Meeting Title: Javy-Data-Engineering-Weekly Date: 2024-10-29 Meeting participants: Nicolas Sucari, Uttam Kumaran, Aman Nagpal, Payas Parab, Robert Tseng
WEBVTT
1 00:02:01.590 ⇒ 00:02:02.450 Nicolas Sucari: Hi guys.
2 00:02:05.170 ⇒ 00:02:06.549 Payas Parab: Hey, Nico! How are you?
3 00:02:06.990 ⇒ 00:02:07.475 Uttam Kumaran: Hey!
4 00:02:09.460 ⇒ 00:02:10.255 Nicolas Sucari: I’m doing good.
5 00:02:10.530 ⇒ 00:02:12.559 Nicolas Sucari: hey? I just realized
6 00:02:12.700 ⇒ 00:02:14.109 Nicolas Sucari: thing that we
7 00:02:15.290 ⇒ 00:02:18.009 Nicolas Sucari: don’t have live deal order items.
8 00:02:18.230 ⇒ 00:02:22.370 Nicolas Sucari: dashboard like the complete one with shopify and Amazon. I’m gonna add it right now.
9 00:02:24.470 ⇒ 00:02:25.480 Aman Nagpal: How’s it going? Guys.
10 00:02:26.750 ⇒ 00:02:27.640 Nicolas Sucari: Hey! Man!
11 00:02:59.280 ⇒ 00:03:01.849 Payas Parab: I’m not sure if Robert’s gonna be joining.
12 00:03:02.458 ⇒ 00:03:08.379 Payas Parab: Yeah, I’m on. We also have a sync with later today with Justin and Jared as well.
13 00:03:08.810 ⇒ 00:03:13.299 Payas Parab: Just kind of going over some of the the like. Interim deliverables with them.
14 00:03:13.820 ⇒ 00:03:29.630 Payas Parab: so for this meeting, I think. Just want to give you an update on like where we’re at. I know the de team was working on integrating some other data sources. Probably have some questions around that we’re building the pipelines getting things together. We also wanted to briefly kind of overview.
15 00:03:29.720 ⇒ 00:03:42.369 Payas Parab: I think when we started like using a bunch of tools to kind of get updates from Jared and Justin. People are getting like a little confused around like what all these tools are and everything. So we put together a little I mentioned we put together like a little
16 00:03:42.640 ⇒ 00:03:53.009 Payas Parab: like, you know, what? What are all these tools like? Why do we need them and stuff like that. So I have a draft of that as well ready. But Nico, I don’t know if you guys had a agenda that you wanted to go through
17 00:03:54.260 ⇒ 00:03:55.000 Payas Parab: in terms of.
18 00:03:55.000 ⇒ 00:03:56.040 Nicolas Sucari: No, no, that’s.
19 00:03:56.293 ⇒ 00:04:02.369 Payas Parab: We can keep this meeting pretty short because I don’t want to also be duplicative. I’m on. I don’t know if you’re later.
20 00:04:03.020 ⇒ 00:04:08.939 Aman Nagpal: Yeah, I will. Try to join that one as well if I can. I guess.
21 00:04:09.240 ⇒ 00:04:16.470 Aman Nagpal: Yeah, the best thing to do is to just go over kind of status updates on what’s next, what we’re doing. So
22 00:04:17.660 ⇒ 00:04:26.289 Aman Nagpal: I mean, I guess I’ll throw out a few random items that are on top of my head, and then what else we had that, you know. Maybe I’m missing. But
23 00:04:26.310 ⇒ 00:04:31.510 Aman Nagpal: one was the Okendo stuff, I believe. Did we start doing that?
24 00:04:32.550 ⇒ 00:04:37.289 Nicolas Sucari: We talked about starting with gorgeous if I didn’t
25 00:04:37.630 ⇒ 00:04:58.419 Nicolas Sucari: remember wrong. But yeah, I I already. I didn’t start with the kendo. I did start looking into gorgeous data source in order to import that. We’re gonna use the light connector that 5 trend has for now. That’s gonna yeah, go through the Api way for now. And we are not gonna use the web hooks.
26 00:04:58.420 ⇒ 00:05:12.879 Nicolas Sucari: I don’t think we need that right now. Like to have live updates from gorgeous, I think, just like one day one time they think it’s gonna be okay for now. So we we need there some access credentials. So, Aman, if you can share
27 00:05:13.348 ⇒ 00:05:20.529 Nicolas Sucari: any access or give me access to gorgeous, we can start importing that into Snowflake, too. Yep.
28 00:05:21.480 ⇒ 00:05:30.209 Aman Nagpal: That sounds good. So with gorgeous, you said, we’ll use 5 trend. Do they have a native integration with gorgeous? Or are we just connecting it to the Api.
29 00:05:32.248 ⇒ 00:05:35.300 Nicolas Sucari: They have a light connector created.
30 00:05:36.240 ⇒ 00:05:40.038 Uttam Kumaran: it connects through the Api, so we’ll just need
31 00:05:40.580 ⇒ 00:05:45.859 Uttam Kumaran: an Api key from you and the subdomain and the username. Basically.
32 00:05:46.010 ⇒ 00:05:48.230 Uttam Kumaran: So I think it’s
33 00:05:48.400 ⇒ 00:05:56.450 Uttam Kumaran: if you just go into gorgeous and you go under your profile, you can actually just generate the Api key right there.
34 00:05:57.530 ⇒ 00:06:02.753 Uttam Kumaran: I mean, even if you have. If you have a minute on this call, too, we could probably just grab it, and I can get it hooked up.
35 00:06:03.360 ⇒ 00:06:10.649 Aman Nagpal: Yeah, I’m actually logging in now. I just sent the sub domain. So I think we can try to knock that. Now, what is a light connector.
36 00:06:11.929 ⇒ 00:06:21.809 Uttam Kumaran: So a light connector is basically they just don’t have like a ton of expansive tables for it. Basically meaning
37 00:06:22.540 ⇒ 00:06:23.670 Uttam Kumaran: likes.
38 00:06:24.120 ⇒ 00:06:35.029 Uttam Kumaran: They. The way 5 Trin works is sometimes they just have like A, they basically support almost like 300 400 connectors, some connectors they want to release, but they’re adding more tables to it. So it’s almost like
39 00:06:36.120 ⇒ 00:06:44.769 Uttam Kumaran: staging versus production. So it’s something that they they just added access to recently. But there’s they’re continuing to kind of beef it up.
40 00:06:45.242 ⇒ 00:06:59.009 Uttam Kumaran: So for us, kind of the goal is we for 5, 10, we get about 14 days of like free data for them. For any connector. We’ll hook it up, find out whether it has all the information. If not, we have a couple of other paths.
41 00:07:01.220 ⇒ 00:07:02.000 Aman Nagpal: Do they have a connector.
42 00:07:02.000 ⇒ 00:07:02.740 Nicolas Sucari: Good evening.
43 00:07:03.010 ⇒ 00:07:05.240 Aman Nagpal: It’s kind of like a trial period, almost.
44 00:07:05.240 ⇒ 00:07:13.055 Uttam Kumaran: Hey? They have, and they call it light. It’s almost. They’re just trying to say that, hey, this may not have every single thing, and it may not be like fully
45 00:07:14.650 ⇒ 00:07:26.409 Uttam Kumaran: like, it may not have been fully fleshed out for Facebook and stuff. Those are. You know, they’ve had those centers for almost like 5 or 6 years now. For gorgeous. I think it’s just something that they developed in the last year.
46 00:07:26.840 ⇒ 00:07:36.370 Aman Nagpal: Got it. Yeah, let me try to get you that now and then in terms of the data. So I know I think I sent a few different scenarios. But
47 00:07:36.920 ⇒ 00:07:38.820 Aman Nagpal: is it going to
48 00:07:39.520 ⇒ 00:07:48.370 Aman Nagpal: keep syncing? I I guess I wanna make sure we’re syncing the right data right from gorgeous. So some of the examples I think I sent over. Yes.
49 00:07:48.370 ⇒ 00:07:49.620 Nicolas Sucari: Yeah. Some questions.
50 00:07:50.140 ⇒ 00:08:12.909 Aman Nagpal: Yeah, like, what macros are being used the most. What ticket fields are being used the most that way we can. You know, we hire a lot of customer service reps abroad, and if they’re not up to date on our cancellation policies or any sort of Cx policy, you know we can kind of track what they’re doing. So if one rep, for example, has an extremely high cancellation rate
51 00:08:13.188 ⇒ 00:08:39.409 Aman Nagpal: we can, you know, see the number of tickets that this person is attached to see what macros they’ve used. So if they’ve used the cancellation Macro over and over and we can see that over time in amplitude, for example. So that’s we just wanna make sure that all of that data is there as well as you know initially, ticket comes in. It has one status. Then by the end it’s closed. Maybe it’ll reopen. So just that all that data will keep updating as we go on.
52 00:08:39.650 ⇒ 00:08:54.684 Uttam Kumaran: Yeah, I’m I’m just looking at like the basically the Erd right now, we’ll have all that information. So not only just macros, but custom fields, the tickets themselves, the team members, and and the messages themselves.
53 00:08:55.090 ⇒ 00:08:58.929 Uttam Kumaran: So I think one of those like what you mentioned, which is like cancellation rate.
54 00:08:59.226 ⇒ 00:09:16.789 Uttam Kumaran: And then also you mentioned a couple of dimensions which is like which macros do they use? So we’ll try to build a couple of these like base tables, which is like the users, the customers, the macros. And then, if you could tell us a little bit of insight into like the dash like the dashboard, but more specifically like what metrics and what
55 00:09:16.790 ⇒ 00:09:31.750 Uttam Kumaran: dimensions you might want to see. For example, I want to see cancellation rate. Let’s see total tickets, like handled or total live tickets, and then also dimensions, which is like the people. If the people organize into any categories, products, things like that.
56 00:09:31.990 ⇒ 00:09:35.710 Aman Nagpal: When you say dashboard, you mean the end result visualization.
57 00:09:35.710 ⇒ 00:09:51.069 Uttam Kumaran: Kind of yeah, and maybe less about like this is a line charts, a bar chart more like, Hey, I want to see cancellation. I want to see Macros by people. Or I want to see this metric by this dimension. That way we can structure the the data model that way.
58 00:09:52.070 ⇒ 00:09:52.660 Uttam Kumaran: So.
59 00:09:52.660 ⇒ 00:10:09.400 Nicolas Sucari: I just tag you in a message a man sent a couple of days ago where he gave us some like dimensions and some questions that he’d like to answer. With this gorgeous date, I think with that we can start working. Once we integrated the data that we have
60 00:10:09.894 ⇒ 00:10:31.490 Nicolas Sucari: on gorgeous. And see from the from that erd you shared with them. That’s the one the same link that I shared for the dogs. We can grab all the information that we have there, and see if we can answer all of those questions. If there is some data missing, we then can take a look into other ways of bringing that information from gorgeous, but I think we will mostly have everything there.
61 00:10:32.590 ⇒ 00:10:33.080 Uttam Kumaran: Cool.
62 00:10:33.080 ⇒ 00:10:49.349 Aman Nagpal: That works. And I guess it’s the same question that you know we, I ask a lot is, What is where is this end result going right? So are we using aptitude for this or not? Because, you know, that was the plan for most of these things. And that’s where a lot of our data is going. For example, all the recharge
63 00:10:49.847 ⇒ 00:10:52.520 Aman Nagpal: cancel events and skip events and things like that.
64 00:10:52.817 ⇒ 00:11:00.500 Aman Nagpal: Is that where we’re planning on sending all this data, or is there a different solution? And that’s something I’ll have to, you know. See what Justin wants to do. Also.
65 00:11:01.640 ⇒ 00:11:27.419 Nicolas Sucari: I think. Well, Robert can jump in here, but I think the idea is to have all of the dashboards in Meta Base so that everyone can go to one to one place and find all of the information we are gonna use real also, for now to explore a little bit on that data and ask and and answer any of those questions. But yeah, I think like the end place to have those dashboards is metabase. For now.
66 00:11:28.130 ⇒ 00:11:35.850 Aman Nagpal: Got it. And if we decide, and I, I do think this should be discussed on the call today with Justin. It’s an important point that hey, this, this and this
67 00:11:35.890 ⇒ 00:11:42.990 Aman Nagpal: we’re doing in Meta Base. We’re doing in real whatever it is, right real. I know you said, is is kind of like a prelim preliminary step, so maybe not real, but
68 00:11:44.280 ⇒ 00:11:48.709 Aman Nagpal: And just to make sure that he’s on the same page. And if for some reason we’re not.
69 00:11:48.760 ⇒ 00:11:53.739 Aman Nagpal: Then we would need to figure out how we’re going to get that data into amplitude right? Which shouldn’t be an issue. I think.
70 00:11:55.520 ⇒ 00:11:57.089 Nicolas Sucari: Yeah, I think.
71 00:11:57.160 ⇒ 00:12:05.590 Nicolas Sucari: yeah, I I don’t. I don’t. I’m not sure about bringing that data into amplitude. But I think there should be a way right. I guess maybe you can help us. There.
72 00:12:07.772 ⇒ 00:12:14.020 Payas Parab: I think I think it’s like bringing it in. I think it’s like bringing the key information from amplitude into
73 00:12:14.290 ⇒ 00:12:25.199 Payas Parab: Meta Base then, or like into the data warehouse. So then you can do joins across. It is like the ideal state. I do need to like, confirm, like that exact integration. How that would work for us. I
74 00:12:26.620 ⇒ 00:12:31.440 Payas Parab: I will need Robert. Maybe we can discuss that after this call real quick, just to see
75 00:12:31.490 ⇒ 00:12:32.989 Payas Parab: how we might do that.
76 00:12:32.990 ⇒ 00:12:42.039 Robert Tseng: Okay, yeah. I mean, I don’t feel like we need to make this decision now. But I mean, I think I want to just backtrack a bit. We did say when we started this.
77 00:12:42.490 ⇒ 00:13:04.120 Robert Tseng: well, let’s kind of go back to the before and after before we started this whole data. Engineering like work stream. All of analytics was flowing through amplitude, and when Pius and I were running the analysis there, we ran into walls with certain things, especially with financial analysis. And so we propose this data warehouse solution, plus being able to have more control of your data
78 00:13:04.120 ⇒ 00:13:25.809 Robert Tseng: so that we could trim down the scope of what’s going into amplitude, and there may come a point to when amplitude is no longer needed. But that doesn’t have to happen like now, I think. We’re moving different things out of amplitude, like the financial reporting, which is why that’s in Meta Base now. And that’s we don’t really want Jared kind of poking around an amplitude for that data anymore.
79 00:13:26.300 ⇒ 00:13:30.169 Robert Tseng: But until, like, we can replace, like the other
80 00:13:30.550 ⇒ 00:13:37.169 Robert Tseng: reports that are in amplitude, which there are some that we can’t replace right now, like we’re not ready to replace the
81 00:13:37.230 ⇒ 00:13:55.128 Robert Tseng: if we wanted to look at any reporting on user engagement on on our pdps, and some of the funnel reports that already exist like the retention dashboard I built like that’s that’s gonna stay there until and I don’t necessarily think it’s a priority to replace it. If that works. So
82 00:13:55.570 ⇒ 00:14:16.900 Robert Tseng: yeah, I think that’s you know, we, we can start to have those converse that conversation today with the rest with with Justin them, and kind of drawing the lines of like where certain types of reporting are going to live. But I also don’t want to go to Justin and be like, yeah, we’re switching off amplitude because I think he would not react well to that.
83 00:14:17.460 ⇒ 00:14:36.080 Aman Nagpal: Yeah, I think, you hit the nail on head. That’s exactly we’re on the same page, you know, we we decided, hey, maybe a bunch of this financial data will move over to Meta Base for something else, and then the rest of the stuff will still, for now live in amplitude. And we’ll, you know, Edit, whatever additional data sources that we’re talking about, I think ideally, if we, if it
84 00:14:36.180 ⇒ 00:14:38.130 Aman Nagpal: does still go into, if it
85 00:14:38.540 ⇒ 00:14:49.190 Aman Nagpal: if gorgeous, for example, goes into amplitude. That would be helpful to have there right now and then, maybe down the road we can discuss elsewhere. But I just wanna make sure.
86 00:14:49.280 ⇒ 00:15:03.669 Aman Nagpal: you know. But it sounds like we’re on the same page with that. So I I am curious. You know the way it is. Now we send an event, for example, ticket created to amplitude. We can’t change it later. So when that ticket is updated, when it’s
87 00:15:04.230 ⇒ 00:15:07.310 Aman Nagpal: closed when there’s a macro used that
88 00:15:07.490 ⇒ 00:15:13.279 Aman Nagpal: initial event. Unless we’re firing more events is not firing into amplitude with this method.
89 00:15:13.340 ⇒ 00:15:22.079 Aman Nagpal: how would that be different? Since, you know, before it was immutable, we’re sending it into amplitude. Now it goes into the data warehouse. The data warehouse I’m assuming is
90 00:15:22.140 ⇒ 00:15:36.709 Aman Nagpal: syncing live updates, or would sync live updates to amplitude for something like this? So what would that process look like where a ticket is created then updated, then closed? And is that just change updating in amplitude.
91 00:15:38.440 ⇒ 00:15:56.620 Robert Tseng: Yeah. So I think with that example, I mean, I could take it. Well, basically, you’re right. I think we would have the. We would have the change logs in the data Warehouse. And so if there is an update, then we would. We can make it so that it overrides the existing record in Snowflake. And then that gets automatically pulled through. Amplitude.
92 00:15:56.889 ⇒ 00:16:11.460 Robert Tseng: Or we just like stored as a replica. So you just have, like the we. You have the history of it, and it kind of depends on the on the specific use case that that we’re talking through. But yeah, I think, to your point, like, yes, things are no longer.
93 00:16:11.460 ⇒ 00:16:19.939 Robert Tseng: They don’t have to remain immutable. And let’s say we want to go back. And we saw like there was like a 1 week gap where data was going. Wonky.
94 00:16:20.000 ⇒ 00:16:26.499 Robert Tseng: Then, if we needed to run a query, pull all of the records from a a week where there have been like
95 00:16:26.847 ⇒ 00:16:36.730 Robert Tseng: Where there was bad data, and we needed to update it. We could actually do that now. And so I think that’s that’s the that’s the kind of flexibility that we have with the system that we’d set up.
96 00:16:37.640 ⇒ 00:17:04.190 Aman Nagpal: That works. So you know, if a ticket, once the ticket is created, it hits amplitude, and it tells us that this ticket was created. This is the information for the ticket. This is the agent that was automatically assigned to the ticket, and then the next. And then we have let’s say, 2 or 3 dashboards or chart reports using that ticket. Now, the next day that ticket is closed by the agent. That event, I assume in amplitude
97 00:17:04.300 ⇒ 00:17:05.470 Aman Nagpal: whether
98 00:17:05.480 ⇒ 00:17:11.450 Aman Nagpal: I guess that existing event will get updated on those charts rather than a duplicate event? Or is that what you mean?
99 00:17:11.730 ⇒ 00:17:24.730 Robert Tseng: Yeah, so we can either go back and rewrite the old one, or we just fire a new event. That’s like close event or something I don’t know, like, I’m just kind of spitballing, but like. And then that would that would prompt the or that would trigger like the update where
100 00:17:25.130 ⇒ 00:17:28.070 Robert Tseng: at the user level, or the whoever the
101 00:17:28.290 ⇒ 00:17:37.480 Robert Tseng: at at the ticket level. Or that would that the status, could. It could be a status change, you know. So yeah, I think there’s a there’s a couple of ways to go and and and do that.
102 00:17:37.890 ⇒ 00:17:43.293 Aman Nagpal: Okay, I think that’ll probably ideal. But I think that’s definitely something to discuss.
103 00:17:43.990 ⇒ 00:17:46.350 Aman Nagpal: you know, once ticket is closed.
104 00:17:46.390 ⇒ 00:17:53.829 Aman Nagpal: I feel like it’s reopened a lot of time also. So maybe instead of firing an open and closed event that might not fulfill
105 00:17:54.163 ⇒ 00:18:06.889 Aman Nagpal: or maybe it will right that I guess that’s something. We just have to go over the different scenarios. But either way, it sounds like we’ll make it work, no matter how we end up doing it? So that’s good. And so it looks like we’re good with gorgeous there. When do you guys think
106 00:18:07.050 ⇒ 00:18:11.580 Aman Nagpal: that could be wrapped up and completed by. So if today’s Tuesday.
107 00:18:11.800 ⇒ 00:18:16.999 Uttam Kumaran: Can I am on while you’re in gorgeous. Can I also get your username Api key?
108 00:18:17.348 ⇒ 00:18:24.040 Uttam Kumaran: And we can set that up? I think, also on this call, Nico. Let’s see if we can set up Okendo
109 00:18:24.170 ⇒ 00:18:28.623 Uttam Kumaran: cause it’s there’s basically yeah, we just don’t. We don’t have to wait on that
110 00:18:29.030 ⇒ 00:18:37.349 Uttam Kumaran: And then let’s see, I think I wanna we got. Let’s take on Nico like a plan for the stuff that I’m on sent before those metrics.
111 00:18:37.598 ⇒ 00:18:45.939 Uttam Kumaran: And then chat with Ryan and see what the scope is. I know we’re still working on some stuff on the finance side, so let’s have a little bit of plan on timeline.
112 00:18:46.930 ⇒ 00:18:47.670 Uttam Kumaran: Yeah.
113 00:18:48.700 ⇒ 00:18:52.499 Aman Nagpal: So do you know, if each user has their own Api key in gorgeous or not.
114 00:18:52.898 ⇒ 00:18:56.479 Uttam Kumaran: That’s my understanding is like your username will be.
115 00:18:56.690 ⇒ 00:18:59.120 Uttam Kumaran: It should be like, probably your email.
116 00:18:59.180 ⇒ 00:19:01.779 Uttam Kumaran: or like whatever emails on the account. And then.
117 00:19:01.970 ⇒ 00:19:05.669 Uttam Kumaran: under Api access, you’ll see that there is an Api key.
118 00:19:06.807 ⇒ 00:19:08.269 Uttam Kumaran: I believe that
119 00:19:08.300 ⇒ 00:19:12.969 Uttam Kumaran: I don’t actually know. If every user has one. It might be just for the entire account. But
120 00:19:14.378 ⇒ 00:19:15.930 Uttam Kumaran: do you see one in there?
121 00:19:16.310 ⇒ 00:19:29.600 Aman Nagpal: Yeah, when I go to rest. Api, it gave me a button to click, create Api key. And then it says, username is my email password is the Api key. My only concern with this is, I don’t have the credentials handy of, like our main admin
122 00:19:29.810 ⇒ 00:19:31.280 Aman Nagpal: a general account.
123 00:19:31.685 ⇒ 00:19:36.274 Aman Nagpal: So this, I guess, would always be attached to my personal, gorgeous account for right.
124 00:19:36.580 ⇒ 00:19:37.930 Uttam Kumaran: That’s correct.
125 00:19:37.930 ⇒ 00:19:38.730 Aman Nagpal: With that.
126 00:19:39.161 ⇒ 00:19:53.489 Uttam Kumaran: No, especially because you’re generating it now. And there isn’t any like. They don’t have any sort of thing around pulling from the Api. There shouldn’t be any issue. Also, what we’ll like, basically what what we’ll see is when we hook it up.
127 00:19:53.940 ⇒ 00:19:55.020 Uttam Kumaran: As long as
128 00:19:55.110 ⇒ 00:20:02.819 Uttam Kumaran: you have access to all the data. We should be fine if, in case we need to say, Hey, you need some like broader access.
129 00:20:02.940 ⇒ 00:20:04.549 Uttam Kumaran: we’ll find out pretty quickly.
130 00:20:05.710 ⇒ 00:20:16.300 Aman Nagpal: Okay. So I guess, if anything were to happen, and it comes to it where we need to switch the Api key to the general accounts. Api key. It’ll it won’t be too
131 00:20:16.380 ⇒ 00:20:21.189 Aman Nagpal: cumbersome of a process to do that right. It’ll just know what is already there and just update the routes.
132 00:20:21.190 ⇒ 00:20:37.869 Uttam Kumaran: Yeah. And we’re purely just reading data. So as long as your if your account has access to everything, then we’ll find that out again. Worst case. If it doesn’t, then we can. We can switch. I think this will just give us enough to start to run towards those metrics. And again, we can sub out the data later, if we need to.
133 00:20:38.800 ⇒ 00:20:41.649 Aman Nagpal: Cool, do you? How do you want me to send this to you?
134 00:20:42.226 ⇒ 00:20:47.909 Uttam Kumaran: If you can, if you’re okay with slacking it. Otherwise I can give.
135 00:20:47.910 ⇒ 00:20:50.370 Aman Nagpal: Let me just back it, and then, once you have it, I can delete it. Does that work.
136 00:20:50.370 ⇒ 00:20:51.430 Uttam Kumaran: Okay, that’s fine.
137 00:20:51.690 ⇒ 00:20:53.130 Aman Nagpal: Here’s the key right here.
138 00:21:01.164 ⇒ 00:21:09.109 Uttam Kumaran: And then, Nico, do you want to try to do Kendo or do? I know we only minutes, or if there’s any other items. We can go through that.
139 00:21:10.820 ⇒ 00:21:13.159 Nicolas Sucari: Are you setting up? Gorgeous? I think 5 already.
140 00:21:13.160 ⇒ 00:21:14.779 Uttam Kumaran: Yeah, I’m gonna do it right now.
141 00:21:17.200 ⇒ 00:21:17.920 Nicolas Sucari: Okay.
142 00:21:18.070 ⇒ 00:21:18.590 Uttam Kumaran: Take the okay
143 00:21:19.580 ⇒ 00:21:20.489 Uttam Kumaran: more involved. So.
144 00:21:22.250 ⇒ 00:21:23.249 Nicolas Sucari: Broken, the wee.
145 00:21:23.250 ⇒ 00:21:26.679 Aman Nagpal: Email, and then just let me know whenever I can delete that feed from slack.
146 00:21:26.680 ⇒ 00:21:30.229 Uttam Kumaran: Yeah, let me just turn it on, and then confirm.
147 00:21:30.510 ⇒ 00:21:31.459 Aman Nagpal: You got it.
148 00:21:32.720 ⇒ 00:21:33.300 Aman Nagpal: You’re okay.
149 00:21:33.300 ⇒ 00:21:35.060 Nicolas Sucari: We need the user Id
150 00:21:35.120 ⇒ 00:21:37.289 Nicolas Sucari: and the recipe key. Also
151 00:21:39.720 ⇒ 00:21:43.220 Nicolas Sucari: in credential section, I think in settings, integrations.
152 00:21:44.850 ⇒ 00:21:47.200 Aman Nagpal: Integrations, credentials
153 00:21:49.380 ⇒ 00:21:51.710 Aman Nagpal: perfect. So I have the Id.
154 00:21:53.290 ⇒ 00:21:55.349 Aman Nagpal: Let’s see that here
155 00:21:57.790 ⇒ 00:22:00.759 Aman Nagpal: and then I have the merchant rest. Api key as well.
156 00:22:00.920 ⇒ 00:22:04.219 Uttam Kumaran: Okay, Nico. I can go ahead, and I’ll do the okendo one, too. Just.
157 00:22:04.220 ⇒ 00:22:04.890 Nicolas Sucari: Okay.
158 00:22:05.975 ⇒ 00:22:06.559 Uttam Kumaran: And then, yeah.
159 00:22:06.560 ⇒ 00:22:07.499 Aman Nagpal: Have my account.
160 00:22:07.500 ⇒ 00:22:08.949 Uttam Kumaran: You can delete the other.
161 00:22:09.610 ⇒ 00:22:11.380 Uttam Kumaran: Api key.
162 00:22:15.860 ⇒ 00:22:17.910 Aman Nagpal: Is there a light connector for Oquendo as well.
163 00:22:17.910 ⇒ 00:22:18.610 Uttam Kumaran: Yes.
164 00:22:19.280 ⇒ 00:22:19.940 Nicolas Sucari: Yes.
165 00:22:50.790 ⇒ 00:22:51.960 Uttam Kumaran: Okay?
166 00:23:03.580 ⇒ 00:23:07.930 Uttam Kumaran: And then, Nico, do we already have requirements for Oquendo.
167 00:23:08.450 ⇒ 00:23:09.080 Nicolas Sucari: No.
168 00:23:09.390 ⇒ 00:23:15.529 Nicolas Sucari: we don’t. But yeah, I don’t know, Aman. If you already have something in mind that you’d like to
169 00:23:15.590 ⇒ 00:23:18.739 Nicolas Sucari: to see from Okando, let us know. Yeah.
170 00:23:19.240 ⇒ 00:23:21.240 Aman Nagpal: I think Okendo
171 00:23:21.500 ⇒ 00:23:24.343 Aman Nagpal: should be a bit simpler than gorgeous.
172 00:23:25.480 ⇒ 00:23:28.679 Aman Nagpal: you know, we need all of the reviews and ratings
173 00:23:28.760 ⇒ 00:23:30.360 Aman Nagpal: synced over
174 00:23:31.034 ⇒ 00:23:44.409 Aman Nagpal: right now we’re doing that with a cloudflare worker that syncs 4 times a day. So would love to turn that off and just rely on this. We already have some reports built out within amplitude. So maybe we could just throw that data in there.
175 00:23:45.392 ⇒ 00:23:48.120 Aman Nagpal: But yeah, reviews, ratings.
176 00:23:49.790 ⇒ 00:23:56.960 Aman Nagpal: what else? We don’t really do surveys on here, so we don’t need that. We don’t do quizzes on here. No loyalty.
177 00:24:02.440 ⇒ 00:24:08.129 Aman Nagpal: yeah, it’s it’s really just a review and ratings. And in terms of what products
178 00:24:08.560 ⇒ 00:24:14.120 Aman Nagpal: they’re attached to, I guess that would be good. So one thing Okendo does is
179 00:24:14.390 ⇒ 00:24:31.550 Aman Nagpal: each review is attached to a specific product which I’m assuming is by product. Id, so that information should already be there. So hopefully, that’s easy. How we display things on the front end is, we’ll group a bunch of products. So, for example, if we have
180 00:24:31.730 ⇒ 00:24:34.210 Aman Nagpal: 5 different protein coffee flavors.
181 00:24:34.643 ⇒ 00:24:48.439 Aman Nagpal: each review is associated with with one of those specific products of that flavor. But maybe on the front end, when we want to display the actual products, we’ll group all of we’ll create a group in Okendo with all of the different products.
182 00:24:48.480 ⇒ 00:25:02.573 Aman Nagpal: flavors in one group. And then when we display it on the website, it just shows reviews for all of the different flavors together. But that aspect of it. I don’t know that that’s something that would need to be pulled into the data warehouse.
183 00:25:03.040 ⇒ 00:25:06.760 Aman Nagpal: unless you guys think of a better reason. I feel like it’s just the reviews and ratings.
184 00:25:08.050 ⇒ 00:25:10.520 Uttam Kumaran: Yeah, we have the products, the orders
185 00:25:10.680 ⇒ 00:25:12.809 Uttam Kumaran: they’re associated with the reviews.
186 00:25:13.280 ⇒ 00:25:15.630 Uttam Kumaran: There’s some attributes about
187 00:25:16.110 ⇒ 00:25:18.380 Uttam Kumaran: the reviewer.
188 00:25:18.906 ⇒ 00:25:22.369 Uttam Kumaran: But of course, anything on the on the rating side.
189 00:25:22.390 ⇒ 00:25:26.470 Uttam Kumaran: But the review object seems to have basically everything
190 00:25:27.860 ⇒ 00:25:28.910 Uttam Kumaran: we need.
191 00:25:30.750 ⇒ 00:25:38.570 Nicolas Sucari: Yeah, I don’t know about the groups, but we can see about that in the product Ids. And if not, we, you can tell us which product ids you’re grouping, and we can
192 00:25:38.660 ⇒ 00:25:40.770 Nicolas Sucari: try to do a workaround in that one. Yep.
193 00:25:40.770 ⇒ 00:25:41.400 Uttam Kumaran: Yeah.
194 00:25:42.260 ⇒ 00:25:43.079 Aman Nagpal: That sounds good. Yeah.
195 00:25:43.080 ⇒ 00:25:49.950 Uttam Kumaran: That’ll be easy as fixed. Then we can just we can just create those and have those in the logic pretty easily.
196 00:25:50.800 ⇒ 00:25:53.118 Aman Nagpal: Yeah, that works for now the
197 00:25:53.700 ⇒ 00:26:00.780 Aman Nagpal: it gives you the you were saying order, and maybe the user info. So I guess it would be just tie it associated with
198 00:26:00.890 ⇒ 00:26:02.999 Aman Nagpal: our other tables. Maybe.
199 00:26:03.050 ⇒ 00:26:09.500 Aman Nagpal: you know, maybe find what percent of I’m just gonna throw random stuff, what percent of
200 00:26:09.980 ⇒ 00:26:16.780 Aman Nagpal: shopify buyers who buy protein coffee. Leave review, for example, right? The fact that it’s all associated will allow us to do that.
201 00:26:16.780 ⇒ 00:26:21.730 Uttam Kumaran: Yeah, Nico, we could add the reviews to the orders. Dash.
202 00:26:21.730 ⇒ 00:26:22.380 Nicolas Sucari: Yeah, of course.
203 00:26:22.680 ⇒ 00:26:23.879 Uttam Kumaran: That way, I think.
204 00:26:24.430 ⇒ 00:26:49.930 Uttam Kumaran: either, like we added to every single order, like a review associated with the order. And then also, if we have a products table, we can add like average and stuff like that there, and then what we’ll do is we’ll also have like a review specific table where we can join in. You’ll see, like all the reviews, you can see the product details associated with it. The customer details associated, the order details associated with it. And then you can just kind of go wild. Basically.
205 00:26:50.630 ⇒ 00:27:09.929 Aman Nagpal: That’s perfect. 2 other things that just popped into my head. We do have a status for each review. It’s pending whether which means have we moderated, moderated it yet, or approved, or rejected, or published, or rejected? I should say so. We will want all of the reviews synced, whether it’s no matter what the status is. But
206 00:27:09.950 ⇒ 00:27:12.540 Aman Nagpal: for some of the dashboards
207 00:27:12.850 ⇒ 00:27:14.589 Aman Nagpal: we probably
208 00:27:14.720 ⇒ 00:27:29.359 Aman Nagpal: would want. I think, which is what we do now is, if we go to. You know the product review? Event in amplitude we can filter by. Is this the status of that review? So is it published? Is it rejected. Is it pending
209 00:27:30.670 ⇒ 00:27:34.970 Aman Nagpal: so that it would be good to make sure we have that capability and.
210 00:27:34.970 ⇒ 00:27:38.230 Nicolas Sucari: I think it’s in the review. Yeah.
211 00:27:38.230 ⇒ 00:27:39.229 Aman Nagpal: In the data.
212 00:27:39.680 ⇒ 00:27:41.100 Uttam Kumaran: Yeah, cool.
213 00:27:41.346 ⇒ 00:27:53.940 Aman Nagpal: And the other thing is sentiment. I think positive or negative. Oquendo may automatically be applying that to each ticket, so I assume that’s in there, too. I don’t know that we use that for now, but if it’s in there, then great, maybe we can use it down the road.
214 00:27:54.780 ⇒ 00:28:04.270 Nicolas Sucari: Yes, it is. It’s sentiment, Field. We have it there, the status, too. So once we have all of the data, we can start poking around and see how we can create all of those tables for you. Yeah.
215 00:28:05.140 ⇒ 00:28:06.070 Aman Nagpal: Sounds good.
216 00:28:06.730 ⇒ 00:28:11.299 Uttam Kumaran: You guys? Just curious like, how’d you guys decide on O Kendo versus like Yaho or
217 00:28:12.810 ⇒ 00:28:15.060 Uttam Kumaran: any of the other like review platforms.
218 00:28:15.340 ⇒ 00:28:19.720 Aman Nagpal: Honestly, not slightest clue. We had okendo when I joined we talked about
219 00:28:19.790 ⇒ 00:28:31.660 Aman Nagpal: maybe using yatbo for their. They have a Tiktok shop integration which allows you to sync reviews there. I think we tried it, but didn’t really go through with it. But yeah, just been on no kendo
220 00:28:32.190 ⇒ 00:28:32.960 Aman Nagpal: cool.
221 00:28:33.530 ⇒ 00:28:38.169 Aman Nagpal: I if I had to guess. I bet Yadpo’s still a lot more expensive. That would be my guess.
222 00:28:38.650 ⇒ 00:28:46.079 Uttam Kumaran: Yeah, that’s what I’ve I’ve seen a lot of people use yak, though. But yeah, it’s I think that their pricing is like pretty pretty huge.
223 00:28:46.370 ⇒ 00:28:47.090 Aman Nagpal: Yeah.
224 00:28:47.450 ⇒ 00:28:55.656 Uttam Kumaran: Yeah, it’s basically 700 for a thousand orders. And then it’s like, if you’re doing anything like, yeah, it just goes crazy pretty much after that.
225 00:28:55.960 ⇒ 00:28:57.009 Aman Nagpal: 700 bucks.
226 00:28:58.183 ⇒ 00:29:02.660 Uttam Kumaran: For a thousand. But if your order volume is like 10,000,
227 00:29:02.980 ⇒ 00:29:08.170 Uttam Kumaran: yeah, I mean, it’s just if you’re doing any sort of numbers. It’s just like gonna be brutal.
228 00:29:08.490 ⇒ 00:29:12.969 Aman Nagpal: All these guys are are these man? I mean, even even our okendo.
229 00:29:13.400 ⇒ 00:29:40.779 Aman Nagpal: we had to negotiate with them to give us a ton extra buffer, because they’re counting all of our renewals, all of our Tiktok shop orders that are still in shopify as new orders within their billing system which makes no sense. So you know, if we have, let’s say, 70 k. New orders in one month. We told them to give us. I think it was like 120 or 150 K. Orders for the month in terms of billing, so that we can compensate for all those extra orders that we shouldn’t be getting charged for, anyway.
230 00:29:40.780 ⇒ 00:29:41.810 Uttam Kumaran: Oh, okay.
231 00:29:41.900 ⇒ 00:29:51.230 Uttam Kumaran: yeah. I mean, I think a lot of these companies like Yapto and these guys, they run out of product innovation. And then they just jack the price, because they’re all like Vc. Backed.
232 00:29:51.510 ⇒ 00:29:52.590 Uttam Kumaran: So
233 00:29:52.610 ⇒ 00:29:54.420 Uttam Kumaran: it’s kind of sad. But
234 00:29:54.949 ⇒ 00:30:03.875 Uttam Kumaran: you know, I think I think hopefully, like a lot of the stuff that we’re doing gives a lot more value to that data. So it’s not just like to functionality, but actually get to like use it.
235 00:30:04.390 ⇒ 00:30:05.216 Uttam Kumaran: and then
236 00:30:05.820 ⇒ 00:30:16.639 Uttam Kumaran: I don’t know. But also the other thing I mean that we’d even do internally is we shop? We try to shop a lot so like these these guys, you go to Yaho, and they may give you like, 50% or 70% off.
237 00:30:16.780 ⇒ 00:30:21.529 Uttam Kumaran: You just kind of have to PIN these guys on each other a little bit, but
238 00:30:21.640 ⇒ 00:30:22.910 Uttam Kumaran: it takes work.
239 00:30:23.600 ⇒ 00:30:33.679 Aman Nagpal: It. It’s funny. When I was I forgot. If I mentioned I was at Obvi last, and those guys get everything for free, just because, you know, people are obsessed with their them on twitter and stuff like that.
240 00:30:34.850 ⇒ 00:30:52.850 Aman Nagpal: You know that software that would cost thousands of dollars. It’s just, oh, yeah, completely free, you know. Let’s let us just put your logo on our site. And now it’s a little bit different where these guys aren’t crazy about the whole personal branding aspect. So I’m always trying to negotiate and say, you know, give give us a break on what we’re already using, and we’ll switch.
241 00:30:53.330 ⇒ 00:30:57.870 Uttam Kumaran: Yeah, I I think. Yes, the whatever price they have on the page is never
242 00:30:57.990 ⇒ 00:31:05.649 Uttam Kumaran: the actual price. You just kind of have to finesse. But it takes time. And I hate talking to the the like Aes at these like software companies.
243 00:31:06.030 ⇒ 00:31:09.630 Uttam Kumaran: It’s the worst like they’re just very annoying. But.
244 00:31:09.630 ⇒ 00:31:11.139 Aman Nagpal: Let’s do it. Never stop!
245 00:31:11.390 ⇒ 00:31:23.980 Uttam Kumaran: I know. But dude, there’s 10. There’s 10 types of companies that probably do this, and you just go to the earliest one who needs the business the most, and then they give you the hat, who maybe, and then maybe you look at how much they raise and then give you the best discount.
246 00:31:24.753 ⇒ 00:31:25.939 Uttam Kumaran: Go ahead.
247 00:31:27.070 ⇒ 00:31:29.930 Aman Nagpal: It’s it’s it’s fun dealing with all this.
248 00:31:32.100 ⇒ 00:31:34.495 Uttam Kumaran: Cool. So both of those are syncing
249 00:31:35.190 ⇒ 00:31:40.229 Uttam Kumaran: and then we’ll get back to you on like timeline. Nico. Anything else on our side.
250 00:31:41.820 ⇒ 00:31:45.419 Nicolas Sucari: I don’t think so. No, we’re still working on
251 00:31:45.480 ⇒ 00:31:50.490 Nicolas Sucari: creating all of those dashboards in real. So I’ll let you know when that is ready so that you can
252 00:31:50.690 ⇒ 00:31:52.289 Nicolas Sucari: don’t poke around it. No.
253 00:31:52.290 ⇒ 00:31:53.920 Uttam Kumaran: Right after this week.
254 00:31:53.920 ⇒ 00:31:54.660 Nicolas Sucari: Research.
255 00:31:54.910 ⇒ 00:31:57.119 Uttam Kumaran: There is some that we added right.
256 00:31:58.020 ⇒ 00:32:00.520 Nicolas Sucari: Yeah, we added, we added one more table.
257 00:32:00.788 ⇒ 00:32:04.400 Nicolas Sucari: But yeah, I’ll I’ll share it through slack so that you can access there.
258 00:32:04.741 ⇒ 00:32:09.379 Nicolas Sucari: And yeah, we are. We created one, just for shopify that we are using
259 00:32:09.826 ⇒ 00:32:13.810 Nicolas Sucari: so that you can take a look directly all of shopify orders there.
260 00:32:13.850 ⇒ 00:32:29.860 Nicolas Sucari: And yeah, I’ll keep adding stuff there and let you know a man so that you can go there and look around. I know maybe Jared Justin and Jared don’t. Don’t want to go very to look at the data, but we are validating and matching, matching everything to metabase
261 00:32:30.187 ⇒ 00:32:40.680 Nicolas Sucari: just to have all of the same information, the same data between both, so that if you want to go there and answer quick questions, you can go there and serve yourself. Okay.
262 00:32:41.560 ⇒ 00:32:42.630 Aman Nagpal: That sounds great.
263 00:32:42.630 ⇒ 00:32:46.450 Uttam Kumaran: Yeah, and we actually have now all 3 platforms, basically.
264 00:32:46.842 ⇒ 00:32:51.810 Uttam Kumaran: separated out. So you can see, tick, tock, shopify, and Amazon all in one area and then
265 00:32:52.266 ⇒ 00:33:03.329 Uttam Kumaran: I think we’ll probably work on the individual dashboards if they have an increased dimensionality. And those are the core order tables. So now that we talked about okendo and gorgeous when we link
266 00:33:03.630 ⇒ 00:33:18.166 Uttam Kumaran: those platforms to the orders it just takes on that all that work. So it all kind of builds on itself for for us to simply be like cool. Join Oquendo to shopify and order Id. That’s like a 1 line thing for us now.
267 00:33:18.640 ⇒ 00:33:23.559 Uttam Kumaran: and so I’m glad. I’m glad we kind of like sorted all of the basic stuff out.
268 00:33:24.254 ⇒ 00:33:27.299 Uttam Kumaran: And we have, like now, all 3 flowing in nicely.
269 00:33:28.190 ⇒ 00:33:33.589 Aman Nagpal: Yeah, all the big stuff looks like, we got it all there. So appreciate you guys, I think
270 00:33:33.630 ⇒ 00:33:35.310 Aman Nagpal: we made a lot of progress.
271 00:33:35.530 ⇒ 00:33:41.809 Payas Parab: I had a quick question for you on just something we’re still trying to figure out like on the back end, like shopify is kind of like
272 00:33:41.940 ⇒ 00:34:07.009 Payas Parab: essentially hiding the shit like in the data like it’s really hard to find. They don’t want you to figure out how much they’re charging you. So they make it really hard, and we’re trying to tie it out to things we see in the shopify portal. When I go into the admin, I use Robert’s login. To shopify. Admin I I’m not seeing like I see there’s like a thing on the side that’s like for finance, but it’s like blurred out. And is it like we don’t have access like.
273 00:34:07.010 ⇒ 00:34:16.800 Payas Parab: I don’t know if we don’t have access to it. Or you guys haven’t activated that module, but, like in the analytics, I can’t seem to pull any ground truth of like cogs or fees
274 00:34:16.800 ⇒ 00:34:31.989 Payas Parab: for me to tie out. You know I can check the individual order, and I was able to check some of our like cogs stuff. But outside of cogs like pick and pack and those types of fees. We’re we’re having trouble finding them in the data. And I was wondering if there’s any finance
275 00:34:32.320 ⇒ 00:34:36.300 Payas Parab: views that have that fulfillment information that we can like
276 00:34:36.380 ⇒ 00:34:41.000 Payas Parab: used to see if we’re finding the right thing. Because we’re not like a hundred percent sure.
277 00:34:41.449 ⇒ 00:34:43.199 Aman Nagpal: It might be a
278 00:34:43.279 ⇒ 00:34:47.639 Aman Nagpal: financing. But some things might not be there. So what specifically
279 00:34:47.959 ⇒ 00:34:51.259 Aman Nagpal: are you looking for that? I can make sure you have.
280 00:34:51.750 ⇒ 00:35:00.479 Payas Parab: In in the Admin portal like there’s like a finances view that, like, I assume, has some additional information about cost like I can share my screen and show you real quick.
281 00:35:01.540 ⇒ 00:35:03.579 Aman Nagpal: Actually, I can’t see.
282 00:35:04.110 ⇒ 00:35:10.150 Aman Nagpal: So as balance credit payouts and bill pay. So that’s the only stuff in there which I don’t have access to either.
283 00:35:10.952 ⇒ 00:35:12.429 Aman Nagpal: But probably nothing.
284 00:35:12.430 ⇒ 00:35:15.110 Payas Parab: But like nothing related to the costs or fulfillment.
285 00:35:15.110 ⇒ 00:35:19.759 Aman Nagpal: No, so fulfillment costs. I sent over to Robert. We have
286 00:35:20.236 ⇒ 00:35:24.040 Aman Nagpal: let me see if I could just share my screen. Really quick.
287 00:35:24.040 ⇒ 00:35:25.740 Robert Tseng: I shared the file. The team. Yeah, this.
288 00:35:25.740 ⇒ 00:35:28.310 Payas Parab: Yeah, we have the file that file you shared. Yeah.
289 00:35:28.310 ⇒ 00:35:35.209 Aman Nagpal: That’s pretty much it. I actually spoke with Jared about this this morning. He’s working on some app for
290 00:35:35.536 ⇒ 00:35:53.720 Aman Nagpal: kind of long term planning in terms of cost, and I asked him maybe we could use that for ad hoc reporting for here. But his opinion was that it? It’s not necessarily what we can use. So we’ll figure out something I know, Robert mentioned. We’ll we’ll find a way to make it easy. Maybe end result is eventually a notion page where they can. Input those assumption
291 00:35:54.272 ⇒ 00:35:55.960 Aman Nagpal: but in terms of
292 00:35:56.030 ⇒ 00:36:09.940 Aman Nagpal: pick and pack costs 3 pl fees. All of that is in that file, and John Jonathan, our CEO and Jared can fill, fill you in on any other stuff. The only thing within shopify
293 00:36:10.150 ⇒ 00:36:13.280 Aman Nagpal: would be the product cost which I think you should be able to see.
294 00:36:13.816 ⇒ 00:36:18.449 Aman Nagpal: I don’t think there’s any other costs that would be in there.
295 00:36:19.870 ⇒ 00:36:22.380 Payas Parab: Product cost, there wouldn’t be anything else in there. Okay.
296 00:36:22.380 ⇒ 00:36:23.529 Aman Nagpal: Yeah, just the cogs.
297 00:36:24.470 ⇒ 00:36:27.360 Payas Parab: Got it. Okay? So we kind of have to like break down this like
298 00:36:27.500 ⇒ 00:36:36.389 Payas Parab: whatever the sources are for the the pick and pack and things like that like from that code, that source code file. We’ll have to kinda figure it out from there outside of cogs.
299 00:36:36.850 ⇒ 00:36:40.370 Aman Nagpal: Yeah, I think. When you guys are ready it would be good to just
300 00:36:40.750 ⇒ 00:36:54.170 Aman Nagpal: maybe do it. Go over in a slack chat with Jared and Jonathan. Just so that we’re not missing anything. Cause that. What I sent you is what we’ve had for so long, and it doesn’t have the latest numbers which I also sent, and there might be other changes, so.
301 00:36:54.170 ⇒ 00:36:55.359 Payas Parab: Got it. Okay.
302 00:36:56.203 ⇒ 00:37:02.199 Aman Nagpal: Also cogs I mentioned to you guys before, a lot of them are probably outdated in shopify. So
303 00:37:02.240 ⇒ 00:37:04.909 Aman Nagpal: if we go in and update those.
304 00:37:05.100 ⇒ 00:37:06.990 Aman Nagpal: We’ll be able to update everything.
305 00:37:07.340 ⇒ 00:37:09.040 Aman Nagpal: you know later on, right.
306 00:37:09.040 ⇒ 00:37:17.129 Payas Parab: Yeah, yeah, that that’s the beauty of, yeah, once you if you update the skews product cost this like whatever we’re building and that aggregates the cogs. It will like
307 00:37:17.290 ⇒ 00:37:19.420 Payas Parab: adjust for it essentially.
308 00:37:19.420 ⇒ 00:37:33.949 Aman Nagpal: Yeah. So once we do, once we go through all the products, everything’s updated for cogs, we do a refresh and all the financials. And then maybe from that point onwards, we’ll say, Hey, look the cogs if they ever change in the future for one product starting this date
309 00:37:34.010 ⇒ 00:37:38.140 Aman Nagpal: charge, or the cogs are different, but before that date the cogs was, you know.
310 00:37:38.140 ⇒ 00:37:39.999 Payas Parab: Sure. Yeah. Yeah.
311 00:37:41.520 ⇒ 00:37:42.250 Payas Parab: Yep.
312 00:37:42.250 ⇒ 00:37:48.959 Aman Nagpal: That works. One thing before I forget. I spoke to Brian about the whole Amazon shopify linking.
313 00:37:49.410 ⇒ 00:37:53.200 Aman Nagpal: I know he was trying to do like something he called Fuzzy Matching. But
314 00:37:53.340 ⇒ 00:37:56.920 Aman Nagpal: pretty much just yeah. Were you guys able to find a way to
315 00:37:56.970 ⇒ 00:38:00.680 Aman Nagpal: match up the Amazon buyers or orders to shopify.
316 00:38:03.210 ⇒ 00:38:17.109 Nicolas Sucari: Yes, I remember that one. I don’t think he was able to manage to to get a lot of users. I think we talked about something about 50 or 60 users that we were trying to match. But yeah, I’ll I’ll try to
317 00:38:17.559 ⇒ 00:38:26.240 Nicolas Sucari: bring back all of that analysis and see where where that is, and if we can continue doing it, I’ll ask Ryan to to continue working on that. Okay.
318 00:38:26.900 ⇒ 00:38:33.690 Aman Nagpal: Thank you. Yeah. If we can make that a priority as well on the data side, just I would think it’s at least a few 1,000, you know, if we look at
319 00:38:34.220 ⇒ 00:38:36.420 Aman Nagpal: email, we don’t have addresses.
320 00:38:36.420 ⇒ 00:38:36.970 Nicolas Sucari: No.
321 00:38:36.970 ⇒ 00:38:47.030 Aman Nagpal: Maybe they wrote St. On one side street on another side. Maybe there’s an apartment all of that. I think that there, you know it’s not easy, but there should be a way to match it up. Still.
322 00:38:47.807 ⇒ 00:38:53.880 Aman Nagpal: That’s, you know, just address alone, and maybe keeping name in mind that customer’s name in mind.
323 00:38:53.910 ⇒ 00:38:57.520 Aman Nagpal: There should be a way to get. At least I would think a few 1,000. I would hope.
324 00:38:57.520 ⇒ 00:39:02.229 Uttam Kumaran: And then, Aman, where does that? What’s the activation for that data like, where
325 00:39:02.240 ⇒ 00:39:04.579 Uttam Kumaran: are you guys? Gonna take that list in the retarget.
326 00:39:05.010 ⇒ 00:39:12.719 Aman Nagpal: Exactly. Yeah. So we’ll throw. We’ll create a list in Klaviyo for buyers that have purchased on Amazon since we don’t have their emails
327 00:39:12.750 ⇒ 00:39:14.549 Aman Nagpal: and then just market them that way.
328 00:39:17.050 ⇒ 00:39:17.750 Uttam Kumaran: Okay.
329 00:39:17.910 ⇒ 00:39:19.170 Uttam Kumaran: yeah, let’s see it.
330 00:39:20.430 ⇒ 00:39:21.450 Uttam Kumaran: it’s tough.
331 00:39:21.600 ⇒ 00:39:31.729 Uttam Kumaran: because sometimes if it’s Amazon, they don’t give us anything. If they if they’ve done a refund or something we can try to match, because the original order is on Amazon.
332 00:39:32.154 ⇒ 00:39:46.155 Uttam Kumaran: But now that we also have, like the customer review data and the gorgeous data, if they if they submitted a ticket, and it’s like linked back to their Amazon order. That’s we did this for another client where we match there because all the ticketing is done through you guys.
333 00:39:46.450 ⇒ 00:39:53.199 Uttam Kumaran: So we’ll have to get a little creative. Yeah, maybe, Nico, it’s probably have some data model. That’s literally just like
334 00:39:53.350 ⇒ 00:39:55.989 Uttam Kumaran: Amazon shopify customer match.
335 00:39:56.180 ⇒ 00:40:02.869 Uttam Kumaran: And we just can put all this fuzzy mass logic. And then, as we add more sources of data, we can just improve that
336 00:40:03.568 ⇒ 00:40:06.161 Uttam Kumaran: it’s probably something like that.
337 00:40:07.570 ⇒ 00:40:10.320 Aman Nagpal: Yeah, that gorgeous idea. Cool. That would be
338 00:40:10.460 ⇒ 00:40:13.569 Aman Nagpal: great to see if we can do it that way. But yeah, even with just
339 00:40:13.580 ⇒ 00:40:17.399 Aman Nagpal: some sort of logic of matching names and addresses.
340 00:40:18.014 ⇒ 00:40:22.939 Aman Nagpal: We should be able to come up with a bunch more hopefully. So just please keep me posted on that.
341 00:40:24.650 ⇒ 00:40:25.310 Aman Nagpal: Okay?
342 00:40:27.446 ⇒ 00:40:32.710 Aman Nagpal: I think that’s the main stuff I had on mind for the data side.
343 00:40:33.081 ⇒ 00:40:37.459 Aman Nagpal: and then you guys will be able to get me timelines. You think later today.
344 00:40:40.410 ⇒ 00:40:48.099 Nicolas Sucari: Yeah, I’m gonna I’m gonna prepare something and try to send you with all of these tasks and see when we can accomplish all of that. Okay.
345 00:40:48.760 ⇒ 00:41:08.340 Aman Nagpal: Thank you. And then Pius and Robert on your guys side, I know you had some discussions with Jared on the gross margin stuff and the tools we were using. I think, Robert, you got on a call with Jared. So for today’s today’s call, later, are we going to kind of do a run through of the different tools? Or what? What do you guys have planned.
346 00:41:08.340 ⇒ 00:41:37.729 Robert Tseng: Main thing. I’m putting together a deck. I’ll share it out soon. It’ll just be a few slides just to kind of walk through, just like overview of, like what we’ve done. We’re gonna summarize like, I mean, I know those guys don’t see the work that we’re doing on the on the data engineering side, but especially Snowflake. Now that Justin is you know, wants to lock. I don’t. I still don’t know if he actually walked in. But we will do a, you know, just a short walkthrough of the of the Snowflake, and then the snowflake environment.
347 00:41:37.860 ⇒ 00:41:58.830 Robert Tseng: And then I think Jared is confused over like Meta base versus real. And I I just wanna start to have that conversation of where we see the different like, what what types of report like, where to go to for different types of reporting. Yeah, I mean, I expect them to just be like, Oh, why can’t we just have it in one place, and I think I want to just take some time to address that.
348 00:41:59.290 ⇒ 00:42:10.980 Aman Nagpal: Yeah. Yeah, I I don’t. I didn’t talk to him about it. My assumption is is, you know, and he was like, Oh, why do we? Why are you sending us something we can’t really use? I mean, yeah. All we can say is, Look, this is here for you
349 00:42:11.010 ⇒ 00:42:32.639 Aman Nagpal: to do self serve, and you don’t need to use it, you know. And obviously, you know, you’re you’re gonna tell them that already. I’m hoping they’ll be good with Meta base. Otherwise, you know, if we need to figure out another visual visualization tool, that’s fine. And then we can discuss and finalize. Well, not really finalize. I think we already discussed about amplitude. That most stuff will still go into amplitude minus
350 00:42:32.810 ⇒ 00:42:35.520 Aman Nagpal: the financial stuff, and we’ll figure out a way to make it all work.
351 00:42:35.840 ⇒ 00:42:36.520 Robert Tseng: Yeah.
352 00:42:38.370 ⇒ 00:42:42.510 Robert Tseng: So it’s gonna be Justin and Jared on. That is that on on your side? Is that.
353 00:42:43.020 ⇒ 00:42:45.739 Aman Nagpal: Just those 2, and then I will try to join as well.
354 00:42:45.740 ⇒ 00:42:46.780 Robert Tseng: Okay. Cool.
355 00:42:48.390 ⇒ 00:42:50.949 Aman Nagpal: Sweet anything else you guys want to go over
356 00:42:51.410 ⇒ 00:42:52.549 Aman Nagpal: or need for me.
357 00:42:57.860 ⇒ 00:42:58.889 Payas Parab: I’ll go to here.
358 00:43:00.120 ⇒ 00:43:02.939 Aman Nagpal: Cool. Thank you. Guys so much. Talk soon.
359 00:43:02.940 ⇒ 00:43:04.759 Robert Tseng: Okay. See you. Bye.