Meeting Title: [Internal] Eden Onboarding Date: 2025-07-09 Meeting participants: Robert Tseng, Amber Lin, Henry Zhao
WEBVTT
1 00:03:25.660 ⇒ 00:03:26.720 Amber Lin: Hi!
2 00:03:27.690 ⇒ 00:03:28.660 Amber Lin: And then.
3 00:03:29.200 ⇒ 00:03:30.140 Robert Tseng: Hey! Amber.
4 00:03:30.510 ⇒ 00:03:34.729 Amber Lin: Hello! I got out of the house in the Starbucks right now.
5 00:03:34.970 ⇒ 00:03:35.880 Robert Tseng: Nice.
6 00:03:35.880 ⇒ 00:03:36.935 Amber Lin: Yeah.
7 00:03:38.860 ⇒ 00:03:45.130 Amber Lin: Are we having this meeting with him, or is he? Is he joining.
8 00:03:45.720 ⇒ 00:03:47.400 Robert Tseng: He did say he was gonna join. So.
9 00:03:47.400 ⇒ 00:03:47.890 Amber Lin: Okay.
10 00:03:47.890 ⇒ 00:03:49.519 Robert Tseng: I’m I’m expecting him.
11 00:03:49.930 ⇒ 00:03:51.040 Amber Lin: Okay, bye-bye.
12 00:03:51.350 ⇒ 00:03:56.049 Amber Lin: Do we want me on Eden, or do we want me on the new clients?
13 00:03:57.510 ⇒ 00:04:04.879 Amber Lin: Take Eden? I won’t be able to take anything else. Eden is so so complicated.
14 00:04:05.580 ⇒ 00:04:11.215 Robert Tseng: Yeah, I talked to you, Tom, he he! We’re not sure yet, but he said to just like loop you in for now.
15 00:04:12.330 ⇒ 00:04:16.280 Amber Lin: Yeah bare minimum. If there’s any fraud management tasks.
16 00:04:16.660 ⇒ 00:04:29.230 Amber Lin: I can help honest, like, I originally thought maybe I could run the stand ups and then do the other rituals with you. But then I saw the standups today, and the stakeholders are very demanding.
17 00:04:31.990 ⇒ 00:04:35.670 Robert Tseng: Yeah, yeah, they they will. They’re pretty hard.
18 00:04:35.670 ⇒ 00:04:40.260 Amber Lin: This language I was like, Wow! Is this, is this a professional meeting?
19 00:04:41.420 ⇒ 00:04:43.510 Robert Tseng: Oh, yeah, well.
20 00:04:43.990 ⇒ 00:04:50.019 Robert Tseng: yeah. And and he’s 1 of the 3 execs, and he’s the nicest of the 3. So.
21 00:04:50.020 ⇒ 00:04:52.670 Amber Lin: Oh, really! Oh, God!
22 00:04:53.060 ⇒ 00:04:56.240 Amber Lin: I guess it’s hard earned money, indeed!
23 00:04:57.036 ⇒ 00:05:07.779 Robert Tseng: Yeah, yeah, wow, let me just quickly check on email. He did say he was, gonna join.
24 00:05:23.150 ⇒ 00:05:24.010 Robert Tseng: what the hell?
25 00:05:35.380 ⇒ 00:05:43.600 Robert Tseng: Okay, I just I just ping them. Give him a couple of minutes join, but otherwise it’s fine. We can get started without him. I’ll just this is recorded, anyway, so I’ll send it to him.
26 00:05:47.080 ⇒ 00:05:50.570 Robert Tseng: Yeah, I mean, what can I say else about it?
27 00:05:50.920 ⇒ 00:05:56.920 Amber Lin: I know all the different projects that’s happening. I guess if
28 00:05:57.100 ⇒ 00:06:15.890 Amber Lin: if we still assume that we want me to take over needed do you want to start with? Is a 1 of the projects? Or do you want to just start with taking over all the stand ups and kind of washing over like what do you think is the best way for us to approach it?
29 00:06:16.560 ⇒ 00:06:19.182 Robert Tseng: Yeah, well, so I think, like
30 00:06:21.620 ⇒ 00:06:29.849 Robert Tseng: the contingency is like, well, so I mean Henry’s Henry’s gonna help with this one project, which is the Cdp thing.
31 00:06:31.230 ⇒ 00:06:36.980 Robert Tseng: now, if he likes it, and he wants to stay on then, like I think he can, I think he’s senior enough to replace me.
32 00:06:37.230 ⇒ 00:06:40.289 Robert Tseng: which would be ideal so that I can go. And
33 00:06:40.652 ⇒ 00:06:44.300 Robert Tseng: I think, yeah, that would that would free me up to go and try to bring in more business.
34 00:06:46.430 ⇒ 00:06:58.956 Robert Tseng: yeah. And so then you would probably shift to like working under him. And like, yeah, I don’t know how much he would want to cover like we. It would kind of be up to his decision, I think.
35 00:06:59.490 ⇒ 00:07:15.788 Robert Tseng: on like, how how much, how much she wants to involve you. But I’m also like I’m not gonna bank on on that like, I think that’s that’s that’s 1 scenario. The other scenario is that like I keep running it? And then, yeah, I would have you eventually kinda come in to help
36 00:07:17.070 ⇒ 00:07:21.722 Robert Tseng: Yeah. Running standups would probably still stay with me.
37 00:07:22.590 ⇒ 00:07:24.630 Robert Tseng: yeah, just because, like the standard.
38 00:07:24.630 ⇒ 00:07:26.249 Amber Lin: A lot of stakeholders.
39 00:07:26.250 ⇒ 00:07:40.690 Robert Tseng: People just come in whenever they want, and they just like they’ll. They’ll like, just talk, whatever they want to say. And yeah, it’s a lot of just thinking on your feet quickly and like kind of just setting expectations and stuff. And I have the relationship. So I think it’s kind of.
40 00:07:41.020 ⇒ 00:07:44.404 Robert Tseng: I I think that’d be pretty hard to hand off like I I don’t think
41 00:07:44.770 ⇒ 00:07:50.200 Robert Tseng: I mean yeah. So I think I would probably stay for that. But then, as far as like.
42 00:07:50.380 ⇒ 00:07:55.159 Robert Tseng: yeah, just Async work of like following up with the rest of the team
43 00:07:55.670 ⇒ 00:08:02.079 Robert Tseng: like kind of grooming grooming tickets and and like managing the backlog, and everything would probably
44 00:08:02.280 ⇒ 00:08:11.229 Robert Tseng: late on you to to help me with that. So maybe it’s more kind of just like internal Pm. And less like client facing. Pm, work. We could talk about that later on.
45 00:08:11.230 ⇒ 00:08:12.770 Robert Tseng: Sure, sure, totally.
46 00:08:13.100 ⇒ 00:08:16.419 Amber Lin: I’ll just sit in. I’ll hear your calls. Hi, Harry!
47 00:08:16.830 ⇒ 00:08:17.760 Henry Zhao: Hey, guys.
48 00:08:19.750 ⇒ 00:08:21.679 Robert Tseng: Hey, Henry, how was your how’s your vacation?
49 00:08:21.680 ⇒ 00:08:26.529 Henry Zhao: It’s good. Thanks. Yeah. I was just there for a wedding. So now that the wedding is done. Yeah.
50 00:08:26.770 ⇒ 00:08:29.460 Robert Tseng: Nice. Where are you in Brazil, or are you in Arizona?
51 00:08:30.111 ⇒ 00:08:31.560 Henry Zhao: Right now I’m still in Brazil.
52 00:08:31.560 ⇒ 00:08:37.689 Robert Tseng: Okay, cool. And then this kind of like, you’re you trying to, are you just gonna stay there? Are you trying to come back to the States?
53 00:08:38.110 ⇒ 00:08:39.869 Henry Zhao: I’m kind of back and forth right now. Yeah.
54 00:08:39.870 ⇒ 00:08:41.890 Robert Tseng: And forth. Okay, yeah, that’s cool.
55 00:08:43.600 ⇒ 00:08:58.129 Robert Tseng: well, yeah, I mean, this is, this is amber. Amber is a Pm. On our on our team. I’m also kind of just like kinda 2 birds, one stone. Try to also bring her onto this project in some surpass capacity as well.
56 00:08:58.473 ⇒ 00:09:15.619 Robert Tseng: Yeah, I mean, I think, like, the main objective is really just to kind of. I I’ve given you kind of the overview of the Cdp kind of work there for Eden. And so that’s like the main project that’s going on that we’re trying to get to a vendor decision. You know, ironically decide whether or not we’re gonna go with broker stack
57 00:09:16.182 ⇒ 00:09:22.950 Robert Tseng: so this been long time coming. And then and there’s a bunch of other parallel work streams that I probably would want to intro
58 00:09:23.239 ⇒ 00:09:50.900 Robert Tseng: to to her and and for you, if you’re if you’re interested, I think it’s kind of like after this project concludes like, if you were to stay on and keep working with us like. And you want to take over more of of this client. I think that would be my ideal scenario. I I think that’s that’s kind of what I’m hoping we’ll we’ll we’ll we’ll get to. But I think for this call. I will just do like an overview of the clients, and then we’ll talk more specifically about this one project.
59 00:09:51.231 ⇒ 00:09:59.599 Robert Tseng: And then any questions that you have of like how we kind of get you kind of plugged in. I think that’s that’s that’s probably what we’ll have time for today.
60 00:09:59.740 ⇒ 00:10:00.439 Robert Tseng: Does that sound good.
61 00:10:00.440 ⇒ 00:10:02.399 Henry Zhao: Sounds. Great. Yeah. Sounds. Perfect.
62 00:10:02.720 ⇒ 00:10:12.160 Robert Tseng: Cool alright, so I will share my screen. And then this, give me a second.
63 00:10:12.610 ⇒ 00:10:23.297 Robert Tseng: Alright. So couple of things I’m just gonna talk through like the various systems that we use as as we’re doing it. So every client has like a notion page. So this is the even notion page. We’ll have access to all of this.
64 00:10:23.640 ⇒ 00:10:29.201 Robert Tseng: I think between you, you and Rico will kind of. He’ll get you sorted out all the logins and everything.
65 00:10:29.960 ⇒ 00:10:43.035 Robert Tseng: but yeah, I think for for this client, we basically are there, we are their data team. So they yeah, we’re basically I have. I guess I’ll update this as we go. But we have.
66 00:10:44.040 ⇒ 00:10:47.860 Robert Tseng: really, that’s active right now.
67 00:10:54.900 ⇒ 00:11:05.009 Robert Tseng: we have like 4 people on this project right now from our side. And then, like, there’s a bunch of stakeholders enough, no names or anything. Yeah. I mean, the idea is that like, we don’t really have like
68 00:11:05.818 ⇒ 00:11:23.719 Robert Tseng: I mean I my, my time is like pretty pretty stretched, thin. And so I think I want I’ve been wanting some more support on the project management side. And then also, just for like a like a head of solutions, or like strategic lead kind of perspective, that’s kind of what I’m trying to to bring bring onto this team.
69 00:11:24.160 ⇒ 00:11:39.829 Robert Tseng: And then as far as like you can go through. Come through any of these artifacts what we’ve done to date for them. We you can kind of look into any of these projects. And linear. So linear is like the the project management tool that we use. And
70 00:11:40.160 ⇒ 00:11:41.220 Robert Tseng: yeah, I think
71 00:11:41.330 ⇒ 00:11:47.139 Robert Tseng: most of these are pretty much up to date. It gets a bit crazy in here, and I think it just at a high level
72 00:11:47.240 ⇒ 00:12:06.270 Robert Tseng: we came in. They were kind of operating their business around with with Google sheets, you know, 6 6 months ago, just had segment didn’t have a data warehouse. We kind of set up a data warehouse for them in bigquery we started routing all that that data from segment into into bigquery. And then.
73 00:12:06.530 ⇒ 00:12:14.389 Robert Tseng: yeah, eventually, we disconnected a bunch of sources from segment. And we’re just like either went direct with them, using an etl tool called polytomic
74 00:12:14.771 ⇒ 00:12:40.320 Robert Tseng: or we just like found other or or we, or we kind of re platform them on on other tools as well. So I think there’s a pretty complicated like we’re just. There’s many tools in that stack that don’t need to necessarily talk about. But it was just all setting up all that, Etl. We built out all their dbt marts, and we do all the orchestration through through a mix of Dbt and dag and daxter
75 00:12:40.915 ⇒ 00:12:56.999 Robert Tseng: and so now and then the end and the main bi tool is in tableau. So I think that’s been the primary kind of work that we’ve been owning up to date. And then in the in the past couple of months, there’s been kind of
76 00:12:57.370 ⇒ 00:12:58.530 Robert Tseng: a gap
77 00:12:58.960 ⇒ 00:13:06.880 Robert Tseng: that to fill where I think the team just wants more self service kind of tooling, and wanting to see
78 00:13:07.851 ⇒ 00:13:21.410 Robert Tseng: like to get more, because there is a mixed panel tool floating around. But it’s not. It’s not heavily used. And so yeah, we just haven’t put too much. We haven’t put many resources at that to to get it ready into a place
79 00:13:21.843 ⇒ 00:13:33.090 Robert Tseng: for 1st kind of non technical stakeholders to go and do exploration. We just have some basic reporting kind of floating around in there. So maybe I’ll just kind of flash a few different things as I’m
80 00:13:33.310 ⇒ 00:13:39.409 Robert Tseng: of like talking through it all. So I’ll flash our mix panel.
81 00:13:41.480 ⇒ 00:13:46.000 Robert Tseng: There’s also an architecture diagram somewhere. This one.
82 00:13:50.580 ⇒ 00:13:53.619 Robert Tseng: I think this is just a generic one. Never mind.
83 00:14:05.170 ⇒ 00:14:07.820 Robert Tseng: 5, 4.
84 00:14:13.170 ⇒ 00:14:20.420 Robert Tseng: Yeah, so, okay, yeah, so this is more or less kind of like, what I was just walking through.
85 00:14:20.755 ⇒ 00:14:49.350 Robert Tseng: These are, now, this is the current state of the raw data sources kind of like how these various tools are being brought into bigquery and then we have, like a whole selection of DVD marks. This probably needs to be updated. We’ve added some stuff since then. But I think we’ve gotten pretty good at reporting out on transactional data. So all things sales by products and orders. And we were able to do. And then on the marketing side of the house. We’ve been able to help with doing kind of just like
86 00:14:49.410 ⇒ 00:14:56.330 Robert Tseng: attribution reporting. It’s pretty basic. There isn’t really too much more than just like making sure that we’re able to
87 00:14:56.870 ⇒ 00:15:10.110 Robert Tseng: consider their entire marketing budget, not just the paid channels, but all offline channels and paid channels as well. Having some sense of like a 1st touch attribution. For that, or like sorry, like a last touch attribution
88 00:15:10.495 ⇒ 00:15:27.069 Robert Tseng: read out for that. Now, there’s like they’re evolving to consider things like incrementality and wanting to consider, predicted Ltv. And these other more forward thinking metrics that our team is assisting with. But that’s not like that’s that’s a parallel effort to what’s going on here
89 00:15:27.090 ⇒ 00:15:42.679 Robert Tseng: and then on the operational side, all things. A customer service related. This is all that Zendesk data coming in from like the other customer experience team and tickets agent performance, we’re able to report on pretty clearly.
90 00:15:43.750 ⇒ 00:15:44.630 Robert Tseng: Order.
91 00:15:44.740 ⇒ 00:15:50.737 Robert Tseng: And then, from the order perspective there’s a couple of different entities that are involved here.
92 00:15:51.230 ⇒ 00:15:55.119 Robert Tseng: basically, the the customer journey is
93 00:15:55.480 ⇒ 00:15:59.549 Robert Tseng: they’ll go onto the website and they’ll
94 00:15:59.610 ⇒ 00:16:28.090 Robert Tseng: they’ll have to fill out an intake form here which basically pre qualifies them to be able to get access to prescription meds, which mostly are glp, one related supplements. And then from there there’s there’s a appointments kind of a telehealth engagement just like a 1 time appointment with a health provider similar to what you would see at like Rose, or hims, or hers, or whatever and then it would be like a subscription kind of Ecom model, where
95 00:16:28.310 ⇒ 00:16:49.689 Robert Tseng: you know, when you purchase your meds, you’re not. It’s not just a 1 time purchase. You’re typically on a plan that’s like a 6 month or annual plan, and then you’re kind of get you get your distributions monthly. So all, all of that is being handled through like a a few different pharmacy pharmacies. That kind of process the orders and then they ship them out
96 00:16:50.191 ⇒ 00:17:10.419 Robert Tseng: and so kind of all of those different entities from like the back end ordering system, which it’s not shopify. They have their own system called bask to the pharmacies. That’s we were originally not going direct to these pharmacies to integrate with them. But we’re heading in that direction now, because there’s been some data quality issues we’ve had to deal with.
97 00:17:11.095 ⇒ 00:17:24.210 Robert Tseng: And then all of the kind of like logistics. Data comes from Shippo, which is just kind of a shipping label provider. They help us capture the Timestamps from once the order leaves the pharmacy to when it gets to the customer doorstep.
98 00:17:24.585 ⇒ 00:17:50.579 Robert Tseng: But that’s pretty much it. It’s a pretty simple business model subscription. Ecom. I think you’re familiar with that, Henry. So I know I’m just kind of like breezing through that but that’s kind of the lay of the land of like the how, what data we have, what we’ve modeled so far, and what’s being used in reporting and then kind of just like very lightly touched on like directionally, like where we’re headed. As you can see, nothing really has been as built
99 00:17:50.925 ⇒ 00:18:04.130 Robert Tseng: for the behavioral analytics side. I think a lot of the tracking already exists. We do have segments set up. We do have mixed panel there. I just think it’s not being fully utilized. And so that’s that’s why we’re doing this kind of Cdp transformation project.
100 00:18:04.888 ⇒ 00:18:14.349 Robert Tseng: But I’ll just pause there. Just kind of see you know, both of you are still following. If you have any questions. Just kind of from from that overview. So far.
101 00:18:16.940 ⇒ 00:18:18.970 Henry Zhao: No, no questions for me on my end.
102 00:18:19.360 ⇒ 00:18:25.219 Robert Tseng: Okay, how about you, Amber? I know you. You Hi! Hi!
103 00:18:25.220 ⇒ 00:18:28.839 Robert Tseng: Looked at our work on urban stems and like other clients as well. So want to know. Kind of.
104 00:18:28.840 ⇒ 00:18:34.150 Amber Lin: I have a general understanding of what what we are doing.
105 00:18:34.500 ⇒ 00:18:34.850 Robert Tseng: Yes.
106 00:18:34.850 ⇒ 00:18:46.229 Amber Lin: If the stakeholder asked me specific question, I cannot answer it yet, like, I have a general idea what we’re doing where things are. But if it’s anything specific, I cannot answer that.
107 00:18:46.490 ⇒ 00:18:57.740 Robert Tseng: Okay, yeah, no, listen, no worries. I know there’s a lot of like, you know, nuances to this that we’ll kind of. We’ll you’ll you’ll you’ll pick up as you kind of sit in on more meetings and get to know the work that we’re actually doing.
108 00:18:58.150 ⇒ 00:19:08.910 Robert Tseng: So yeah, I think with that in mind. I’ll just kind of quickly show kind of where we’re at mixed panel. Wise because I think this is the visualization tool that we use for behavioral stuff.
109 00:19:08.960 ⇒ 00:19:31.880 Robert Tseng: So what I’ve set up so far is really just like a basic like, kind of Google analytics replacement. More or less, you can think of it that way. This is a web. Analytics, dashboard kind of just shows them like basic kind of like site visit trends, conversion trends kind of like what we call site intake site to check out site to order completed and then just like basic things on like volume page, so that we could track.
110 00:19:31.930 ⇒ 00:19:51.590 Robert Tseng: you know, as they’re launching new piece of content across the page, like how that’s actually doing. There’s no really, for by experiments this is just like the only real experimentation that’s going on at this company is not really from like a landing page or design kind of perspective. Those things do get iterated on, but doesn’t don’t happen very quickly.
111 00:19:52.850 ⇒ 00:19:53.600 Robert Tseng: Huh!
112 00:19:54.120 ⇒ 00:20:19.741 Robert Tseng: I think it’s because there’s a cookie that kind of automatically brings you back to the intake. But it’s really the main experiments are in these intakes. So there are various landing pages. That get triggered that you know that we attach to, you know, ads. And then once users come in, they all get, you know, some funnel, which is just like a V various forms of questionnaires. So that’s like the main thing that they look at.
113 00:20:20.490 ⇒ 00:20:24.070 Robert Tseng: And then, yeah, we have some like basic performance metrics around that.
114 00:20:24.691 ⇒ 00:20:42.859 Robert Tseng: From there, I think. There, there’s some legacy stuff that’s been sitting in here. I’ve already kind of shut down a couple of things before the team was using mixed panel before bigquery was set up to look at sales data as well, which was totally off base, and, like the accuracy, was not great. So I shut that down. But yeah, I would say
115 00:20:42.860 ⇒ 00:21:12.270 Robert Tseng: the brand team, or like kind of one side of the marketing house uses mix panel and just uses it for purely as like a Google analytics tool. There is like a Google analytics obviously in this company as well. And we use Google tag manager to kind of set up all the other pixels and all the Martech integration stuff that this team doesn’t touch. I kind of rely on the engineering engineers to do that, because I think tagging and tracking is kind of just like really tedious work that
116 00:21:12.270 ⇒ 00:21:31.822 Robert Tseng: is very, not sexy and gets very under appreciated. So I don’t usually let my team touch that but yeah, so that’s kind of the ex the extent of what we have so far. And then if I just kind of pop into segment, and I’ll do like a quick kind of like, stand through there. So you guys can see where we’re at
117 00:21:32.700 ⇒ 00:21:33.430 Robert Tseng: oops
118 00:21:35.190 ⇒ 00:21:43.100 Amber Lin: Robert. Henry hopped off for a bit for a quick second, but I think everything is recorded, so he will be back soon.
119 00:21:43.100 ⇒ 00:21:47.839 Robert Tseng: Okay, yeah, no worries, I think. I kind of need like a sec to transition, anyway.
120 00:21:48.020 ⇒ 00:21:48.600 Amber Lin: Yeah.
121 00:21:51.990 ⇒ 00:21:58.640 Robert Tseng: Yeah, I know, I know, amber. You’re just kind of getting a crash course on like the full. It’s like a full like data function like less.
122 00:21:59.087 ⇒ 00:22:00.750 Robert Tseng: what? What we, what we did.
123 00:22:01.210 ⇒ 00:22:11.050 Amber Lin: Yes. Do we inform? Say we were just talking about mixed panel of landing pages? Does our work inform those decisions.
124 00:22:12.930 ⇒ 00:22:15.419 Amber Lin: I was just making sure it actually works.
125 00:22:16.270 ⇒ 00:22:23.159 Robert Tseng: Like do we influence like landing page, like what landing pages that they keep or like? How they launch it.
126 00:22:23.160 ⇒ 00:22:32.550 Amber Lin: Yeah, like, are we making any strategic recommendations? Or are we? Are we just making ins ensuring that their data functions are like
127 00:22:32.850 ⇒ 00:22:34.100 Amber Lin: functional.
128 00:22:34.340 ⇒ 00:22:35.610 Robert Tseng: Yeah. So we don’t.
129 00:22:35.610 ⇒ 00:22:36.637 Amber Lin: Plays our role.
130 00:22:36.980 ⇒ 00:22:56.888 Robert Tseng: Yeah, we’re not making recommendations on everything I would say with Brand, I’ve been, yeah, like, I just put this in their hands, and they’re doing whatever they want with it. I would like to. I think conversion rate optimization and basically driving new sales from the website is like the closest way that we can tie our work to revenue.
131 00:22:57.820 ⇒ 00:23:09.029 Robert Tseng: I just wait. I haven’t spent enough time there. And where they’ve needed me more is on like what we call core business intelligence, which is just like around.
132 00:23:09.460 ⇒ 00:23:27.960 Robert Tseng: you know, just making sure that when we, when we’re talking about orders and revenue in various different ways that, like, we’re very clear about it. We can break it out by product. We can break it out by cohort and all those things. And and so I I provide rec from. I provide recommendations on how the how the team should should look at the business from that perspective.
133 00:23:27.960 ⇒ 00:23:28.330 Amber Lin: I see.
134 00:23:29.110 ⇒ 00:23:32.799 Amber Lin: So in that way. That’s kind of like 4 parts. So I understand.
135 00:23:33.000 ⇒ 00:23:39.290 Robert Tseng: Okay, cool. Yeah, it’s just that they probably launch a lot more. They’re launching new products probably every 2 weeks. And so.
136 00:23:39.290 ⇒ 00:23:39.609 Amber Lin: Go home.
137 00:23:39.610 ⇒ 00:23:40.200 Robert Tseng: It’s probably.
138 00:23:40.560 ⇒ 00:23:53.929 Robert Tseng: you know, like the the feedback loop is probably a bit tighter than full parts. But yeah, okay, so we’re gonna jump to segment. And just because I think this is largely where the Cdp kind of like work started, maybe Henry will have more questions here.
139 00:23:54.662 ⇒ 00:24:14.860 Robert Tseng: But yeah, I would say, you know, this is kind of like our usage monthly usage. I’m sure it’s like peanuts compared to what you’re used to seeing in previous companies. But I would say the whole like timing and urgency of this project is just that you know the segment contracts finally up for renewal end of this month. I’ve kind of like laid out a plan to kind of like
140 00:24:15.424 ⇒ 00:24:20.835 Robert Tseng: get us ready to make a decision on whether we’re gonna stay on segment or move off of segment.
141 00:24:21.800 ⇒ 00:24:26.180 Robert Tseng: we’re not gonna sign the same contract, because right now we get billed by Mtus, which
142 00:24:26.509 ⇒ 00:24:54.009 Robert Tseng: it looks like we’re tracking a lot of users, obviously, because within the 1st week of July we’ve already Mac almost met our our quota, and then we get charged overages on top of that but in a health E-com business, you know, over 90% of your users are not identifiable. And so we’re basic. This pricing structure doesn’t make any sense for this business. Because if you look at the actual volume of events or objects that are processed we’re very, very small. So like.
143 00:24:54.060 ⇒ 00:25:08.580 Robert Tseng: And this just tells you that the product is not very complex. Users aren’t doing that much. It’s really just you buy. You buy something. Maybe you log in you, you update your subscription. You’re not really transacting that much on the platform. So there really isn’t that much.
144 00:25:09.002 ⇒ 00:25:19.719 Robert Tseng: There’s not really that much stuff going on in terms of like a user journey perspective. It’s probably much simpler than other Sas companies even like some of the clients that we other clients we work with.
145 00:25:20.170 ⇒ 00:25:32.766 Robert Tseng: So one, we’re trying to obviously shift to a different plan. We want to do event based pricing, not Mtus, and then I think there’s some other limitations with segment.
146 00:25:33.480 ⇒ 00:25:35.660 Robert Tseng: one is that.
147 00:25:36.320 ⇒ 00:25:55.010 Robert Tseng: yeah, like their basic product starts from connectors. And so you can think of amber. You can think of segment similar to polytomic where they’re. They started off as an Etl tool, meaning that they’re they have these like preset connectors that help you to bring in data from different sources. And
148 00:25:55.480 ⇒ 00:26:19.139 Robert Tseng: yeah, I think that’s that was like their main core offering. And this is a requirement to doing business with them that you’re paying for at least that level of service over time. I think the product is kind of like not kept up, and they don’t maintain the the most number of sources. And, yeah, just generally, we. We prefer other options to be our our Etl Atl tool, just because it’s much cheaper elsewhere. At this point.
149 00:26:19.430 ⇒ 00:26:39.210 Robert Tseng: however, with the current setup you can see with Bask there are a bunch of random web hooks that are set up because bask is not, doesn’t integrate with anyone. They don’t have an Api. They only send out web hooks. And so we’ve just used segment to kind of create custom web hooks to be able to bring in the transaction events that we need to do a lot of the core reporting.
150 00:26:39.680 ⇒ 00:26:42.517 Robert Tseng: So between Basque as one source
151 00:26:43.200 ⇒ 00:27:04.049 Robert Tseng: Facebook ads, this could probably be shut off soon, like, I don’t really even so understand why we’re holding this on. But the volume so low I haven’t really touched it. And then customer I/O because that’s the main platform. We use to send emails and SMS, so what we consider customer engagement and then ship out so those are the only 3 sources that are actually coming through segment.
152 00:27:04.160 ⇒ 00:27:06.080 Robert Tseng: Currently.
153 00:27:06.830 ⇒ 00:27:29.020 Robert Tseng: I think, obviously, these are all mission critical, especially bask additional context is that they are moving off of Bast. It was supposed to be next month, but it’s probably the window is extended to. Not, I would say, probably not for another 2 to 3 months, as they build their own custom. Kind of what we call the Emr, or just kind of like a a system to process orders.
154 00:27:29.417 ⇒ 00:27:46.902 Robert Tseng: So eventually, like none of these will be necessary anymore. But it’s gonna that’s that’s like another kind of like, that’s probably a future kind of data like, kind of digital transformation, big big scale project to to work on as well.
155 00:27:48.000 ⇒ 00:27:58.889 Robert Tseng: but yeah, for now we we need to, whatever we decide with this Cdp vendor we have to just make sure that all of these web hooks are still firing. We’re still capturing all this data. So there is kind of like in.
156 00:27:59.160 ⇒ 00:28:26.630 Robert Tseng: you know, it’s it is easier to stay on segment and just try to adjust pricing to try to break. Make it more reasonable rather than kind of like setting all this up again in rudder stack, which is doable. It’s not that hard, but that would just, you know. Take take some more. Take some more time for us. I’ll just pause there. Just kinda give you kind of see both of you have any reactions to on the on like the sources side, like, how we’re using segment. Currently.
157 00:28:33.500 ⇒ 00:28:36.389 Robert Tseng: Okay? Seems like no questions.
158 00:28:36.390 ⇒ 00:28:37.399 Henry Zhao: That makes sense.
159 00:28:37.850 ⇒ 00:28:39.940 Robert Tseng: Cool, alright.
160 00:28:40.620 ⇒ 00:28:41.080 Henry Zhao: Yeah.
161 00:28:41.350 ⇒ 00:29:00.371 Robert Tseng: Okay, great, yeah. Feel free to cut me off whenever Henry. So I know I’m just kinda trying to. I’m just talking all over the place and the destinations. And we’re just for your knowledge. This is like, Okay, data comes in Eden’s or a Cdp or segment is not actually a data warehouse. They don’t store. They do store data, which is kind of
162 00:29:01.110 ⇒ 00:29:12.416 Robert Tseng: like misleading. But I mean, just conceptually, you can think of it as this data comes in, and their purpose is really just to to reroute it to wherever it needs to go.
163 00:29:12.740 ⇒ 00:29:32.010 Robert Tseng: what we’ve introduced is the that all that data 1st goes into bigquery, which is the data warehouse that we have set up for them. And then we’re doing whatever we want to it, doing transformations, we’re able to combine different sources. Do all that very flexibly. That’s where Dbt comes in, and we’re able to to orchestrate all of that
164 00:29:32.325 ⇒ 00:29:39.209 Robert Tseng: and then we let a segment kind of pull back from the warehouse and then push the data wherever it needs to go. And so
165 00:29:39.615 ⇒ 00:29:48.600 Robert Tseng: there’s a few different sources that you know that or destinations that it goes to. Obviously all the social channels dwo is their experimentation platform.
166 00:29:48.630 ⇒ 00:30:12.616 Robert Tseng: So they do run some light experimentation from a design perspective for the for the ui, and that’s where that goes. And then obviously tick tock and you’ll see a few other things. Hubspot doesn’t exist anymore. So I think some of these are shut off. Send us. We’re also that’s also not around. So I would only call out that the main things are, yeah, it gets pushed in the customer. I/O, the mix panel and
167 00:30:14.090 ⇒ 00:30:33.334 Robert Tseng: And yeah, I think those are the. Those are the 2 main 2 2 main things that the segment we really rely on to push push data into everything else is more or less kind of a a wash. And like we’re like, whether we don’t really need it. Like, go high levels. Another tool. It is like they’re lightly being used.
168 00:30:33.940 ⇒ 00:30:51.279 Robert Tseng: and and so that’s why I think that retaining a tool like a segment is helpful, because there are these like random in my opinion, like low value or very experimental ways where we can send data quickly into a tool that marketing is testing
169 00:30:51.563 ⇒ 00:31:18.780 Robert Tseng: but these things are revolving door. I feel like this set of tools changes like every every quarter, every 2 quarters. And so, yeah, just being able to kind of like, have a plug and play way of sharing the data that we are kind of more rigorously maintaining in the data warehouse. I do. I do like the being able to plug plug that just just to use segment for that and not think about it too much. And we’re not syncing that much data, anyway. So it’s it’s quite cost effective.
170 00:31:19.237 ⇒ 00:31:25.320 Robert Tseng: Yeah. So that’s from a from a destination perspective, where where data is going to from segment. Currently.
171 00:31:25.809 ⇒ 00:31:38.580 Robert Tseng: But yeah, I would say, Customer I/O mixed panel, though the 2 most important ones that we both. If you read through my really long notion, Doc, we have like success criteria tied to both of these sources.
172 00:31:40.580 ⇒ 00:31:50.259 Robert Tseng: yeah. And so I’ll kind of double click into that a bit more on the mix panel side. Obviously these are the same transactions that are kind of coming in from mask, and those are just getting pushed to the mix panel
173 00:31:51.192 ⇒ 00:32:18.399 Robert Tseng: combined with some of the web flow data, which is, you know, just really kind of web analytics data. So we can at least, even if we don’t identify all of the all the users coming in. We’re able to kind of just see what they’re doing on the sites. Obviously, look at basic funnel reports where they’re dropping off. If they start questionnaires, if they drop out, etc. And like where you know what what products they’re they’re they’re gravitating towards. If
174 00:32:18.630 ⇒ 00:32:42.479 Robert Tseng: I would say, most of the traffic is still very hyper focused, driven through landing pages. It’s not like a lot of most of our new customers do not come from organic search, or whatever. So but anyway, mixed panel is kind of that environment really for marketer to go and like figure all that out, and I do think it’s not that hard to set up. We generally have all that there. So directionally, I think the insights are pretty solid there.
175 00:32:42.720 ⇒ 00:32:48.720 Robert Tseng: I think it’s just a matter of like thinking through what else we need to do to activate and train people to use it more
176 00:32:49.947 ⇒ 00:33:15.499 Robert Tseng: on the customer. I/O side. It’s a very different directive. It’s less. How do we get? How do we understand new people, new users, and convert them to customers. It’s more. I mean, there is that directive as well, because, you know, people that do share data with us. They get onto an email list. There’s an onboarding campaign and everything. But I would say the bigger push from customer I owe. The directive is increase, you know, repurchase rates, and
177 00:33:15.791 ⇒ 00:33:36.170 Robert Tseng: I would say it’s a pretty sticky product already. I think the repurchase rate is something around 70. I have yet to validate that, but that’s the number that the company is anchored towards. And so it’s like, okay, how do we drive that number from 70 to 80%? And that would be a huge win, we’d be able to ask for a ton, more money, or whatever so like. There’s a lot of things that I think are low hanging fruit that we could do there.
178 00:33:36.981 ⇒ 00:34:00.568 Robert Tseng: But yeah, just the way that customer I/O is set up is kind of messy and I kind of if in in the doc that I shared. There’s all this kind of stuff that’s here. If I were to just summarize like, and I’ve I’ve been spending my time the past couple of days in customer, really trying to understand, like what the problem is. I’ll touch on that in a sec. But yeah, I think there’s just redundance.
179 00:34:01.040 ⇒ 00:34:16.890 Robert Tseng: there’s just redundant things that are being pushed into customer. I/O that are all tied back to how we I, how we define a customer, and I guess what I’m considering a customer profile. There’s 3 different sources of truth for that right now.
180 00:34:17.050 ⇒ 00:34:27.439 Robert Tseng: and that’s where I’ll jump into next. There’s in in segment. There already is kind of like some profiles that are set up. There’s some traits that are just kind of
181 00:34:27.739 ⇒ 00:34:33.083 Robert Tseng: yeah, these are out of the box traces. Segment comes with some very like light
182 00:34:33.620 ⇒ 00:34:45.440 Robert Tseng: computer traits that are not used anywhere. These are from the guy that I inherited this from. We’ve never used it since. Like, no one actually really believes these these traits. And so these are not actually being used anywhere.
183 00:34:46.310 ⇒ 00:35:02.299 Robert Tseng: But all of this data is like what we could say. Every customer we have, we’re able, like segment is able to identify all these different things for us and that does get that that is being piped into bigquery. Currently.
184 00:35:02.480 ⇒ 00:35:19.839 Robert Tseng: however, we haven’t really done much data with that oasis, or did. Our analyst engineer is going in and actually doing something with it. Now. As I kick this off with him this week. But yeah, in in bigquery. We’ve only been using our dim customers model which?
185 00:35:27.560 ⇒ 00:35:54.399 Robert Tseng: yeah. So I know there’s a lot of stuff here. But if you go in here and you just look at this kind of, you know, obviously AI generated here, but just gives you a quick snapshot of lineage of how we have been modeling customers in the big in bigquery. It’s really just been coming from vast events. We can get some of the same identifiers from what Bask already sends us. And then there’s like one field that comes in from like Facebook, somehow, that I’m not fully under understanding, like why we even need to do that
186 00:35:54.890 ⇒ 00:36:06.979 Robert Tseng: but the the point is that our internal dim customers model is a separate model from like what segment puts gives us, and then, if we go into customer. I/O
187 00:36:07.150 ⇒ 00:36:11.410 Robert Tseng: and I kind of tease this out in one of the objectives.
188 00:36:14.350 ⇒ 00:36:16.430 Robert Tseng: I can find it.
189 00:36:16.650 ⇒ 00:36:23.550 Robert Tseng: Yeah. So within customer, I/O, there has. It has its own set of profile properties. They all call them different
190 00:36:23.870 ⇒ 00:36:39.650 Robert Tseng: segment calls and traits customer attributes we call them. You know what dimensions or fields in in bigquery. But the the concept is the same. These are all 3 different ways of looking at, like what an ideal customer data model is, and they’re all disconnected.
191 00:36:39.660 ⇒ 00:36:59.479 Robert Tseng: That’s the biggest problem that we’re trying to solve for so just bring it into a single enriched customer data model. That we end up using to push into all these other tools. So mixed panel needs to be read off of bigquery. And so there’s that this shift of moving the data model out of segment and not going straight into segment from or straight into mixed panel from here.
192 00:37:01.000 ⇒ 00:37:19.340 Robert Tseng: and moving that model into bigquery, having it enriched there and then. We’re also use the same one to push into customer. I/O. So that’s not all gonna happen in the next 2 weeks. We’re just doing a proof of concept here just to show that like, okay, we can actually reconstruct the customer data model in a warehouse native approach.
193 00:37:19.664 ⇒ 00:37:31.020 Robert Tseng: It is able to bring in stuff in segment. We’re able to enrich stuff that we already have in bigquery, and then we can also achieve all the same capabilities that are in customer.
194 00:37:31.050 ⇒ 00:37:39.860 Robert Tseng: The I/O there are a few fields where they hired some other random people to just basically add
195 00:37:40.010 ⇒ 00:37:55.688 Robert Tseng: add stuff to customer directly without going through this whole process. So that’s like a whole. Another black box that I’ve been trying to dig into today. But yeah, so that’s that’s kind of the the main. That’s the main problem that we’re trying to crack with. With this approach.
196 00:37:56.820 ⇒ 00:38:03.640 Robert Tseng: I’ll pause. There, I know, like that’s that’s probably a lot to consume. But just want to make sure that we’re kind of on the same page. There.
197 00:38:04.750 ⇒ 00:38:13.969 Henry Zhao: Yeah, it makes sense. I think the identity stitching and like centralized customer model is very important. I’ve done that a lot, and I see the value of that, and it seems like this is a good use case for it.
198 00:38:14.530 ⇒ 00:38:15.939 Robert Tseng: Okay. Great. Yeah.
199 00:38:15.940 ⇒ 00:38:19.914 Henry Zhao: Also, I think it’d be nice also to understand if they’ve done any Nps surveys
200 00:38:20.740 ⇒ 00:38:25.430 Henry Zhao: to understand how to increase that 70% repurchase rate.
201 00:38:25.910 ⇒ 00:38:28.359 Robert Tseng: Yeah. Yeah. So they don’t really do
202 00:38:28.630 ⇒ 00:38:42.580 Robert Tseng: post no post purchase service currently. Or like, if they do, it’s just like via email, like, I don’t really think it’s formally like, and there hasn’t been hasn’t really been been set up. So I think that’s like a great initiative to even recommend later on.
203 00:38:43.970 ⇒ 00:38:44.800 Robert Tseng: Okay.
204 00:38:45.130 ⇒ 00:39:14.269 Robert Tseng: cool. Last thing, I’ll cover on this just because I don’t really like talking for an hour straight is on the audience decide. And so there’s this idea of like, okay, well, what do you do? Once we have a customer data model. And you have all these like, you have your dream state of like you have your table, of your customer, and you have every single thing you would want to know about them. Like, what do you do with that? Well, like that’s it’s probably too unwieldy. And from a marketer’s perspective they don’t really care about that. They just want to be able to
205 00:39:14.370 ⇒ 00:39:26.350 Robert Tseng: to to create cohorts out of them, which your segment calls audiences. But the idea is the same. You can just think of it, amber and more speaking to you, I guess this may be newer concept to you.
206 00:39:26.650 ⇒ 00:39:44.167 Robert Tseng: yeah, you’re just creating filters of like the customer based off, you know, using kind of like sequel logic pretty much to use the the fields that you’ve defined, your customer data model to figure out like, what’s what are the right. What are the best cuts of users that you can go after to retarget?
207 00:39:44.510 ⇒ 00:39:48.617 Robert Tseng: this is pretty green field. They don’t really do much other than
208 00:39:49.510 ⇒ 00:40:12.759 Robert Tseng: stuff that that’s like, Oh, send an email to a customer that ordered a product 7 days ago, and just ask them how I was doing and like stuff like that. So I would say, this is probably like the most inhibitive kind of piece for us, where we really do need a strong marketing partner to really guide us through, like what is, what are the best audiences that they want to like. Look at and test. And
209 00:40:12.860 ⇒ 00:40:29.189 Robert Tseng: I mean, I have my own point of view. We have recommendations that we can give them, but ultimately, if they’re not adopting them. Which you know, this previous dude like came in and he built out some stuff. But none of these are being used. Then it’s basically a wash, right? And so yeah, I think
210 00:40:29.190 ⇒ 00:40:55.990 Robert Tseng: where he fell short, I think, was that he tried to get the marketing team to move their workflow into segment, to build audiences in here and then to like kind of just like export them into into customer. I/O. But then the marketing team never did that. They just like prefer to do everything in customer I/O. And so that tug of war kind of was there. And this has just been like a dead product that has never been used in the past year. So I think that’s
211 00:40:55.990 ⇒ 00:41:01.640 Robert Tseng: that’s like another hurdle to overcome once again. Not in this one to 2 week window. But that’s kind of like.
212 00:41:01.640 ⇒ 00:41:16.806 Robert Tseng: once we’ve done all this cleanup, we’ve built the centralized customer data model. How do we measure the success of like what we’ve done. It’s really to drive adoption of audiences in a and to kind of push them into using audiences in a better in a better fashion.
213 00:41:17.110 ⇒ 00:41:36.220 Robert Tseng: I I like the rudder stacks approach of like we don’t have to change how they do the workflow. If they’re already working in customer I owe like, let them do that there. And it’s just about kind of just exposing the right data to them. And so I’m I’m totally fine with that approach. But I think that’s that’s probably something that we’ll probably we’ll get to, you know.
214 00:41:36.300 ⇒ 00:41:37.769 Robert Tseng: a couple weeks from now.
215 00:41:39.360 ⇒ 00:41:47.292 Robert Tseng: but yeah. So I think that’s in a nutshell like what this Cdp kind of like optimization project is.
216 00:41:48.200 ⇒ 00:41:54.489 Robert Tseng: yeah, I think I’ve kind of distilled it down to 4 core objectives and amber. This is kind of where?
217 00:41:54.700 ⇒ 00:42:03.770 Robert Tseng: Yeah, from your perspective. Yeah, I haven’t been great about ticketing it all out in like this way, like I do have a project in linear
218 00:42:04.953 ⇒ 00:42:17.270 Robert Tseng: this is called. It is the highest priority project for us. And I do have issues. And like stuff that’s like very tactical that we’re working through. But yeah, I don’t know.
219 00:42:17.270 ⇒ 00:42:36.870 Amber Lin: I went through the project. At 1st I was overwhelmed. I used gpt so, and with that, and this I understand this project now in terms of tickets. I think I’ll have to work with Henry. See what works the best for him, because I can create my tickets, and then it doesn’t make sense for Henry, so I think we’ll we’ll work together to see what’s best
220 00:42:36.870 ⇒ 00:42:47.780 Amber Lin: if he needs tickets at all. Well, you can see the size of tickets as needed. And then we’ll go from there. I think I do understand what we’re trying to achieve.
221 00:42:48.248 ⇒ 00:42:53.410 Amber Lin: In this timeframe just confirm. Is it a 2 week timeframe or 3 week.
222 00:42:54.228 ⇒ 00:43:17.499 Robert Tseng: It’s, I would say, 2 weeks. And we’re kind of already started earlier this week. So like oasis already building the customer data model in the warehouse. You should have that ready by today or tomorrow. So that’d be great. I think that would kinda get us much farther along. Obviously, stuff like this cutting tooling costs. I think that’ll that’s not really like anything actionable. That’s just me
223 00:43:17.920 ⇒ 00:43:24.260 Robert Tseng: kind of knowing how to turn it turn things off and on like I I think I I know the roadmap to get there so I can. I’ll
224 00:43:24.380 ⇒ 00:43:32.290 Robert Tseng: like that’s not really engineering work. And then, yeah, also, like mixed panel stuff. This is, yeah, probably.
225 00:43:32.350 ⇒ 00:43:46.069 Robert Tseng: Still, like, yeah, I I think this is not really engineering work, either. Like, it’s it’s just like Qa, making sure that we have the right views that we have and mixed battle like, how do we debut that like? Okay, this.
226 00:43:46.070 ⇒ 00:44:09.299 Robert Tseng: the state has been cleaned up, and it’s not ready for use like go and run free. And so like kind of thinking about like, how do we roll that out. So I would say, these are not objectives that we’re accomplishing these 2 weeks. But these 2, yeah, we are. We’re not actually gonna drive any actual risk. But yes, definitely, we’re gonna have the customer data model ready. And then also, like, be ready to kind of work with the marketing team
227 00:44:09.300 ⇒ 00:44:22.610 Robert Tseng: to recommend, like what audiences to to build with them. And so there is going to be some like client like stakeholder back and forth there to kind of build out like, what? What? That? What? Those audiences will look like.
228 00:44:23.550 ⇒ 00:44:24.440 Robert Tseng: Yeah.
229 00:44:30.280 ⇒ 00:44:35.242 Robert Tseng: Okay, cool. Any other questions on this?
230 00:44:36.460 ⇒ 00:44:51.070 Robert Tseng: Nope, okay, great. So yeah, I think next steps would just be obviously Henry and Henry both. You’ve both consume this. Henry will get access to everything. Here, take your time. Consume it all. You know you can cut up this document. Do whatever you want with it. I think I’m kinda just
231 00:44:51.426 ⇒ 00:45:01.273 Robert Tseng: I’ll be here, and I I think maybe the handoff will be a bit kind of off. But I’ve already prepped the team. They they know that you’re you’re coming in and and kind of kind of taking the reins.
232 00:45:01.600 ⇒ 00:45:22.269 Robert Tseng: yeah, I kind of just got the party started. And now now you’ve walked in, and you can keep it going, I guess. So. Yeah, I think I’ll I’ll be able to support you on any questions you have, and then, just to give you just a snapshot into other things that are in flight. So that’s 1 work stream we’re doing. The team knows that. Like I’ve
233 00:45:22.540 ⇒ 00:45:23.205 Robert Tseng: I’ve
234 00:45:24.410 ⇒ 00:45:45.239 Robert Tseng: yeah, like, this is the biggest priority thing like, I’m deflecting a lot of other requests. And so obviously with this type of like, fractional data engagement. A lot of it is like from a project management perspective. How do we manage our team’s hours? Because, yeah, if we all just like locked in and just like worked a lot. We sure we get a lot of stuff done. But I think I’m always like looking to.
235 00:45:45.460 ⇒ 00:46:01.039 Robert Tseng: I I under like I have the best sense of like budget profitability, kind of like what makes sense for us to take on how many hours you want to put into things, and so I think it’ll take maybe some time for you to kind of maybe figure out like how to utilize our team best. But
236 00:46:01.562 ⇒ 00:46:19.959 Robert Tseng: point is you, I mean, I think I think you’ll catch on to that pretty quickly. And then amber like I mean, obviously like you, you know where where we stand here and what our path to just bigger margin with this client looks like where we need to scale back and and the efficiency you need to create. So that’s just gonna always be in the background of like
237 00:46:20.490 ⇒ 00:46:45.519 Robert Tseng: that to me, is working on the business and not working in the business like figuring out, how do we run? Run our teams more more smoothly? Yeah. And then there’s just ad hoc stuff that comes up. They’re always launching new products, and so that always challenges or data models. And so things will get fixed there and then. As they use all the reports that we’ve already built out and I’ll give you access to that blow so you can go and look out everything in there.
238 00:46:46.266 ⇒ 00:46:59.443 Robert Tseng: Yeah, we’re we’re kind of asked to make some refinements to to our tableau reports here and there so, and we do maintain like quite a number like oh, probably over 20 at this point. So,
239 00:47:00.150 ⇒ 00:47:14.793 Robert Tseng: yeah, those are the things that are running in the background aside from you. And and I want to, you know, set you up for success, so you don’t have to worry about anything except for the Cdp stuff. And then but you’ll kind of be a fly on the wall kind of absorbing like, what else is going on?
240 00:47:15.220 ⇒ 00:47:16.080 Robert Tseng: so
241 00:47:16.500 ⇒ 00:47:33.235 Robert Tseng: yeah, I think that’s that’s all I wanted to cover from this session. I’ll just leave some open space to just ask any questions about about anything related to this project or not. For both of you. Obviously, this record, this, this video is recorded. And
242 00:47:33.720 ⇒ 00:48:02.279 Robert Tseng: I think I’ll share one cool thing with Henry. This is like internal tooling we’ve built for ourselves. So every Zoom Meeting gets pulled into here. And then you can pretty much just chat with anything. So we have it broken out by client. So if you wanted to go and reference a meeting there. You know, it’s already videos. There, you can see the transcript. We automatically can generate tickets that get can get. You know, just like a template that can get pushed to linear
243 00:48:02.734 ⇒ 00:48:18.439 Robert Tseng: and then for any client cons, we have like some summarization there. So we’re, I think we’re pretty AI forward company. We have a dedicated AI engineering team internally to basically as you’re working through anything like if you think something is slower and efficient, like.
244 00:48:18.470 ⇒ 00:48:46.959 Robert Tseng: we’re pretty loose about it, like I think we maybe follow agile practices. But we don’t. We’re not that religious about it. I think we’re very much about taking shortcuts, trying to figure out how to how to make things faster for ourselves. So yeah, I think you’ll hopefully you get a kick out of like, yeah, you’ll you’ll continue to help us build build better tooling to speed up your workflows. And you know, a lot of this stuff ends up going into sales calls and Utam and I, we just
245 00:48:47.120 ⇒ 00:49:05.559 Robert Tseng: we, you know, it helps because people can see that we we kind of. We dock through and everything that we built for clients. And yeah, anyway, I think this is this is something that you know the team the team uses daily at this point. So yeah, I think that’s
246 00:49:06.380 ⇒ 00:49:10.059 Robert Tseng: yeah. I’ll I’ll just pause there. Any other kind of thoughts or questions, for now.
247 00:49:11.060 ⇒ 00:49:19.709 Henry Zhao: I have a Cdp related question. So have you guys already reached out to rudder, stack, and segment like the salespeople to talk about pricing, and like what that looks like.
248 00:49:19.840 ⇒ 00:49:41.979 Robert Tseng: Yep. So with respect, they understand that we’re we have this like, I have this slack channel with them. I can bring you in there, I mean, if you remember Ryan I’ve heard with them before. I think he was at the event as well. And then Megan is kind of their salesperson. So I kind of let them know, like what we’re doing in the next week or 2. They’re not gonna run pocs for us like exactly. I think I was.
249 00:49:42.060 ⇒ 00:49:58.663 Robert Tseng: I don’t think they have a poster team or anything, but they’re they’re aware that we can ask any questions if we wanted to kind of like test what we’ve built out from the modeling side with rudder, stack, or segment. Once that’s there, they’re they’re on. They’re ready to kind of help us, and pricing wise.
250 00:49:59.120 ⇒ 00:50:14.045 Robert Tseng: I have a ballpark pricing from them. I think they they haven’t really like committed to a number yet which I expecting them to within a week or 2 and then on the segment side, we also have a channel that I’ll add into
251 00:50:14.550 ⇒ 00:50:22.450 Robert Tseng: which I have a call with them, actually. But sag met.
252 00:50:23.650 ⇒ 00:50:32.339 Robert Tseng: Yeah. So we have called them on like, kind of tomorrow, really? And so I’m I’ll I’ll keep kind of having those conversations with them like I have.
253 00:50:33.420 ⇒ 00:50:41.499 Robert Tseng: This was the contract that they originally have. It expires obviously end of the month. And then we’re we’re kind of in the middle of talking about pricing now.
254 00:50:42.480 ⇒ 00:50:46.570 Henry Zhao: Okay. So I’ll try to sign the contract today. So maybe I can even be part of that call if you think it’s beneficial.
255 00:50:46.870 ⇒ 00:50:51.109 Robert Tseng: Okay, cool. Yeah, I think, that’s
256 00:50:52.920 ⇒ 00:50:56.259 Robert Tseng: yeah. I’ll I’ll add you to any to to these.
257 00:51:01.390 ⇒ 00:51:02.120 Robert Tseng: Alright,
258 00:51:03.520 ⇒ 00:51:09.860 Robert Tseng: yeah. Any other. Any other thoughts, questions, things we want to talk about. We don’t have to take the full time as well. But I like to
259 00:51:10.220 ⇒ 00:51:13.082 Robert Tseng: sleep, not not run to the end.
260 00:51:16.470 ⇒ 00:51:17.590 Henry Zhao: I’m good on my end.
261 00:51:17.590 ⇒ 00:51:19.380 Robert Tseng: Okay, how about you? Amber?
262 00:51:20.219 ⇒ 00:51:32.749 Amber Lin: We can stay over a little bit to talk about other parts for this meeting. How you want me to help. We could do another meeting, so that this recording doesn’t get extended too long. That helps.
263 00:51:32.960 ⇒ 00:51:39.970 Robert Tseng: Yeah, that’s okay, I mean, and maybe we could just figure out like, what, what do you feel like you would want to know next, and I’m sure I’ll set up another call.
264 00:51:39.970 ⇒ 00:51:40.450 Robert Tseng: maybe.
265 00:51:40.450 ⇒ 00:51:46.289 Robert Tseng: Yeah, with you do separately, or both of you together again. What? What do you feel like? You would want to cover in the next one.
266 00:51:47.368 ⇒ 00:52:01.369 Amber Lin: For the next one. As you said, you’re still gonna be handling customers facing standups, and probably Demos. I know, planning and grooming are currently internal meetings and retros as well. I could assist with those.
267 00:52:01.370 ⇒ 00:52:01.880 Robert Tseng: Yeah.
268 00:52:02.290 ⇒ 00:52:10.079 Amber Lin: I want to know how specifically I can help you with, say, internal pm, when it’s not client facing.
269 00:52:10.250 ⇒ 00:52:25.560 Amber Lin: does it? Does that mean helping you group the tickets, making sure that they have all the requirements? Or does it mean checking in on people on? Are they completing the tickets on time. Of what does my responsibility cover.
270 00:52:25.840 ⇒ 00:52:45.739 Robert Tseng: Yeah, I would say, definitely, so yeah, I think Mondays and Fridays we do a daily stand up with them Mondays and Fridays. I usually don’t let the Eden team join. Monday is usually a planning call for us. We talk about kind of like from a project perspective. What we’re prioritizing. I try to let the team know like these are our guardrails.
271 00:52:45.860 ⇒ 00:52:56.560 Robert Tseng: Some ad hoc stuff inevitably comes up. I still say 25% of the work that we take on is pretty ad hoc but at least like this helps me to say no to things when the team come, when they come on.
272 00:52:56.560 ⇒ 00:52:59.030 Amber Lin: I see you’re running a 1 week spread.
273 00:52:59.967 ⇒ 00:53:14.400 Robert Tseng: They’re actually 2 weeks. So typically they don’t. But we we do adjust some things because some things don’t take a full 2 weeks to do. Yeah, we’re still operating pretty light, like we’re we don’t like. No, nobody on this full time on this client. Right? Well, we also have quite a few people.
274 00:53:15.390 ⇒ 00:53:40.360 Robert Tseng: yeah. And then on Fridays we do retros which I haven’t been doing very well, which I think I’ve seen you do with urban sense and stuff. So being able to apply something like that to our team, just to make sure that the engineers feel good about how we’re running things. I think the common complaint is always just like I I probably not super organized all the time I come in. And I basically ask them to to do stuff that
275 00:53:40.370 ⇒ 00:53:45.919 Robert Tseng: like, I mean, a lot of it is just they don’t update their tickets. They’re like, there’s a lot of redundant things like.
276 00:53:46.332 ⇒ 00:54:02.950 Robert Tseng: yeah, you can just look at some of the deadlines. They’re not like all. It’s not all clean. So like grooming is kind of just like. When I have extra time like I’ll just I’ll do some of it. But otherwise, like I, I do limit my my time on this client, and so I I don’t think I’m giving them
277 00:54:03.080 ⇒ 00:54:07.540 Robert Tseng: like a great experience. So.
278 00:54:07.540 ⇒ 00:54:20.059 Amber Lin: I I mean for your projects and your projects like whatever you can do with your time. It’s already great. I don’t like you guys shouldn’t really be running projects. I think for me, I can.
279 00:54:20.260 ⇒ 00:54:49.169 Amber Lin: First, st I can help with grooming. Now that I have context, I’m going to go into all the projects you have. Make sure tickets are. Have all the contacts. I’ll probably book a meeting grooming session with the team. Maybe I’ll run through it with you first, st and then check with the team to make sure. At least in the backlog, or things still needed. Can they be cleaned up, and so that when it’s when it comes to planning, it will be a lot easier for you to pull in tickets.
280 00:54:50.190 ⇒ 00:55:08.883 Robert Tseng: Yeah, okay, I think that sounds good. Definitely. Start with the stuff that’s in progress. I don’t need you to go through all the historical stuff, and then backlog and plan we can ignore. For now, too, I think that’s just kind of something we’ll we’ll build together like as you as you think into it, but I think this should be plenty enough for you to to get started with
281 00:55:09.140 ⇒ 00:55:15.970 Amber Lin: Good and honestly for grooming. I can. Sorry for retros. I can run the session, so you don’t have to.
282 00:55:16.070 ⇒ 00:55:25.870 Amber Lin: You don’t have to coordinate it. You can listen in and chime in, but I think if I’ve run it similarly, as I do for time, it’ll be a lot easier for your brain.
283 00:55:26.350 ⇒ 00:55:29.809 Robert Tseng: Okay. Great. Yeah. I think I would appreciate that.
284 00:55:30.370 ⇒ 00:55:31.910 Amber Lin: Yeah, I’ll start with that.
285 00:55:33.010 ⇒ 00:55:44.649 Robert Tseng: Okay, yeah. And then if you decide that like, Hey, we don’t need to meet daily or like, you feel like there are some inefficiencies. Like, I mean, I think that’s also why I want you as a 3, rd you know, like another, another set of eyes here.
286 00:55:45.296 ⇒ 00:55:46.270 Robert Tseng: Okay, okay.
287 00:55:46.270 ⇒ 00:56:00.299 Amber Lin: I guess my question, Henry, is, do you want me to help you? Set up this project in linear? How do you prefer to track things. Keep track of tasks and keep track of deadlines. What do you prefer.
288 00:56:00.500 ⇒ 00:56:04.229 Henry Zhao: I think we can maybe do start out meeting weekly and discuss that.
289 00:56:06.060 ⇒ 00:56:07.680 Amber Lin: Go ahead. Do you want.
290 00:56:07.680 ⇒ 00:56:13.999 Henry Zhao: I decide now, I think I decide usually after looking at the data, looking at how people work together. And then I make a better decision.
291 00:56:14.590 ⇒ 00:56:21.480 Amber Lin: Okay, so should I put stuff? Should I move the project stuff from notion to linear now, or do you? You say you want to wait like.
292 00:56:21.480 ⇒ 00:56:24.623 Henry Zhao: I don’t care either one. I don’t care if it’s 0 notion linear. I don’t care.
293 00:56:24.820 ⇒ 00:56:31.750 Amber Lin: Okay, okay, so I’ll move it. If we we’ll we’ll meet again, maybe next week, and if you don’t like it we’ll just crash. We’ll flash that.
294 00:56:31.750 ⇒ 00:56:34.509 Henry Zhao: I think for me. What’s more important, just the communication is
295 00:56:34.890 ⇒ 00:56:46.080 Henry Zhao: like. So I work with some people that like they won’t do anything unless there’s a ticket for it. I’ve worked with people that you know don’t even look at tickets. So I think we just need to align on what your work, what are each person’s working styles are, and.
296 00:56:46.080 ⇒ 00:56:48.210 Robert Tseng: We have a mix of those on our team. So.
297 00:56:48.421 ⇒ 00:56:50.749 Henry Zhao: Think the world would give you a mix of those right.
298 00:56:51.319 ⇒ 00:56:51.889 Robert Tseng: Okay.
299 00:56:51.890 ⇒ 00:56:56.730 Henry Zhao: Everybody’s person is like. So that’s why I say, I don’t decide that now. Usually it’s after
300 00:56:56.910 ⇒ 00:56:59.230 Henry Zhao: the 1st one how people are. Yeah.
301 00:56:59.230 ⇒ 00:57:16.740 Amber Lin: I see. This is a fundamental basic question. It’s for Robert. So for the Cdp project, the the structure of this engagement is Henry is gonna be the strategist of doing some most important tickets. And there’s still other team members involved in this project, too. Right?
302 00:57:16.950 ⇒ 00:57:31.749 Robert Tseng: Yeah, I would say, engineering work. He can. He can tap our team to do and then I mean, he- he’s just gonna take over like what I was doing on that project which was just like designing the the strategy and the basically the solutions architect or whatever. If you want to call that.
303 00:57:31.750 ⇒ 00:57:32.550 Amber Lin: Okay, I understand.
304 00:57:33.050 ⇒ 00:57:35.559 Amber Lin: There you go. Okay. Sounds good.
305 00:57:36.520 ⇒ 00:57:37.130 Robert Tseng: Okay.
306 00:57:37.130 ⇒ 00:57:38.650 Henry Zhao: Perfect good to meet you.
307 00:57:39.040 ⇒ 00:57:44.119 Robert Tseng: Cool. Alright. Well excited to yeah. Excited for this to kick off.
308 00:57:45.370 ⇒ 00:57:48.660 Robert Tseng: Okay? Well, talk to you later. Y’all later.
309 00:57:49.040 ⇒ 00:57:50.910 Henry Zhao: Thanks, Robert, thanks, amber bye, take care.