Meeting Title: Data Model and Refunds Sync Date: 2025-07-10 Meeting participants: Awaish Kumar, Fireflies.ai Notetaker Tigran, Robert Tseng, Annie Yu, Amber Lin, Demilade Agboola, Josh
WEBVTT
1 00:02:21.190 ⇒ 00:02:22.470 Robert Tseng: Hello!
2 00:02:22.470 ⇒ 00:02:23.680 Amber Lin: Hi.
3 00:02:27.430 ⇒ 00:02:30.289 Robert Tseng: Is my speaker better today, Annie?
4 00:02:31.280 ⇒ 00:02:36.750 Annie Yu: Yes, much better. What is that headphone, new headphone.
5 00:02:37.340 ⇒ 00:02:46.129 Robert Tseng: It’s the same. It’s old headphones. But I’m just, I’m gonna use the wire when I’m taking calls. I think I think seems like it’s more consistent.
6 00:02:47.230 ⇒ 00:02:47.990 Robert Tseng: Yeah.
7 00:02:50.700 ⇒ 00:03:00.869 Robert Tseng: okay, let’s jump into it. Got a lot to cover, I believe Henry is also going to join, but I’ll give him some time to triple in.
8 00:03:05.780 ⇒ 00:03:18.780 Robert Tseng: Okay, so I’m gonna I have a grooming session with amber later today on these tickets. So
9 00:03:19.716 ⇒ 00:03:28.310 Robert Tseng: I think the quality of our tickets, and the organization of this board is gonna get better now that we have actual Pm support. And it’s not just
10 00:03:29.016 ⇒ 00:03:30.010 Robert Tseng: me just
11 00:03:30.140 ⇒ 00:03:52.519 Robert Tseng: kind of doing doing whatever so, you know, just be able to look out for that. We. I’ve kind of given her and Henry the intro to kind of where we’re at on our current projects. But for now I think kind of how this handoff is. Gonna go, I’ll still be running these calls. But I think I’m just gonna get some kind of more support on the back end for now.
12 00:03:54.260 ⇒ 00:03:57.885 Robert Tseng: yeah. So just fyi. And then,
13 00:04:00.420 ⇒ 00:04:07.789 Robert Tseng: yeah, I guess for this call. I just want to kind of run through some of the things that I know are kind of outstanding from yesterday, and then I think I want
14 00:04:07.960 ⇒ 00:04:13.700 Robert Tseng: to give some time for a ways to kind of follow up on the data model since that’s what we were waiting on yesterday.
15 00:04:15.970 ⇒ 00:04:22.260 Robert Tseng: so yeah, I think let’s just jump into it. 1st thing that comes to mind is the vial size thing.
16 00:04:23.170 ⇒ 00:04:28.510 Robert Tseng: yeah. Can you like I I spoke with Andy briefly, about this yesterday.
17 00:04:28.720 ⇒ 00:04:40.379 Robert Tseng: what like, what’s stopping us? I thought, we kind of got to a place where we were basically like, Okay, well, for several products we can. We can do this. But I’m not really sure where where we’re at.
18 00:04:40.950 ⇒ 00:04:45.360 Robert Tseng: I thought we had this conversation earlier this week, and then, like, it’s just been a couple of days.
19 00:04:47.900 ⇒ 00:04:51.749 Demilade Agboola: Yeah, for some products we could.
20 00:04:52.040 ⇒ 00:04:54.450 Demilade Agboola: So the the roll up is done.
21 00:04:57.260 ⇒ 00:05:03.740 Demilade Agboola: Based off like all time. So it’s at this current moment. It’s not like a a granular thing.
22 00:05:04.332 ⇒ 00:05:11.060 Demilade Agboola: To do. Like to break it to the shipment per day, vowed.
23 00:05:11.654 ⇒ 00:05:23.670 Demilade Agboola: That can happen. But there was. I figured there was an easier way, which is what I shared with and yesterday. So there is a dosage, and it’s more what’s the word?
24 00:05:23.870 ⇒ 00:05:32.230 Demilade Agboola: It’s more expansive than the previous solution, because this comes directly from past so they’re using dosage
25 00:05:32.740 ⇒ 00:05:34.519 Demilade Agboola: column in the.
26 00:05:34.520 ⇒ 00:05:37.275 Robert Tseng: Does that line up with the assumptions that
27 00:05:38.040 ⇒ 00:05:40.370 Robert Tseng: Rebecca gave you via the Google sheet.
28 00:05:41.100 ⇒ 00:05:48.830 Demilade Agboola: No, not that which is kind of what I also like. Put as a limitation on this. So we we might be able to get more like
29 00:05:48.990 ⇒ 00:05:50.280 Demilade Agboola: values.
30 00:05:50.500 ⇒ 00:05:52.269 Demilade Agboola: The problem is.
31 00:05:52.810 ⇒ 00:05:58.239 Demilade Agboola: it doesn’t necessarily be like we also have quantity. But the problem is sometimes you know how
32 00:05:59.110 ⇒ 00:06:03.390 Demilade Agboola: I would have to check. But you know how sets in orders. It’s shipped. Multiple things are shipped at once.
33 00:06:04.540 ⇒ 00:06:07.920 Demilade Agboola: The orders might not necessarily be like representative, as you know.
34 00:06:09.030 ⇒ 00:06:16.089 Demilade Agboola: If also the 1, 2, and 3 were shipped. For instance, it might just be one of those values, but this is much closer to
35 00:06:16.380 ⇒ 00:06:21.319 Demilade Agboola: what we will need, and it will be about tweaking this to get closer to our final solution.
36 00:06:23.160 ⇒ 00:06:32.460 Robert Tseng: Okay, I I still don’t I? I hear some changes. But can you like just grab time on my calendar later today, I wanna like
37 00:06:32.620 ⇒ 00:06:39.850 Robert Tseng: I wanna respect. I have an understanding of like where I think we left off. I get that. You’re getting direct data directly from Bask. But like.
38 00:06:40.580 ⇒ 00:06:49.310 Robert Tseng: I’m not really sure what what the hangup is, so I don’t. We don’t have to cover it on this call. But if you could just find time on my calendar today, and just we’ll just talk through it.
39 00:06:50.560 ⇒ 00:06:51.660 Demilade Agboola: Okay. Sounds good.
40 00:06:52.420 ⇒ 00:06:55.053 Robert Tseng: Yeah. So we’ll do that.
41 00:06:55.700 ⇒ 00:07:04.409 Robert Tseng: yeah, just because I think that’s there. There’s still that I would say, this is high priority on Rebecca side. She’s trying to basically size, like.
42 00:07:04.750 ⇒ 00:07:22.410 Robert Tseng: well, there’s there’s 2. There’s 2 investigations from from her side. One is one. Is this on the file size wanting to do like, you know, file forecasting and be able to know how we’re being overcharged from the booth and farm pharmacy. And then the second one is the refunds issue that we kind of. We’re discussing on a few threads yesterday.
43 00:07:23.130 ⇒ 00:07:26.540 Robert Tseng: I know you pointed out an example of where a refund
44 00:07:26.670 ⇒ 00:07:36.080 Robert Tseng: was missed or like. It’s not coming in from mass data. I asked to kind of just for you to size like how big is what, how much money are we leaving on the table?
45 00:07:36.360 ⇒ 00:07:39.270 Robert Tseng: I don’t know if you’ve gotten a chance to to follow up on that yet.
46 00:07:52.470 ⇒ 00:08:00.640 Demilade Agboola: I just realized I was muted. I said I. I asked, if are we looking at refunds specifically, or are we looking at all others.
47 00:08:03.385 ⇒ 00:08:12.224 Robert Tseng: Well, so I mean, this is the thread, the original question from Katie. I I think you can find that in the even, in in the even slack as well. But
48 00:08:12.510 ⇒ 00:08:13.050 Demilade Agboola: Yeah.
49 00:08:13.810 ⇒ 00:08:17.270 Robert Tseng: Yeah, I mean, in my understanding, you’re showing me.
50 00:08:17.690 ⇒ 00:08:30.340 Robert Tseng: Here’s an order. It got refunds. But it’s we’re we’re missing that refund data because it didn’t come through. And so I’m asking you like how much I mean, I I feel like this is adjacent to some of those other stuff. I just want you to kind of put the story together like
51 00:08:30.470 ⇒ 00:08:47.769 Robert Tseng: we understand that there’s a delay and refunds. It’s not like process immediately doesn’t come through. I’ve asked Katie to pull her source, and let us know within one week’s worth of data, how much you know. How many refunds is she seeing? How often is that from what we’re seeing? So if there’s a delay problem, it’s like a moving target. We can isolate that.
52 00:08:47.770 ⇒ 00:09:01.760 Robert Tseng: But if it’s really just that we’re missing data. I want to know how much, how much money we’re just like misreporting on on the refund side. So I think that’s that we need to be able to say that I can’t just like isolate one example and bring it up to Zack like he’s not going to do anything about it.
53 00:09:02.800 ⇒ 00:09:06.099 Demilade Agboola: Yeah, yeah, fair fair. It’s it’s a larger. It’s about.
54 00:09:06.300 ⇒ 00:09:10.369 Demilade Agboola: So there are 28 refunds, and over the last 3 days.
55 00:09:10.470 ⇒ 00:09:17.650 Demilade Agboola: But we are only reporting, or we’re only saying 17, or which some of them are duplicates, and
56 00:09:17.900 ⇒ 00:09:28.250 Demilade Agboola: they don’t appear as duplicates in the role which is kind of point 3 that I mentioned. So on in the raw data, it’s showing like 2,200
57 00:09:28.700 ⇒ 00:09:37.859 Demilade Agboola: But in the, you know, like we have, we have 2 different rows that sum up to about $4,000, which is not true, because the auditor does not like validate that
58 00:09:38.604 ⇒ 00:09:41.259 Demilade Agboola: so that’s kind of what I would need to
59 00:09:42.130 ⇒ 00:09:52.240 Demilade Agboola: isolate. So we will. We are both missing numbers which will show the sum, and then in some cases we might have duplicates of sets of numbers which will inflate the sum.
60 00:09:53.600 ⇒ 00:10:11.809 Robert Tseng: Yeah, no, I I get that. You have the scenarios in your mind. I think you’re you’re probably right, like I. I think that makes sense to me. I just yeah in order for me to. If you could build the story like, I just like, let me know, like what what the what we think the estimate impact is you don’t have to be right. But you could make an assumption off of, like.
61 00:10:12.300 ⇒ 00:10:21.299 Robert Tseng: you know, over the past week we missed whatever. I’m just like going off of the numbers that you’re giving me. We missed, you know, 8 to 10 refunds and
62 00:10:21.300 ⇒ 00:10:49.259 Robert Tseng: you know it’s off by 4,000. If we were to abstract that to like over over a month, you know, we could be misreporting on 50 refunds, and it could be, you know, $20,000 leaving on the team. I mean, I’m just putting those numbers out there. But like I, I need, I need something. I need something like that to kind of stitch what you’re what you’re saying, and from like, where where things could be broken to like what the impact is on what the on on. The report that Katie is is is asking about right.
63 00:10:50.260 ⇒ 00:10:51.300 Demilade Agboola: Yeah, sounds good.
64 00:10:51.840 ⇒ 00:10:59.175 Robert Tseng: Okay, yeah. So please follow up on that. Just like, help me build a story. Let me like we, we need to kind of follow up with on that one.
65 00:11:00.090 ⇒ 00:11:08.030 Robert Tseng: Okay, and then on cog stuff. I see it’s still sitting here. Any movement on these things.
66 00:11:09.944 ⇒ 00:11:17.849 Demilade Agboola: I’m so, submitted Christian on today. I tried to move her yesterday, but unfortunately she didn’t have any time. So we’ll meet today and I’ll present
67 00:11:18.020 ⇒ 00:11:21.320 Demilade Agboola: the 2 options that we talked about yesterday to her.
68 00:11:22.040 ⇒ 00:11:23.110 Demilade Agboola: Okay, thanks. A link.
69 00:11:23.110 ⇒ 00:11:23.500 Demilade Agboola: Yeah. One.
70 00:11:23.500 ⇒ 00:11:25.429 Demilade Agboola: I think the pharmacy data.
71 00:11:25.430 ⇒ 00:11:25.750 Robert Tseng: Yeah.
72 00:11:25.750 ⇒ 00:11:27.040 Demilade Agboola: But yeah.
73 00:11:27.910 ⇒ 00:11:40.400 Robert Tseng: Okay, yeah, I’ll do my best when I meet with amber later today, and we’re grooming these tickets. I’m gonna consolidate these down and try to be more clear about like what’s outstanding here. But I think if you you run that meeting with with Christian, I’m not gonna be there.
74 00:11:41.930 ⇒ 00:11:51.030 Robert Tseng: yeah, okay, I’ll move on to Annie. Just really quick. Yeah. So I we. We went through. We whiteboarded out the change to the
75 00:11:51.230 ⇒ 00:11:53.782 Robert Tseng: order journey. Dash yesterday.
76 00:11:54.600 ⇒ 00:11:58.800 Robert Tseng: That was towards the end of the day. But kind of where? Where are we? On on that.
77 00:12:00.250 ⇒ 00:12:11.870 Annie Yu: Yeah, I was able to add a chart. And still I’m working on the tables. But in the meantime I also found another data issue. If we can click into my ticket.
78 00:12:13.610 ⇒ 00:12:15.160 Annie Yu: It’s 4, 40.
79 00:12:16.910 ⇒ 00:12:17.560 Robert Tseng: Okay.
80 00:12:21.580 ⇒ 00:12:25.669 Annie Yu: So in the table we want to flag all the orders that
81 00:12:25.890 ⇒ 00:12:41.819 Annie Yu: were placed, but not yet sent to pharmacy. But I found some examples that you see, this order was actually delivered on the desk side as well, but we never received a sent to pharmacy date. So it’s flagged. As
82 00:12:41.820 ⇒ 00:12:48.138 Annie Yu: yeah, that happens, I don’t think vast gives us every everything perfectly. I think, in this example.
83 00:12:49.860 ⇒ 00:12:51.060 Annie Yu: I mean, well, we basically
84 00:12:51.060 ⇒ 00:13:04.970 Annie Yu: right? Yeah, I was thinking, I can write some logic. If it’s delivered, we can flag it as true flag as sent to pharmacy as true. But you won’t have the date, and if it’s not yet delivered, we can’t catch that.
85 00:13:05.720 ⇒ 00:13:22.205 Robert Tseng: Yeah, we don’t need to build this into the report. But if you could have the query that catches I wanna know how many times is it happening? Because this is basically we’re telling Zach. Look, you, gave us no details about this order until it was delivered right? Like that’s, you know, that’s obviously a problem. So
86 00:13:22.480 ⇒ 00:13:22.930 Demilade Agboola: So.
87 00:13:22.930 ⇒ 00:13:24.689 Robert Tseng: I don’t know how many. Yeah.
88 00:13:24.980 ⇒ 00:13:31.630 Demilade Agboola: I want to clarify delivery status data and delivery status. Information is from Shiple.
89 00:13:31.990 ⇒ 00:13:32.760 Robert Tseng: Yeah, yeah.
90 00:13:32.760 ⇒ 00:13:40.519 Demilade Agboola: The ships. The ship date. So that’s why we have it. The ship date is from bask, and the center pharmacy date is from bask.
91 00:13:42.920 ⇒ 00:13:57.137 Robert Tseng: Yeah. So that’s why I’m I’m saying that. Okay, well, from our own data that we’re able to get directly from Shippo, we see that this order was delivered. But from bask side they didn’t tell us anything about this order. Right? That’s that’s a problem. Because,
92 00:13:57.670 ⇒ 00:14:07.704 Robert Tseng: obviously, like they were, we’re missing order updates for that for that order like, we didn’t get anything from bask until it was actually processed, and or until until we saw it ship off.
93 00:14:10.000 ⇒ 00:14:18.060 Annie Yu: Then my question would be cause. We don’t want to falsely call an order not yet sent to pharmacy right.
94 00:14:18.630 ⇒ 00:14:37.449 Robert Tseng: Yeah, we should exclude that from from the from this table that you’re talking about. But I also separately want, like a list of like those orders, so we can see how many times that’s happening. Where like, Shipbo is telling us information about the order that, like Bass never told us about. That’s, you know, simply what I what I’m saying.
95 00:14:39.198 ⇒ 00:14:44.930 Annie Yu: So what if I’m I mean, I can look into that. But I’m suspecting there are orders that
96 00:14:45.230 ⇒ 00:14:46.639 Annie Yu: don’t have.
97 00:14:47.230 ⇒ 00:14:52.649 Annie Yu: like, if it’s not delivered yet, and we don’t have shipped date, and we don’t have sent to pharmacy date.
98 00:14:52.770 ⇒ 00:14:57.409 Annie Yu: But then, if compared to bask, they are actually already shipped.
99 00:14:58.620 ⇒ 00:15:01.969 Annie Yu: and I don’t really know what to do with those, and I don’t want to.
100 00:15:02.470 ⇒ 00:15:03.930 Annie Yu: Not yet shipped.
101 00:15:06.060 ⇒ 00:15:17.750 Robert Tseng: Yeah, I I mean, I would say, for this table, if the delivery status doesn’t matter right. It’s just from when the order is placed to when it leaves the pharmacy so is shipped equals? True? Right? Because that’s all data that comes from mask.
102 00:15:17.910 ⇒ 00:15:36.620 Robert Tseng: I mean, this is good, that we have this field that lets us know. Like I’m looking at this one row, one should be excluded from that table. But I still want to know that row one happened like, and I want to know how many of those, how many of those orders we have, because that’s that should not be counted against the pharmacy team and their sla
103 00:15:37.700 ⇒ 00:15:52.200 Robert Tseng: situation because didn’t have any control of that. It just asked, didn’t, didn’t, didn’t share any data there, so it should be excluded from this table. But I I want the I want the. I want a separate list of those orders where this situation is happening.
104 00:15:54.370 ⇒ 00:15:55.040 Annie Yu: Yeah.
105 00:15:57.290 ⇒ 00:15:57.790 Robert Tseng: Okay.
106 00:15:58.290 ⇒ 00:15:58.810 Annie Yu: Okay.
107 00:15:59.030 ⇒ 00:16:02.290 Robert Tseng: Does does that make sense like, I know, I’m I basically split that request.
108 00:16:02.623 ⇒ 00:16:13.280 Annie Yu: That makes sense. I think one thing now is cause. Remember, in the dashboard we want to have 3 different tables. One is focusing on orders that haven’t made it to pharmacy.
109 00:16:13.950 ⇒ 00:16:19.290 Annie Yu: And then one focuses on orders that made it to pharmacy, but haven’t been shipped.
110 00:16:20.080 ⇒ 00:16:26.540 Annie Yu: and I I just don’t want to show any orders that were actually already shipped. But we just
111 00:16:26.900 ⇒ 00:16:31.549 Annie Yu: because we don’t have the sent to Pharmacy date, so we flag it as not yet
112 00:16:32.350 ⇒ 00:16:34.719 Annie Yu: center for me. Does does that make sense.
113 00:16:37.750 ⇒ 00:16:38.270 Robert Tseng: Alright.
114 00:16:38.270 ⇒ 00:16:39.079 Demilade Agboola: But I think for okay. So
115 00:16:40.290 ⇒ 00:16:52.729 Demilade Agboola: well, those scenarios we can just easily, once it has a ship date or a delivery date like we can exclude it from the sent to pharmacy list or not sent to Pharmacy List, because we know that there’s no way they were shipped
116 00:16:53.070 ⇒ 00:16:56.269 Demilade Agboola: or delivered without having sent to a pharmacy.
117 00:16:58.380 ⇒ 00:16:59.230 Annie Yu: Okay. Okay.
118 00:17:02.940 ⇒ 00:17:09.009 Robert Tseng: Okay, cool. So I think we’re live here. I know you were still working on this. Yeah, the 3 tables we talked about adding.
119 00:17:09.470 ⇒ 00:17:14.429 Robert Tseng: yeah, I guess when when that’s ready for you, just let let me know. But
120 00:17:14.930 ⇒ 00:17:17.240 Robert Tseng: okay, anything else on this.
121 00:17:18.659 ⇒ 00:17:19.079 Annie Yu: No.
122 00:17:19.540 ⇒ 00:17:19.909 Robert Tseng: Okay.
123 00:17:20.780 ⇒ 00:17:27.810 Robert Tseng: cool. Then. Yeah. I wish I wanna kind of, do you have an update on the customer data model in that? We moved.
124 00:17:28.079 ⇒ 00:17:30.260 Robert Tseng: That’s kind of reconstructed in the warehouse.
125 00:17:30.870 ⇒ 00:17:32.089 Awaish Kumar: I saw you not pushing.
126 00:17:32.090 ⇒ 00:17:32.890 Robert Tseng: The Pr.
127 00:17:33.370 ⇒ 00:17:34.096 Robert Tseng: I have.
128 00:17:34.600 ⇒ 00:17:39.720 Awaish Kumar: I’ve created like a 3 different models.
129 00:17:41.110 ⇒ 00:17:44.569 Awaish Kumar: One is called into user trades.
130 00:17:49.350 ⇒ 00:17:52.070 Awaish Kumar: Let me accomplish your.
131 00:17:52.670 ⇒ 00:17:58.719 Robert Tseng: If you have it handy, you can share your screen. Otherwise it’s gonna take me a while to go and look for all these tables.
132 00:18:00.000 ⇒ 00:18:01.400 Awaish Kumar: Yeah, I can share my screen.
133 00:18:01.850 ⇒ 00:18:02.450 Robert Tseng: Okay.
134 00:18:10.830 ⇒ 00:18:12.103 Awaish Kumar: So we
135 00:18:21.730 ⇒ 00:18:25.254 Awaish Kumar: one is called infuser trace.
136 00:18:27.320 ⇒ 00:18:36.739 Awaish Kumar: So this is basically like generating the order you are trying to generate with this metrics for each trade.
137 00:18:36.740 ⇒ 00:18:39.429 Robert Tseng: Like my really big union? Query.
138 00:18:39.789 ⇒ 00:18:44.460 Awaish Kumar: Yeah, based on that, I added few more metrics on top of it.
139 00:18:45.122 ⇒ 00:18:50.929 Awaish Kumar: And yeah, change a little bit. But you know, if I concept is the same.
140 00:18:52.300 ⇒ 00:18:56.990 Robert Tseng: Oh, wow, sweet, you got this to work. Okay, yeah.
141 00:18:58.070 ⇒ 00:19:04.460 Awaish Kumar: So we have all these trade name data type, no normal percentages
142 00:19:04.610 ⇒ 00:19:10.560 Awaish Kumar: listing user for this specific trade and few other features. Yeah.
143 00:19:10.840 ⇒ 00:19:15.020 Awaish Kumar: And then I applied a some
144 00:19:17.180 ⇒ 00:19:25.700 Awaish Kumar: like filters on top of it to figure out which can be the meaningful columns. And these are the ones which I actually
145 00:19:28.100 ⇒ 00:19:39.790 Awaish Kumar: right? So like, I’m saying, at least like 5% of the record have nominal values.
146 00:19:40.690 ⇒ 00:19:46.939 Awaish Kumar: And it’s listing users. And it’s it’s not having like us.
147 00:19:47.310 ⇒ 00:19:54.840 Awaish Kumar: There’s a variance in the data. So not not all the the for all the users is the same. There is like kind of meaningless.
148 00:19:55.650 ⇒ 00:20:06.879 Awaish Kumar: And it must have more than one unique values. And by using the Nh being updated regularly by doing that, I got like around 40 columns
149 00:20:07.540 ⇒ 00:20:09.640 Awaish Kumar: which are useful.
150 00:20:10.060 ⇒ 00:20:13.603 Awaish Kumar: And then I included some of them
151 00:20:14.110 ⇒ 00:20:22.559 Awaish Kumar: like, user Id and email and phone, like, some of these things are just identity columns. I just edited also.
152 00:20:22.690 ⇒ 00:20:29.050 Awaish Kumar: And then we have a a final model which is basically called user profiles.
153 00:20:30.350 ⇒ 00:20:34.720 Awaish Kumar: And we just okay, who’s
154 00:20:39.440 ⇒ 00:20:42.970 Awaish Kumar: so it will have a kind of a who’s
155 00:20:43.810 ⇒ 00:20:46.439 Awaish Kumar: unique, like, single row per user.
156 00:20:47.078 ⇒ 00:20:51.350 Awaish Kumar: with the best possible value we could find in the data
157 00:20:55.730 ⇒ 00:20:58.659 Awaish Kumar: like, we have user Id email.
158 00:21:04.730 ⇒ 00:21:19.649 Robert Tseng: Okay. My next question is, how does this compare to dim customers? Right? I guess you may not know the answer now, but I wanna know with this new model that we’re pulling from segment. We’ve redone the stitching. How many more users are we customers? Are we identifying?
159 00:21:19.870 ⇒ 00:21:31.949 Robert Tseng: And like. I mean, you can work with Henry on this like I’ll I’ll probably have you guys meet in a couple of hours because he’s he’ll he’ll yeah. He’s gonna he’s gonna take over this part of the of the kind of profiles thing. But
160 00:21:32.410 ⇒ 00:21:35.689 Robert Tseng: yeah, I mean, he’s I think he’s gonna ask the same questions. He’s gonna know.
161 00:21:36.520 ⇒ 00:21:37.330 Robert Tseng: I think at that point.
162 00:21:37.330 ⇒ 00:21:44.520 Awaish Kumar: This is basically, the main difference is that in the dim customers will only have the the customer data
163 00:21:45.046 ⇒ 00:21:49.610 Awaish Kumar: which are basically the customers, and we have a lot more information about them.
164 00:21:49.850 ⇒ 00:21:56.980 Awaish Kumar: So because they have become customer, they have signed up. We have email birthdays, gender, everything we can get
165 00:21:57.702 ⇒ 00:22:03.980 Awaish Kumar: but for these users in for this table we have a large pool of users who.
166 00:22:04.576 ⇒ 00:22:12.180 Awaish Kumar: might have visited our website, or something like that, they might not be the customer. They might just be a visitor of the website.
167 00:22:13.920 ⇒ 00:22:18.176 Awaish Kumar: and they might not have made purchases as well. So it’s like a big
168 00:22:19.297 ⇒ 00:22:27.170 Awaish Kumar: super set of users who somehow kind of interacted with the bus, but maybe do not convert it.
169 00:22:27.410 ⇒ 00:22:34.219 Awaish Kumar: They visited the website, but didn’t actually signed up, or actually made an order.
170 00:22:34.440 ⇒ 00:22:42.210 Awaish Kumar: and things like that. And this for those those which are customers, we could see them. We we can see, like
171 00:22:42.330 ⇒ 00:22:44.019 Awaish Kumar: some of the things which
172 00:22:45.686 ⇒ 00:22:53.249 Awaish Kumar: like like these columns, I don’t know like how segment is calculating them, but it’s the likelihood of
173 00:22:53.380 ⇒ 00:22:59.170 Awaish Kumar: likelihood to purchase is greater than 60% things like that.
174 00:22:59.570 ⇒ 00:23:05.810 Awaish Kumar: So like this is a field which is calculated on segment. We don’t have this and dream customer
175 00:23:07.822 ⇒ 00:23:11.750 Awaish Kumar: likelihood to churn or things. In some of these columns.
176 00:23:12.090 ⇒ 00:23:16.399 Awaish Kumar: and basically all other problems are, are there like
177 00:23:16.900 ⇒ 00:23:31.269 Awaish Kumar: for existing customers, all other information which we are getting from this user profile table is basically, we already have them because we already have what? What source they come from. What was their 1st visit.
178 00:23:31.770 ⇒ 00:23:39.300 Awaish Kumar: 1st visit thing might change, because and in our do you know what
179 00:23:39.570 ⇒ 00:23:50.689 Awaish Kumar: dem customer our 1st 1st visit is when order was placed, and here 1st visit might be that somebody visited last website in January, made an order in February.
180 00:23:51.450 ⇒ 00:23:53.710 Awaish Kumar: So 1st visit will still be January.
181 00:23:57.230 ⇒ 00:24:01.749 Awaish Kumar: and but the 1st order was in February. So in dim customer.
182 00:24:01.910 ⇒ 00:24:06.629 Awaish Kumar: when we when I, when somebody says, Okay, when we’ve seen this customer 1st time.
183 00:24:06.880 ⇒ 00:24:08.560 Awaish Kumar: they can only see that they’re
184 00:24:08.770 ⇒ 00:24:12.969 Awaish Kumar: this customer made an order in February. It was the 1st time we saw it.
185 00:24:13.380 ⇒ 00:24:16.160 Awaish Kumar: but actually, maybe we saw it in January, because
186 00:24:16.540 ⇒ 00:24:19.070 Awaish Kumar: at that time he he visited the Basquen thing.
187 00:24:20.950 ⇒ 00:24:41.009 Robert Tseng: It’s yeah, yeah, okay, alright. So I’m clear, I’m clear on this. So next step. So we have. Yeah, I mean, I’m calling segment and runner stack later today. So I think this models in production, I’m gonna show them. This is now where the customer data model sits. It’s not in its final form yet. We’re gonna I think over the next.
188 00:24:41.500 ⇒ 00:24:54.409 Robert Tseng: I mean, I don’t think it’s gonna happen in the next day. Probably like to today tomorrow, like kind of leading into early next week, we’re gonna refine this set of of fields. But I think now it’s in a form where it’s easy to kind of like, make these changes.
189 00:24:54.910 ⇒ 00:25:03.677 Robert Tseng: And yeah, but yeah, this is the base. This is the base model that I want so great. I think we’re glad we have this. I can, we can continue on
190 00:25:04.590 ⇒ 00:25:07.980 Robert Tseng: yeah, we have to basically take this and push it into customer I/O.
191 00:25:08.550 ⇒ 00:25:13.200 Robert Tseng: And I’m gonna do that, either, through rudders are going to do that through rudder stack and through segment.
192 00:25:13.440 ⇒ 00:25:35.410 Robert Tseng: And we’re just gonna see kind of like performance. We’re gonna look at cost. We’re gonna see? Like what you know those trade offs. But getting that raw data from segment, and then kind of combining it with our under existing data in a single customer data model. That’s what this is right. I think you have total orders here. So that obviously came from our, our.
193 00:25:35.410 ⇒ 00:25:40.309 Awaish Kumar: No like this is like user profile from segment. So total orders in this.
194 00:25:40.830 ⇒ 00:25:46.960 Awaish Kumar: the table is also coming from this segment data, because this.
195 00:25:46.960 ⇒ 00:25:57.639 Robert Tseng: Okay? Well, so then, that’s that’s what we need to. Well, that’s what’s gonna change right? Like Ltv. And orders like, I wouldn’t trust segment for this, we would. I want to use our our data from big grid. Yeah.
196 00:25:57.640 ⇒ 00:26:02.150 Awaish Kumar: Yeah, like, this is not the final, enriched, enriched model. We, we said.
197 00:26:03.250 ⇒ 00:26:09.480 Awaish Kumar: from user profile, from segment. Now I have to create one more which will join this one with our
198 00:26:09.980 ⇒ 00:26:12.409 Awaish Kumar: kind of gym customer, and then
199 00:26:12.890 ⇒ 00:26:19.579 Awaish Kumar: which, like Ltv. And the total orders and things like that, I had like few more fields here, and then it will be
200 00:26:20.700 ⇒ 00:26:26.820 Awaish Kumar: so. Then we will have this customer interest model, which will have everything from segment and from our team customer.
201 00:26:27.520 ⇒ 00:26:41.979 Robert Tseng: Great. Okay? So that clears it great. So you basically trim down like 340 like fields in in segment down to like less than 40 the definitions you use to free filter. Can you add that to the notion, Doc? So I can review it, and I want
202 00:26:42.360 ⇒ 00:26:42.950 Robert Tseng: get it to you.
203 00:26:42.950 ⇒ 00:26:43.889 Awaish Kumar: So it is.
204 00:26:43.890 ⇒ 00:26:44.760 Robert Tseng: Trees of it.
205 00:26:45.530 ⇒ 00:26:57.129 Awaish Kumar: It is in the I haven’t put in motion, but it is here in this, like Github Repository. If you want, I can put it in motion as well. It’s basically these.
206 00:26:57.360 ⇒ 00:27:08.799 Robert Tseng: Yeah, if you could just paste in a notion, just because Henry hasn’t looked at the code base yet, I understand. Like I’ll just. I’ll pull. I’ll pull this branch, and then I’ll be able to reference this model like moving forward. Just fine.
207 00:27:09.407 ⇒ 00:27:18.320 Robert Tseng: But yeah, okay, so yeah, we, I think I have enough here to continue the conversation with respect and and segment today.
208 00:27:18.540 ⇒ 00:27:30.749 Robert Tseng: And then, yeah, we’re gonna keep working on the on the enriched data model. I’m gonna yeah, that between you, basically, Henry will help you help you arrive at that final enriched model.
209 00:27:31.533 ⇒ 00:27:32.819 Robert Tseng: And then.
210 00:27:33.770 ⇒ 00:27:47.769 Robert Tseng: yeah, we’re just gonna test test the push to to customize. So okay, that gives me a good sense of where we’re at on. This probably won’t finish until early early to mid next week, which is what which is fine. We still have some time for that decision as we made.
211 00:27:49.510 ⇒ 00:27:50.980 Robert Tseng: Okay, cool.
212 00:27:51.920 ⇒ 00:27:54.186 Robert Tseng: Alright. Well, good progress. Everyone.
213 00:27:55.970 ⇒ 00:27:57.453 Robert Tseng: yeah, I mean, I think,
214 00:27:58.030 ⇒ 00:28:05.399 Robert Tseng: it would be great to get some of those updates out. I think we haven’t pushed an update out in the analytics channel and in a while. So
215 00:28:05.570 ⇒ 00:28:16.177 Robert Tseng: I’m sure, like any. Once you’re done with kind of that, I’ll review your your, the dash update so we can share that. And then, Dave a lot. I know you’re you’re still kind of meeting with people. So
216 00:28:16.680 ⇒ 00:28:17.560 Robert Tseng: okay.
217 00:28:23.850 ⇒ 00:28:35.390 Robert Tseng: anything else. I know we didn’t cover every ticket like I said, Amber and I are. Gonna go and groom this in a couple of hours, so she’s gonna help me clean up the board, and she’ll kind of just make make it better moving forward. But
218 00:28:36.780 ⇒ 00:28:40.430 Robert Tseng: yeah, anything else that needs kind of like urgent feedback that you’re blocked on.
219 00:28:46.990 ⇒ 00:28:47.860 Robert Tseng: Okay?
220 00:28:48.060 ⇒ 00:28:55.880 Robert Tseng: If not, then. Yeah, I mean, there’s a few things that I’m not gonna add into this cycle for this week. But next week I’ve I’ve been.
221 00:28:57.420 ⇒ 00:29:02.250 Robert Tseng: There’s yeah. There’s there’s there’s some things coming. I don’t. I’m not gonna talk about it right now. I don’t. Wanna
222 00:29:03.203 ⇒ 00:29:09.820 Robert Tseng: people. So yeah, okay, cool. Alright. Well, then, that’s it, and we’ll we’ll keep talking slot.
223 00:29:09.820 ⇒ 00:29:18.689 Josh: Get the. Did we get the data back to make sure that those reports are represented correctly.
224 00:29:19.386 ⇒ 00:29:19.880 Josh: know that.
225 00:29:19.880 ⇒ 00:29:23.639 Robert Tseng: For the products like the Hrt. And everything they want. It.
226 00:29:26.360 ⇒ 00:29:32.600 Demilade Agboola: Yeah, right now they are favoring Hrc, as the bundles
227 00:29:32.940 ⇒ 00:29:37.919 Demilade Agboola: like they’re favoring the bundles about the product. Id, so you should represent the values as bundles.
228 00:29:38.100 ⇒ 00:29:52.460 Josh: But are you guys like pulling in like the new naming conventions for certain things like it? Just, I feel like it’s under reported on the sales. I think that we probably like. They’ve probably been adding things faster, and they’ve been communicating them to you.
229 00:29:53.170 ⇒ 00:29:53.949 Josh: That’s my.
230 00:29:54.366 ⇒ 00:30:01.350 Demilade Agboola: That. Yeah, that’s a possibility. But I I did have Cotter did tag Christina
231 00:30:01.870 ⇒ 00:30:13.419 Demilade Agboola: 2 days ago, and she sent the new Hrt specifically Hrt values. And those are what are in the reports right now. So unless there’s a new like value that she didn’t send.
232 00:30:13.700 ⇒ 00:30:14.990 Demilade Agboola: it should be here.
233 00:30:16.770 ⇒ 00:30:23.919 Robert Tseng: Yeah, I mean, we can ask that Zack to send the product. Csv, I mean, I followed up yesterday. He didn’t send anything, so
234 00:30:24.670 ⇒ 00:30:31.630 Robert Tseng: you know, I think that would be our only other way of of pulling in the new the new product names, or any data around that.
235 00:30:34.740 ⇒ 00:30:36.220 Josh: Interesting.
236 00:30:39.485 ⇒ 00:30:40.190 Josh: Okay.
237 00:30:41.590 ⇒ 00:30:49.309 Robert Tseng: Yeah, I mean, if you see something that’s if you feel like you’re missing something. I mean, I don’t like that. It’s reactive. But like. I mean, I’ll I’ll I’ll
238 00:30:49.590 ⇒ 00:30:58.199 Robert Tseng: I’m gonna bump Zack again you’re gonna meet with Christiana. So just ask her again to see if there was anything else that was added, since the last time you got a.
239 00:30:58.460 ⇒ 00:31:12.940 Josh: There’s gotta be like a or something that is getting like part of their process when they add stuff. Because I feel like they’re just forgetting because they don’t think about the data needs. And like, you guys, gotta be loud and be like, Hey, you guys, gotta add these things or we’re not gonna report them.
240 00:31:14.130 ⇒ 00:31:14.730 Josh: Yeah.
241 00:31:14.730 ⇒ 00:31:27.519 Josh: casual reminder for them. Because I just think I just have a gut feeling man like, usually I go with my instincts and my instincts are telling me that we’re like forgetting some stuff like it’s it’s not on you 100. But I look to you guys to just
242 00:31:27.700 ⇒ 00:31:31.699 Josh: continue to reiterate to them, hey, we need this in order to report it.
243 00:31:33.873 ⇒ 00:31:45.999 Demilade Agboola: So I would. I would look into that today just like majority products and try to try and see if there any like new product. Ids that that have been ordered, that don’t seem to be represented properly in our.
244 00:31:46.150 ⇒ 00:31:49.950 Josh: Cool daily reports cool. Thank you.
245 00:31:50.910 ⇒ 00:31:52.290 Robert Tseng: Alright! Thanks everyone.