:49.000 00:48:54.930 Robert Tseng: I like added like a couple of squares that like, I don’t know how to delete man

495 00:48:55.380 00:48:58.899 Robert Tseng: or whatever I’ll I’ll figure it out. I’ll just re redo it. Okay.

496 00:48:59.240 00:48:59.590 Uttam Kumaran: Okay.

497 00:48:59.940 00:49:05.920 Robert Tseng: All right. Yeah. I mean, that’s why I’m gonna be working on finishing the Javi proposal. And.