Meeting Title: FAST Platform Order Status Sync Date: 2026-04-15 Meeting participants: Robert Tseng, Jasmin Multani, Amber Lin
WEBVTT
1 00:01:13.690 ⇒ 00:01:17.100 Robert Tseng: Hey, can you join the Zoom with, me and Jasmine?
2 00:01:23.300 ⇒ 00:01:24.170 Robert Tseng: Thanks.
3 00:01:24.280 ⇒ 00:01:24.990 Robert Tseng: Yep.
4 00:02:06.290 ⇒ 00:02:13.790 Amber Lin: Yeah, I think the one that pops. It doesn’t pop, but it’s like a garment. It’s coming from Jello. Okay.
5 00:02:13.790 ⇒ 00:02:15.810 Robert Tseng: Yeah, so I have the.
6 00:02:15.810 ⇒ 00:02:16.579 Amber Lin: There’s a long time ago.
7 00:02:16.580 ⇒ 00:02:21.330 Robert Tseng: The whole, like, the Slack thread pulled up here.
8 00:02:21.480 ⇒ 00:02:37.280 Robert Tseng: I think I’m calling… I just… I called because I feel like we went down a rabbit hole without really unders… I mean, unless things have changed, but from the beginning, we’ve always known that orders start and abandon. So, even if they are pending, if you’re just relying on order statuses.
9 00:02:37.280 ⇒ 00:02:37.610 Amber Lin: I was.
10 00:02:37.610 ⇒ 00:02:44.450 Robert Tseng: abandoned and petting, it, like, makes sense to me. So, I mean, there’s probably some more, like, filtering that we’ve added since then.
11 00:02:44.620 ⇒ 00:02:56.359 Robert Tseng: But, yeah, I think I… I saw somebody communicate that, like, hey, basically the expectation is we should only be looking at pending, and I don’t think that’s… I don’t think that’s right.
12 00:02:56.360 ⇒ 00:02:57.580 Amber Lin: So…
13 00:02:57.580 ⇒ 00:03:03.769 Robert Tseng: I just… I feel like there was a knowledge transfer gap here, like.
14 00:03:03.770 ⇒ 00:03:04.300 Amber Lin: Thank you.
15 00:03:04.300 ⇒ 00:03:10.670 Robert Tseng: I mean, I would have expected that everyone on this account would have understood the…
16 00:03:10.770 ⇒ 00:03:22.260 Robert Tseng: like, the order status isn’t vast, but it doesn’t seem like we understand, so I’m just kind of… I mean, I don’t know if I agree with the resolution here.
17 00:03:24.250 ⇒ 00:03:30.320 Amber Lin: Starting from my side, I think… There’s a misalignment between
18 00:03:30.490 ⇒ 00:03:33.399 Amber Lin: the order status in the FAST platform.
19 00:03:33.670 ⇒ 00:03:36.470 Amber Lin: and what we get through the API.
20 00:03:36.670 ⇒ 00:03:39.260 Amber Lin: We do get the status abandoned.
21 00:03:39.690 ⇒ 00:03:52.309 Amber Lin: from the API, when we get order completed, that event. When we get that event, it’s usually either abandoned or pending. That’s why we also get the status abandoned in Omni. Yep.
22 00:03:52.580 ⇒ 00:03:59.090 Amber Lin: That’s something that me and Jasmine’s investigating, there’s some misalignment there, but I agree, there’s… I think that’s just…
23 00:03:59.090 ⇒ 00:04:06.799 Robert Tseng: That’s just fast. BAS just will not… order statuses are just inaccurate, which is why we… we kind of default keep things to pending.
24 00:04:07.300 ⇒ 00:04:22.910 Robert Tseng: Until we’re clear that it’s canceled, or abandoned, or whatever. Yeah, so, like, yeah, anyway, so a 30% pending order rate, slightly higher than what it used to be. I think it used to be, like, 15 to 20%, so it should be…
25 00:04:23.120 ⇒ 00:04:30.979 Robert Tseng: I mean, data engineering should have an alert going off. Anything above 20% they should be telling us, which I feel like has just gone off the rails, so I will…
26 00:04:30.980 ⇒ 00:04:31.440 Amber Lin: I see.
27 00:04:31.440 ⇒ 00:04:32.410 Robert Tseng: To them as well.
28 00:04:32.590 ⇒ 00:04:32.980 Amber Lin: Okay.
29 00:04:32.980 ⇒ 00:04:37.690 Robert Tseng: for some reason, I mean, this 30% concerns me, because it’s usually not that high.
30 00:04:37.890 ⇒ 00:04:44.739 Amber Lin: Yeah, I… I agree. It is a quite high number. It… I checked in the Basque UI,
31 00:04:45.050 ⇒ 00:04:56.390 Amber Lin: it’s an even higher number. It’s at, like, 300-something total for, all orders, and that is also about 30%.
32 00:04:56.560 ⇒ 00:05:12.789 Amber Lin: So, like, our numbers are close, but, like, that’s something their farm ops team, or something that they need to look into. But going back to your point, I do think there’s a lack of knowledge transfer, and, like, as me and Jasmine work on this, I think we really want to
33 00:05:13.240 ⇒ 00:05:29.290 Amber Lin: we set up the prop of the data documentation that you’ve been the only person doing. I think that’s something that me and Jasmine can put our hands on and do, because, like, I was also very lost. I was like, I think I know what it is, but I’m not sure what it is.
34 00:05:29.500 ⇒ 00:05:30.280 Amber Lin: So…
35 00:05:30.280 ⇒ 00:05:30.820 Robert Tseng: out.
36 00:05:32.040 ⇒ 00:05:49.489 Robert Tseng: Okay, yeah, I know that, yeah, I’m definitely a little bit more heavy-handed with documentation than, like, who Tom is, like, he’s a lot more just like, whatever, just look at cursor, but I’m very much like, no, unless somebody’s maintaining it, like, nobody understands what’s going on, so…
37 00:05:49.960 ⇒ 00:05:53.269 Robert Tseng: I think you should… you should take that over. I mean, I…
38 00:05:53.430 ⇒ 00:06:02.759 Robert Tseng: I think this week I got derailed quite a bit because I had to get pulled into Eden a couple times, so… and that, you know, I’m a little bit, like, kind of…
39 00:06:03.040 ⇒ 00:06:22.290 Robert Tseng: I mean, I can come back in in a more involved way, but, like, you know, I’m just a little bit more sensitive on this client, because, like, it’s not just this situation, it’s, like, data engineering dropped the ball, Greg didn’t do anything on product analytics the past week, like, we basically, like, looked like there was a lot of
40 00:06:22.780 ⇒ 00:06:26.329 Robert Tseng: in the channels, but I don’t think we really delivered much as a team in this past.
41 00:06:26.330 ⇒ 00:06:27.180 Amber Lin: I hear you.
42 00:06:27.180 ⇒ 00:06:32.760 Robert Tseng: I think we’re just trying to… and I think they’re… they’re on… they’re on my… they’re on my
43 00:06:33.100 ⇒ 00:06:34.149 Robert Tseng: because of that, so I’m…
44 00:06:34.150 ⇒ 00:06:48.370 Amber Lin: I hear you. And you’re the only person getting the pressure, because you’re the, like, they talk to you, and, like, we’re all kind of behind you, so, like, if you want to put pressure on us, like, I think that’s very understandable. And you should.
45 00:06:49.790 ⇒ 00:07:04.680 Robert Tseng: Yeah, so I believe that these are solvable problems, you know, this, like I said, you know, 30% is not, like, outlandish. I think this takeaway that you’re telling me makes sense. I would have preferred to hear that in the Slack thread, I would say, because I think.
46 00:07:04.680 ⇒ 00:07:05.070 Amber Lin: I see.
47 00:07:05.070 ⇒ 00:07:16.130 Robert Tseng: there’s all this, like, breakdowns of, like, what your findings are. I don’t really need to read all of that. I think I would prefer more, like, hey, I looked into it, I actually think that, like, pharmacy just needs to go, like, check their… check their orders, and, like.
48 00:07:16.130 ⇒ 00:07:16.830 Amber Lin: Yeah.
49 00:07:16.830 ⇒ 00:07:33.140 Robert Tseng: if you didn’t have the baseline of the 20% before, I mean, ideally, data engineering would have jumped in and been like, hey, look, you know, baseline should be 20, 30% is higher than what we would expect. Like, this is probably something that we should revisit in our logic. Are we missing something?
50 00:07:33.140 ⇒ 00:07:37.340 Robert Tseng: and you’re calling out, hey, there could be an API, like, issue here.
51 00:07:37.340 ⇒ 00:07:43.999 Robert Tseng: I’m okay with, like, those are all good… those are all good things that I’m… I’m looking for, because it helps me to kind of respond quickly to them. Got it.
52 00:07:44.000 ⇒ 00:07:58.010 Amber Lin: Gotcha. So, communications to you will just… it’s be similar to, like, client-level extrapolation, and I’ll give, like, the details to Jasmine. I think me and Jasmine were writing out our thought process as we were writing this.
53 00:07:58.010 ⇒ 00:08:08.560 Robert Tseng: Yeah, which is good. I think that helps me to audit, kind of, where things are going, where you guys… so I’m okay with you writing the full thing out. I just… I need the… I need the kind of point of view at the end of it, I guess.
54 00:08:08.560 ⇒ 00:08:08.970 Amber Lin: -
55 00:08:08.970 ⇒ 00:08:14.339 Robert Tseng: I already kind of just made assumptions, and I told them… I think someone mentioned.
56 00:08:14.340 ⇒ 00:08:14.810 Amber Lin: similar.
57 00:08:14.810 ⇒ 00:08:22.340 Robert Tseng: Yeah, Jasmine’s question was, hey, do you think it was from the API? Like, I just ran with that. I immediately kind of went back to Josh and Matasha was like.
58 00:08:22.480 ⇒ 00:08:23.000 Amber Lin: I’m teasing.
59 00:08:23.000 ⇒ 00:08:33.040 Robert Tseng: thinks it’s an API issue. They’re like, okay, whatever. They’re not gonna, like, poke around more into it. So, that helps me to de-escalate it immediately, because it shows that, like, okay.
60 00:08:33.039 ⇒ 00:08:45.680 Robert Tseng: we have some way to root cause what’s going on, they’ll get us a little more time, so I think that was… that was the right move. I… I guess I understand that both of you were kind of just digging into it more later and just leaving the paper trail, so…
61 00:08:45.680 ⇒ 00:08:46.160 Amber Lin: Yeah.
62 00:08:46.160 ⇒ 00:08:56.529 Robert Tseng: Anyway, so I don’t think I’m, like, trying to criticize, kind of, like, the, like, the, like, what happened here. I just am trying to just get to the, get to the point of, like, okay, well, actually.
63 00:08:56.790 ⇒ 00:09:00.699 Robert Tseng: maybe we’re just misunderstanding, like, pending and abandoned, because I…
64 00:09:00.700 ⇒ 00:09:03.319 Amber Lin: Yeah, I think, I think so. Yeah.
65 00:09:03.620 ⇒ 00:09:06.709 Amber Lin: To conclude, like, to finalize that understanding.
66 00:09:07.250 ⇒ 00:09:17.079 Amber Lin: Are we… what is our final decision? Are we using abandoned, or are we using pending and abandoned grouped together? Because…
67 00:09:17.210 ⇒ 00:09:24.159 Robert Tseng: Yeah, I am a bit… like, last time I made this decision was when I created the For Josh report back in, like.
68 00:09:24.160 ⇒ 00:09:24.610 Amber Lin: Yeah.
69 00:09:24.610 ⇒ 00:09:25.260 Robert Tseng: last year.
70 00:09:25.260 ⇒ 00:09:25.710 Amber Lin: Yeah.
71 00:09:25.710 ⇒ 00:09:28.110 Robert Tseng: So, it’s okay if things have changed. Like, I think.
72 00:09:28.450 ⇒ 00:09:29.100 Amber Lin: Okay.
73 00:09:29.100 ⇒ 00:09:40.219 Robert Tseng: I wouldn’t say that, like, I’m attached to what we defined before. I think we just have to draw a line in the sand and say, this is how we should do it moving forward. And, like, I think I can…
74 00:09:40.220 ⇒ 00:09:40.640 Amber Lin: I see.
75 00:09:40.640 ⇒ 00:09:45.240 Robert Tseng: talk to them with that. So, I think it is a time for us to reinvent it a bit, yeah.
76 00:09:45.240 ⇒ 00:10:01.989 Amber Lin: Yeah, okay, sounds good, and that’s good to know we have flexibility. I think moving forward, I’ll have both the pending and abandoned there, because I remember in the 7 days report, there was one that excluded those orders, so I guess they want to see both of them.
77 00:10:02.160 ⇒ 00:10:07.569 Amber Lin: So, the last thing I want to confirm is, do we want, like, pending orders, or do we want pending
78 00:10:07.690 ⇒ 00:10:12.649 Amber Lin: customers, because that’s what they’re used to seeing. It was customers, it was not orders.
79 00:10:12.650 ⇒ 00:10:17.529 Robert Tseng: Exactly. Pending customers is what they care about. Pending orders, they don’t actually care about, yeah.
80 00:10:17.530 ⇒ 00:10:23.309 Amber Lin: Cool, okay. So, pending, like, new customers or pending customers?
81 00:10:23.310 ⇒ 00:10:29.389 Robert Tseng: Well, in that report, they’re looking at first-time customers. That’s… that’s typically what they care about the most. I mean, it’s…
82 00:10:30.700 ⇒ 00:10:37.760 Robert Tseng: important to… I mean, you should have both, but, like, they’re really looking at that report daily to see, like,
83 00:10:38.240 ⇒ 00:10:39.220 Robert Tseng: They’re…
84 00:10:39.220 ⇒ 00:10:42.640 Amber Lin: Like, anyone yesterday that coming in for us.
85 00:10:42.640 ⇒ 00:10:51.269 Robert Tseng: Right, yeah, it’s like, okay, we saw they were pending yesterday, did they actually convert to new today? And like, you know, and I think that’s… that’s the signal that they’re paying attention to.
86 00:10:51.270 ⇒ 00:11:08.140 Amber Lin: Sounds good. Yeah, and I know FarmOps looks at it more detailed. I’m, like, before we stopped, I was working with them on, like, their pending report. Okay, so clear on that, I’ll… I’ll do it based on customers, new customers especially, and they’ll… I’ll include both, pending and abandoned.
87 00:11:08.840 ⇒ 00:11:16.050 Robert Tseng: Yeah. I mean, you looked at the AMBLE data, like, what… what… what are they doing? Have you seen… have you, have you… did you ever notice that?
88 00:11:16.200 ⇒ 00:11:20.679 Amber Lin: What are they doing on what? Like, statuses, or…
89 00:11:20.680 ⇒ 00:11:24.440 Robert Tseng: Yeah, like, are they… are they filtering on statuses? I never looked at their data models.
90 00:11:24.670 ⇒ 00:11:31.179 Amber Lin: I never saw their reports. All I can see is their… is their data.
91 00:11:32.010 ⇒ 00:11:36.740 Robert Tseng: Oh, okay, so everything you’re seeing is just order of revenue, it’s not really, like, at the order status level, I see.
92 00:11:36.740 ⇒ 00:11:37.340 Amber Lin: No.
93 00:11:37.650 ⇒ 00:11:52.909 Robert Tseng: Okay, well, my guess is, since it’s the same team that built this… they’re probably not… don’t really… they didn’t solve this problem. Okay, anyway, I was just… I was just curious to see if there is another… a better way of doing this in… in VASC, but it seems like…
94 00:11:52.910 ⇒ 00:12:00.990 Amber Lin: No, not really. The… if you want to hear… do you want to know what’s wrong with the API, I can give it to you in one sentence.
95 00:12:00.990 ⇒ 00:12:02.310 Robert Tseng: Sure.
96 00:12:02.620 ⇒ 00:12:08.050 Amber Lin: Okay, there’s multiple statuses in BASC. Phys status, pharmacy status, etc.
97 00:12:08.090 ⇒ 00:12:27.399 Amber Lin: And we get from the API once when it’s order completed event and the various order updated events. The order updated events include… our hypothesis is that it includes all these statuses, so our fields are pretty messed up. It does not map to order status.
98 00:12:29.060 ⇒ 00:12:34.210 Robert Tseng: Order updated does not update our order status.
99 00:12:34.730 ⇒ 00:12:44.900 Amber Lin: order updated does… might have… it doesn’t directly map to order status field updating. Something may have changed in pharmacy status.
100 00:12:45.430 ⇒ 00:12:49.160 Amber Lin: Just so go in. And then it’ll… it’ll mess it up.
101 00:12:51.610 ⇒ 00:12:54.450 Robert Tseng: Yeah, it’s because they keep changing the,
102 00:12:55.480 ⇒ 00:12:57.869 Robert Tseng: What they’re calling order updated status.
103 00:12:58.260 ⇒ 00:12:58.890 Robert Tseng: Whatever.
104 00:12:59.270 ⇒ 00:13:00.970 Robert Tseng: Yeah.
105 00:13:02.260 ⇒ 00:13:08.040 Robert Tseng: Yeah, so then we… then we, then we are no longer to map the new things to the existing, status.
106 00:13:08.390 ⇒ 00:13:08.790 Robert Tseng: before.
107 00:13:08.790 ⇒ 00:13:14.799 Amber Lin: And that also leads to, like, the 30-day revenue discrepancy that me and find.
108 00:13:14.800 ⇒ 00:13:18.230 Robert Tseng: Yeah, it creates, like, a fan-out for order statuses, people are kind.
109 00:13:19.440 ⇒ 00:13:27.459 Amber Lin: Yeah, and because they filter out, like, canceled and error, but our canceled and error is different, so it just gets…
110 00:13:27.800 ⇒ 00:13:30.050 Amber Lin: It’s… is not going to align.
111 00:13:30.920 ⇒ 00:13:31.600 Robert Tseng: Okay.
112 00:13:33.740 ⇒ 00:13:40.980 Robert Tseng: Okay, well, this is what I think you should take back to data engineering with, which is…
113 00:13:40.980 ⇒ 00:13:41.659 Amber Lin: being, like…
114 00:13:41.660 ⇒ 00:13:45.870 Robert Tseng: similar to what we did with the product lists. We don’t get this anymore, but, like.
115 00:13:46.780 ⇒ 00:13:53.559 Robert Tseng: Before, if you remember, we used to be… I used to have Demulade ask Bast every week for a product, updated product list.
116 00:13:54.120 ⇒ 00:14:03.140 Robert Tseng: That way, we would be able to tell when product names have changed, and we can go and move products to the right bucket in case we needed to do that reactively.
117 00:14:03.160 ⇒ 00:14:12.880 Robert Tseng: We kind of need a similar, like, regular mapping for order statuses, that is. I mean, it’s probably not every week, like, I doubt they’re adding new statuses every week on, like, the products.
118 00:14:12.880 ⇒ 00:14:23.210 Robert Tseng: So maybe it’s a little bit less tedious, but at least we need to be getting… we need to be getting something from BASS that’s, like, telling… telling us, like, what order statuses are… are… are… are coming.
119 00:14:23.230 ⇒ 00:14:30.450 Robert Tseng: So I feel like we should ask for that… we should ask data engineering to request that from BASC and get that monthly. I feel like that would at least help.
120 00:14:30.590 ⇒ 00:14:32.660 Robert Tseng: Us, like, not have to…
121 00:14:32.780 ⇒ 00:14:35.719 Robert Tseng: Like, yeah, react to it as quickly.
122 00:14:35.720 ⇒ 00:14:36.270 Amber Lin: Yup.
123 00:14:36.400 ⇒ 00:14:37.530 Amber Lin: Sounds good.
124 00:14:37.790 ⇒ 00:14:44.199 Amber Lin: Okay, I’ll… I’ll think about… I’ll draft that with Jasmine, and then we’ll move forward.
125 00:14:44.520 ⇒ 00:14:52.799 Robert Tseng: Okay. Yeah, and then this whole, like, shifting to the API numbers is a… is a new… is a new thing. Like, this obviously wasn’t there 3 months ago, so…
126 00:14:52.940 ⇒ 00:15:02.099 Robert Tseng: Yeah, I understood that, like, that it’s gonna just be a little bit funky, man, a mix of webhooks and APIs, like, you don’t really know which one to trust.
127 00:15:02.230 ⇒ 00:15:07.820 Robert Tseng: Yeah, I understand that this is just, like, a challenge, so… I don’t…
128 00:15:07.820 ⇒ 00:15:16.889 Amber Lin: I think the good thing is that they understand, too. It’s just, like… like, as long as we tell them what is wrong, they would understand.
129 00:15:17.350 ⇒ 00:15:18.000 Robert Tseng: Okay.
130 00:15:18.270 ⇒ 00:15:18.860 Amber Lin: Yeah.
131 00:15:19.910 ⇒ 00:15:20.910 Robert Tseng: Okay, cool.
132 00:15:20.910 ⇒ 00:15:21.820 Amber Lin: Okay.
133 00:15:22.040 ⇒ 00:15:29.259 Robert Tseng: I think that’s all I got. I feel good with the takeaway. I already kind of told them that, like.
134 00:15:29.440 ⇒ 00:15:38.530 Robert Tseng: I think, if anything, I will also suggest to them that, hey, pending orders, there’s a new… like, we’re focusing on pending customers, and, like.
135 00:15:38.530 ⇒ 00:15:38.910 Amber Lin: Excellent.
136 00:15:38.910 ⇒ 00:15:48.789 Robert Tseng: That’s, yeah, I don’t know, if you want to draft me that message, I want to clarify, like, I want to… I think you guys should make a decision on what… what… how it should be modeled, and so I can set that expectation with that.
137 00:15:49.110 ⇒ 00:15:57.780 Amber Lin: Yeah, okay. Yeah, and I’ll… I’ll send that over, and I’ll make a draft dashboard so that I can see what it looks… what it can look like.
138 00:15:58.660 ⇒ 00:15:58.980 Robert Tseng: Okay.
139 00:15:59.540 ⇒ 00:16:00.060 Jasmin Multani: Okay.
140 00:16:00.770 ⇒ 00:16:01.329 Amber Lin: Thank you.
141 00:16:02.110 ⇒ 00:16:02.780 Robert Tseng: Thanks.
142 00:16:03.020 ⇒ 00:16:03.880 Robert Tseng: Bye.