Meeting Title: CTM Data Integration Project Check-in Date: 2026-04-29 Meeting participants: Robert Tseng, acromie, Adam Kittleson
WEBVTT
1 00:00:08.460 ⇒ 00:00:09.820 acromie: Hello!
2 00:00:09.820 ⇒ 00:00:10.930 Robert Tseng: Hey, Allison.
3 00:00:11.310 ⇒ 00:00:12.330 acromie: Go ahead!
4 00:00:13.240 ⇒ 00:00:14.359 Robert Tseng: Good, how are you?
5 00:00:14.800 ⇒ 00:00:19.550 acromie: Not too bad! Not too bad. Trying to enjoy a little time outside while the weather’s nice.
6 00:00:19.900 ⇒ 00:00:20.590 Robert Tseng: Nice.
7 00:00:20.880 ⇒ 00:00:24.640 Robert Tseng: Sorry, I will keep my camera off on this one.
8 00:00:24.640 ⇒ 00:00:25.470 acromie: phrase. Slightly.
9 00:00:25.470 ⇒ 00:00:29.799 Robert Tseng: Embarrassing, but, like… the cafe that I was at was, like…
10 00:00:30.090 ⇒ 00:00:32.810 Robert Tseng: Yeah, they had, like, a water pipe burst or something, so…
11 00:00:32.810 ⇒ 00:00:34.160 acromie: Oh no!
12 00:00:34.160 ⇒ 00:00:35.400 Robert Tseng: Yeah.
13 00:00:35.400 ⇒ 00:00:36.110 acromie: Oh… I’m just flying.
14 00:00:36.110 ⇒ 00:00:39.760 Robert Tseng: letting me sit in, like, the stairwell right now, so…
15 00:00:39.760 ⇒ 00:00:41.090 acromie: Oh my gosh.
16 00:00:41.580 ⇒ 00:00:44.760 Robert Tseng: Yeah, kind of… not ideal situation.
17 00:00:44.760 ⇒ 00:00:46.130 acromie: Sprite? Oh my.
18 00:00:46.130 ⇒ 00:00:48.590 Robert Tseng: I would rather not show you my background right now.
19 00:00:48.730 ⇒ 00:00:50.500 acromie: It’s okay.
20 00:00:51.940 ⇒ 00:00:57.920 acromie: Well, hopefully we can keep this quick, but I just saw the email you sent over.
21 00:00:57.920 ⇒ 00:00:58.470 Robert Tseng: Yep.
22 00:00:58.690 ⇒ 00:01:03.109 acromie: And so, I was just starting to digest it, really.
23 00:01:03.110 ⇒ 00:01:06.010 Robert Tseng: Yeah, yeah, I figured, like, it probably didn’t have much time to…
24 00:01:06.010 ⇒ 00:01:06.460 acromie: Yeah.
25 00:01:06.460 ⇒ 00:01:08.310 Robert Tseng: Probably make more sense to walk through it anyway.
26 00:01:08.430 ⇒ 00:01:26.689 Robert Tseng: So, yeah, I mean, I guess, like, you know, a lot of back and forth, so just to kind of frame the situation, I think the best artifact was, like, your call tracking calls, kind of connection flow diagram. I think that generally is… yeah, that makes sense. Like, I think we agree with you that, like.
27 00:01:26.810 ⇒ 00:01:30.649 Robert Tseng: Yeah, we need to send the data to…
28 00:01:30.900 ⇒ 00:01:36.119 Robert Tseng: Or, like, I think we need to send the data first to CTM, enrich it there, and then send it to Amplitude.
29 00:01:36.120 ⇒ 00:01:47.149 acromie: Yeah, that’s my preference, and unfortunately, they haven’t gotten back to me yet this week, so I’m hoping they’re gonna come back and be good with that, and if not, I’ll try to make the case that they need to be. Yeah.
30 00:01:47.320 ⇒ 00:01:57.320 acromie: So, because otherwise, it’s just, how many more ways do we need to route it, right? Or, like, in your email, I noticed the last option was the batch import, you know.
31 00:01:57.320 ⇒ 00:01:57.860 Robert Tseng: Correct.
32 00:01:57.860 ⇒ 00:02:01.950 acromie: means we would probably have to do it that way, in that case, so…
33 00:02:01.950 ⇒ 00:02:07.560 Robert Tseng: Yeah, which is the, like, I tried to order that in preference. I think Allison 1 would be ideal.
34 00:02:07.560 ⇒ 00:02:10.120 acromie: Least ideal, but would work.
35 00:02:10.660 ⇒ 00:02:15.989 Robert Tseng: Yes. So, I mean, I guess, like, I left off a few details there, so if we want to start with the worst-case scenario.
36 00:02:15.990 ⇒ 00:02:32.769 Robert Tseng: Worst case scenario, it’s not going to be real time. It’s just going to be able to pull, kind of, like, batches, right, into your data warehouse, and then we can, you know, from there, it’s pretty straightforward. It’s just, like, kind of doing, scheduling a job from the warehouse into Amplitude. You know, this is actually better for…
37 00:02:32.770 ⇒ 00:02:42.279 Robert Tseng: like, doing backfills, if we need to, like, you know, and it’s just… it’s just slower, and the hard part is really just the identity matching. So, because…
38 00:02:42.670 ⇒ 00:02:47.559 Robert Tseng: Yeah, being able to… it’s gonna be, like, an orphaned, kind of, like.
39 00:02:47.790 ⇒ 00:02:53.739 Robert Tseng: call ID that, like, we may have trouble matching to the amplitude user. I can imagine, like.
40 00:02:53.920 ⇒ 00:03:10.769 Robert Tseng: there’s going to be multiple user IDs. We may… I mean, separate, but related. If we do build a better identity model, and we can… that’s more robust, then we can always… even if the, you know, whatever device or session they’re using, we’re just kind of having this, like, way to map it consistently. Yep, yep.
41 00:03:10.900 ⇒ 00:03:29.409 Robert Tseng: then, like, from the user side, we can keep that under control, but then the caller ID may continue to just be a bunch of, like, random IDs, and that may continue to fan out. So, yeah, so it’s like, we can only control, like, the… what we… the identity side of what we do have, which is not on the call center side.
42 00:03:29.470 ⇒ 00:03:31.779 Robert Tseng: But, you know, assuming that that’s…
43 00:03:32.200 ⇒ 00:03:36.960 Robert Tseng: I mean, yeah, I think this is quite… a little bit abstract at this point. We’re kind of…
44 00:03:36.960 ⇒ 00:03:37.670 acromie: Yeah, no.
45 00:03:37.670 ⇒ 00:03:38.810 Robert Tseng: I noticed that it’s like…
46 00:03:39.000 ⇒ 00:03:53.690 acromie: Totally understood. I think we’ve got this little piece of data we need to get from the call center side, and, you know, once we have that, we’ll be able to more fully, kind of, map everything and understand what variables we have available to us, but…
47 00:03:53.690 ⇒ 00:03:54.250 Robert Tseng: Right.
48 00:03:54.250 ⇒ 00:03:58.579 acromie: We’re getting closer to figuring out what this actually means.
49 00:03:59.100 ⇒ 00:03:59.830 acromie: So…
50 00:04:00.100 ⇒ 00:04:12.310 Robert Tseng: One thing I didn’t mention was, like, okay, if that worst-case scenario happens, we basically have to… we can manage the identity model… I mean, we can help your team identify… manage the identity model from, like, kind of, fanning out too much.
51 00:04:12.320 ⇒ 00:04:23.819 Robert Tseng: even if they are shooting a bunch of random, like, you know, call IDs, what we could do is we could hash the phone number and use that as, like, CTM’s version of, like, a stable user ID.
52 00:04:23.820 ⇒ 00:04:24.430 acromie: Exactly.
53 00:04:24.430 ⇒ 00:04:30.259 Robert Tseng: numbers probably don’t change that much, so that could be better, and I do believe that
54 00:04:30.360 ⇒ 00:04:47.170 Robert Tseng: is… you could, like, your privacy policy should support that. Like, I was, like, checking with a couple, friends I know in the industry, and they’re like, yeah, that seems doable. So, you know, all that to just kind of give you some reassurance that, like, I think if we even got to the worst case.
55 00:04:47.170 ⇒ 00:04:53.590 Robert Tseng: I still think it’d be okay. Yeah. Yeah, it would just be more limited compared to option one, yeah.
56 00:04:54.010 ⇒ 00:04:54.550 acromie: Yep.
57 00:04:55.080 ⇒ 00:04:55.630 Robert Tseng: Yeah.
58 00:04:58.290 ⇒ 00:05:02.200 acromie: Okay, cool. Alright, so I think,
59 00:05:02.670 ⇒ 00:05:08.089 acromie: You know, option one is obviously where I want to be, right?
60 00:05:08.090 ⇒ 00:05:08.460 Robert Tseng: Right.
61 00:05:08.460 ⇒ 00:05:25.310 acromie: the most sense to me. My question for you is this small service in between, and based on my notes I sent over, does it make… can that small service be in Tag Manager on a server-side install?
62 00:05:25.790 ⇒ 00:05:35.710 Robert Tseng: I think we would recommend against that. Okay. Yeah, just like… it’s…
63 00:05:36.480 ⇒ 00:05:52.590 Robert Tseng: using it to route calls, you’re basically relying on it to be, like, your main container for routing calls from CTM to Amplitude. It’s just, like, another failure point that, like, we don’t really have much visibility into, frankly. So,
64 00:05:52.870 ⇒ 00:05:59.070 Robert Tseng: I mean, I guess, Aria, our preference is to just run it on your existing
65 00:05:59.170 ⇒ 00:06:11.579 Robert Tseng: I mean, I think you’re in Azure already, but, like, we, you know, in my… it’s like, whether you’re using Lambda functions, GCP Cloud Functions, whatever it is, you know, I think you already have your data in Azure, they have their version of, like, a cloud function.
66 00:06:11.580 ⇒ 00:06:12.390 acromie: Yeah, yeah.
67 00:06:12.390 ⇒ 00:06:16.680 Robert Tseng: And yeah, like, that should… I mean, that would be more reliable.
68 00:06:16.680 ⇒ 00:06:21.220 acromie: And we would, I mean, essentially, we would want this to be the same
69 00:06:21.620 ⇒ 00:06:24.649 acromie: Oh, wait, no, this is before we get to amplitude, so…
70 00:06:24.650 ⇒ 00:06:26.349 Robert Tseng: Yeah, this is before, this is routing the data.
71 00:06:26.350 ⇒ 00:06:30.829 acromie: with Google, because Google’s automatically going, okay, okay, sorry, we’re getting…
72 00:06:31.220 ⇒ 00:06:49.880 Robert Tseng: Because you’re basically creating two endpoints. One is to take the webhook, transform it into something that Amplitude can accept, and then you’re sending it to Amplitude. So, instead of hosting that in GTM, yeah, probably would… that’d be better to… to do a, like a… like a cloud function. Yep.
73 00:06:50.270 ⇒ 00:06:51.640 acromie: Okay. Okay.
74 00:06:51.910 ⇒ 00:06:52.460 Robert Tseng: Yeah.
75 00:06:55.210 ⇒ 00:06:55.740 acromie: Awesome.
76 00:06:56.470 ⇒ 00:06:56.819 acromie: the other.
77 00:06:56.820 ⇒ 00:07:01.940 Robert Tseng: Yeah, that way it’ll be, like, as close to being real-time as, you know, you can, yeah.
78 00:07:02.190 ⇒ 00:07:03.779 Robert Tseng: Besar, yeah, you were saying?
79 00:07:04.040 ⇒ 00:07:12.130 acromie: Oh, no, I was just saying, then this… then from there, it would get into amplitude and run the… through the same kind of,
80 00:07:12.360 ⇒ 00:07:15.820 acromie: What we have set up now for online scheduling events.
81 00:07:15.820 ⇒ 00:07:16.240 Robert Tseng: Right.
82 00:07:16.240 ⇒ 00:07:27.509 acromie: It would just be another set of events, and actually, we’ll probably reuse some of the events you’ve already set up… Right. …in there for these, because some of them don’t… aren’t even existing anymore, so…
83 00:07:27.790 ⇒ 00:07:28.210 Robert Tseng: Exactly.
84 00:07:28.210 ⇒ 00:07:29.630 acromie: Okay, cool.
85 00:07:32.760 ⇒ 00:07:37.280 acromie: Now, I guess, so maybe where I was thinking is…
86 00:07:37.450 ⇒ 00:07:45.020 acromie: does it make sense for that piece that you guys built out for us to move into a GTM?
87 00:07:46.620 ⇒ 00:07:52.589 Robert Tseng: Yeah, we could move that, if you would prefer to, but since we already set it up,
88 00:07:53.280 ⇒ 00:07:59.400 Robert Tseng: I mean, well, it’s a cloud function currently, we could just… Okay. We could just… Yeah. Yeah.
89 00:07:59.750 ⇒ 00:08:03.890 Robert Tseng: if that’s the only thing we’re moving, I don’t really feel like that’s already… it seems…
90 00:08:04.110 ⇒ 00:08:07.260 Robert Tseng: reliable for now, I… yeah, so… Okay. Yeah.
91 00:08:08.240 ⇒ 00:08:15.029 acromie: Cool. So, but not necessary.
92 00:08:16.660 ⇒ 00:08:24.890 acromie: Sweet. Because, yeah, because I think, what we’re trying to do is figure out how we could also potentially send events over to Meta.
93 00:08:25.320 ⇒ 00:08:26.780 acromie: Eventually.
94 00:08:27.590 ⇒ 00:08:46.120 Robert Tseng: Yeah, so that’s kind of maybe where option B is a little bit more, like… yeah, I mean, that’s another middleware, but, like, that’s what a CDP is there for. Like, they have these connections out of the box, so rather than us having to build multiple custom cloud functions, we could just use, like, a segment or rudder stack, probably.
95 00:08:46.170 ⇒ 00:09:02.209 acromie: Yeah, well, and what irritates me is that Amplitude is a CDP, it’s just they… they… because we have these multiple accounts, which we have in Facebook too, you know, it’s hard to send the data to the right one, which is why we had to create this…
96 00:09:02.570 ⇒ 00:09:06.170 acromie: Outside, you know, solution.
97 00:09:06.400 ⇒ 00:09:08.760 acromie: for these Google conversions.
98 00:09:08.940 ⇒ 00:09:09.470 acromie: And so.
99 00:09:09.470 ⇒ 00:09:10.230 Robert Tseng: Yeah.
100 00:09:10.230 ⇒ 00:09:19.219 acromie: what I’m trying to figure out is how can we use the same infrastructure to do something similar over there through the CAPI?
101 00:09:19.400 ⇒ 00:09:25.140 acromie: thing, or is it gonna require a service like in number 2?
102 00:09:25.880 ⇒ 00:09:27.010 acromie: Yeah.
103 00:09:27.010 ⇒ 00:09:37.410 Robert Tseng: Yeah, I mean, Amplitude’s CDP functionality is quite limited, you know, we kind of explored that the first time, so, I mean, it really is just about trying to…
104 00:09:37.860 ⇒ 00:09:43.070 Robert Tseng: kind of… yeah, I mean, I don’t think it was intended for your use case, like, where multiple.
105 00:09:43.070 ⇒ 00:09:43.600 acromie: Right.
106 00:09:43.600 ⇒ 00:09:46.419 Robert Tseng: and all these different things. So.
107 00:09:46.420 ⇒ 00:09:47.979 acromie: You know? Yeah.
108 00:09:49.160 ⇒ 00:10:01.789 acromie: If we were to ever consider, moving, do you have any recommendations on what would make sense to, like, so that all of this, like, we do have better control over CDP?
109 00:10:02.330 ⇒ 00:10:18.839 Robert Tseng: Yeah, I mean, frankly, I think, I mean, I think Segment and Rudderstack are two… are probably the two leading experts. I mean, I understand there’s HIPAA. Rudderstack would definitely do… it’s… like, I… we implement Rutterstack for other health companies. Segment, I mean, I think it’s kind of case-by-case. You may not be able to adopt Segment.
110 00:10:18.840 ⇒ 00:10:25.940 Robert Tseng: But assuming that you… you clear… you clear that hurdle, Rudder stack kind of needs to be maintained by your… by your technical team. Like, it is, like.
111 00:10:25.940 ⇒ 00:10:26.500 acromie: Yeah.
112 00:10:26.500 ⇒ 00:10:29.140 Robert Tseng: purely, like, no UI kind of thing.
113 00:10:29.150 ⇒ 00:10:31.740 acromie: That’s kind of what we liked about Amplitude, is that…
114 00:10:31.740 ⇒ 00:10:32.140 Robert Tseng: Right.
115 00:10:32.140 ⇒ 00:10:37.279 acromie: We are able to get in there and do more of that reporting and whatnot, so…
116 00:10:37.280 ⇒ 00:10:37.930 Robert Tseng: Yeah.
117 00:10:38.090 ⇒ 00:10:39.170 acromie: Okay, okay.
118 00:10:39.170 ⇒ 00:10:55.559 Robert Tseng: But, yeah, I mean, frankly, I mean, this is like, you know, if I were to rebuild from scratch, like, I don’t even think you need, segment, rubber stacking amplitude anymore. Like, we are… a client we’re working with… I mean, they’re not in a regulated space, so there’s a little more… and they’re more, like, willing to let us
119 00:10:55.560 ⇒ 00:11:00.079 Robert Tseng: kind of try a new, like, a simpler infrastructure. We, like…
120 00:11:00.080 ⇒ 00:11:15.619 Robert Tseng: whether you choose Databricks or Snowflake, they can do this entire thing from the ground up. So, like, we’re just using Snowflake microservices to kind of run the whole thing. Like, they do everything from kind of event… event tracking to date routing, like, yeah, so, like, we didn’t…
121 00:11:15.620 ⇒ 00:11:27.459 Robert Tseng: that allowed us to set it up without having to use any of this middleware. Although, you know, I wouldn’t necessarily recommend that for you. It’s… yeah, I’m just saying that I do think that
122 00:11:27.530 ⇒ 00:11:46.890 Robert Tseng: because I understand that a lot of these different tools, like, are overlapping more and more, like, the consolidation, like, makes sense, like, intuitively, that you would want to move towards that, but, you know, it’s kind of hard to, like, piecemeal it unless you just kind of rebuild it from scratch, which, frankly, oftentimes is faster.
123 00:11:47.320 ⇒ 00:11:49.520 acromie: Right, right.
124 00:11:49.690 ⇒ 00:12:01.530 acromie: Oh, what we could have known back then. Okay, so, what about, do you have a sense of pricing on these options?
125 00:12:02.360 ⇒ 00:12:09.020 Robert Tseng: Yeah, I think I’ll probably revisit it. I’m pretty sure, like, we were intending to do option one, so, like, I’m pretty…
126 00:12:09.020 ⇒ 00:12:09.389 acromie: kind of…
127 00:12:09.500 ⇒ 00:12:14.959 Robert Tseng: yeah, I pretty, like, I think that’s what we priced for. So,
128 00:12:15.110 ⇒ 00:12:18.369 Robert Tseng: yeah, I guess, like, I can go back to, kind of.
129 00:12:19.510 ⇒ 00:12:23.590 Robert Tseng: confirm it, but, like, I’m pretty sure that’s what we… to do.
130 00:12:23.800 ⇒ 00:12:29.370 Robert Tseng: Okay. Yeah. Yeah, in the sense that we were expecting to build a custom pipeline. So…
131 00:12:30.010 ⇒ 00:12:38.160 Robert Tseng: Yeah, and then… I mean, as far as, like, kind of, you know, we jumped back in for a few hours, fixed something, like, we kind of, like, caught…
132 00:12:38.310 ⇒ 00:12:43.429 Robert Tseng: We remember where we left off on. It sounds like there are still some…
133 00:12:43.590 ⇒ 00:12:53.899 Robert Tseng: architecture decisions that your team is making that may change, like, the course of what we’re doing, so I think that makes me feel a little bit, like, hmm, like, maybe it’s better to just do, like, a
134 00:12:54.590 ⇒ 00:12:59.489 Robert Tseng: Like, a minimum retainer kind of thing, and then we just kind of…
135 00:12:59.960 ⇒ 00:13:12.459 Robert Tseng: like, I don’t know when these decisions are made, you could pivot, like, it just gives us a little bit more flexibility, rather than locking in, like, in the contract, okay, this is… we’re going to option one, and then it ends up just, like, you know,
136 00:13:12.660 ⇒ 00:13:25.789 Robert Tseng: the… your CTM comes back, and they’re like, no, and we have to pivot again, and I… you know, I guess I wonder with, kind of, the changes, if it even makes sense to structure the way that I did last time.
137 00:13:26.410 ⇒ 00:13:31.799 acromie: Yeah, right, because we’ve gone through a few different options in terms of the structure.
138 00:13:31.800 ⇒ 00:13:32.410 Robert Tseng: Yeah.
139 00:13:32.580 ⇒ 00:13:39.569 acromie: Because, like, I mean, the way I’m currently looking at this is, like, this is… this is the project, right? Like, this is the thing.
140 00:13:39.570 ⇒ 00:13:39.960 Robert Tseng: Yeah.
141 00:13:39.960 ⇒ 00:13:59.119 acromie: to get done for a very specific need. And then, what I was hoping to move into was more of this retainer where we’re, you know, working together to continue enhancing and building and, you know, doing the things we want to do with this outside of just these conversion
142 00:13:59.120 ⇒ 00:14:08.599 acromie: events. And so, I don’t know if that… Makes sense still, or…
143 00:14:09.250 ⇒ 00:14:20.689 Robert Tseng: Yeah, like, I agree. I think the way we were thinking was, like, okay, we do it more kind of focused, scope-based, like, kind of outcome-based for just, like, getting the visibility that you want up again.
144 00:14:20.750 ⇒ 00:14:33.640 Robert Tseng: And then, like, you know, we could transition to a, like, a retainer when we’re, you know, assuming that all the data is trustworthy again, and we can, like, actually start, like, layering things on top of that, so… Right.
145 00:14:33.640 ⇒ 00:14:46.539 acromie: Yeah, and that’s the piece where I don’t have a sense of timing yet, but, like, this scope here, like, I need to get the other side of things, and then I need to go get approval, because I need this stuff connected now.
146 00:14:46.940 ⇒ 00:14:53.320 acromie: Yeah. You know, 6 months ago. But things keep, you know, getting…
147 00:14:54.210 ⇒ 00:15:05.650 acromie: ruffled back up. So, this is more of, like, a, if I get the okay on this, because now there’s going to be the other side of development to send the data to CTM,
148 00:15:05.650 ⇒ 00:15:26.319 acromie: So obviously now we’re looking at, you know, more scope in total, so I need to get that approval. If I, for some reason, I don’t get that approval, then that probably would mean we’d be looking at option 3, right? Because that data’s already coming down to the data warehouse. Well, now our only option is to, you know, do that batch.
149 00:15:26.450 ⇒ 00:15:27.700 acromie: Yeah.
150 00:15:27.730 ⇒ 00:15:47.139 acromie: And so, that’s… that’s what would happen in that case. But hopefully I can go in and be like, hey, here’s what we’re gonna get, here’s what it’s going to, you know, affect on the side of, you know, advertising and optimization and all of these things. Get the approval, and then come to you and saying, okay, when can we start?
151 00:15:47.250 ⇒ 00:15:51.579 acromie: So that we can be, like, on top of it.
152 00:15:51.580 ⇒ 00:15:52.090 Robert Tseng: Yeah.
153 00:15:56.240 ⇒ 00:15:56.795 acromie: Yeah…
154 00:15:57.810 ⇒ 00:15:59.980 Robert Tseng: Sorry, just taking some notes. Yeah, no worries.
155 00:16:02.170 ⇒ 00:16:06.010 Robert Tseng: Yeah, so…
156 00:16:08.320 ⇒ 00:16:12.779 Robert Tseng: I guess, since we’re not sure, I mean, you’re expecting to hear back from them sometime soon.
157 00:16:12.780 ⇒ 00:16:19.269 acromie: You were supposed to send me stuff by Monday at the latest, so I sent him a note this morning and was like, hey!
158 00:16:19.500 ⇒ 00:16:28.450 acromie: Like, what’s going on? And I haven’t heard anything yet, so… Yeah. Hopefully… oh, oh, I did get a somewhat response.
159 00:16:29.310 ⇒ 00:16:33.740 acromie: Oh, gotcha. They’re all in Vegas right now.
160 00:16:38.610 ⇒ 00:16:40.039 Robert Tseng: Yeah, well, I guess, like…
161 00:16:40.040 ⇒ 00:16:42.040 acromie: They’re still looking into things, so he’s.
162 00:16:42.040 ⇒ 00:16:42.410 Robert Tseng: Yeah.
163 00:16:42.410 ⇒ 00:16:45.440 acromie: Get back to me later this week with a final option, so…
164 00:16:45.440 ⇒ 00:16:46.010 Robert Tseng: Okay.
165 00:16:46.650 ⇒ 00:16:52.689 Robert Tseng: I, I think, so just kind of tying it to something, can I kind of just talk a little bit about the pricing?
166 00:16:52.690 ⇒ 00:16:53.420 acromie: Yeah, go ahead.
167 00:16:53.420 ⇒ 00:17:01.879 Robert Tseng: Okay, so, like, I’m just looking back through our, our emails now. So, the original proposal was, like, what, 14, 14.4 up front.
168 00:17:02.300 ⇒ 00:17:14.599 Robert Tseng: Then we’re, like, kind of making adjustments based on, like, okay, we actually need this, like, additional complexity. Like, you know, that’s kind of how we do the fixed pricing, or we could just go just, like.
169 00:17:15.010 ⇒ 00:17:18.289 Robert Tseng: Retainer, like, just… we just go hourly, and it’s like.
170 00:17:18.560 ⇒ 00:17:23.810 Robert Tseng: Our hourly rates up, like, starting at 15 a month, and we just kind of…
171 00:17:24.020 ⇒ 00:17:34.840 Robert Tseng: do that month to month until, like, we… until we… until we… until we finish this. Like, I… that feels like it’s the easier thing to get over the line. I don’t want to be, like, adjusting, like.
172 00:17:34.930 ⇒ 00:17:51.089 Robert Tseng: I mean, to keep tweaking the… the fixed price. Like, that was meant to, like, give you something predictable. Right. And yeah, like, ideally, it kind of caps… caps, like, what the cost would be. I mean, there’s a little bit more variability on the… on the… if we… if we go hourly, because then we’re not really
173 00:17:51.810 ⇒ 00:17:57.110 Robert Tseng: scope it down as much, but, like, I don’t know, I think that… that seems like that would be better for now.
174 00:17:57.280 ⇒ 00:18:08.480 acromie: It could be. My only concern is that I’m going to get a budget for this, and it’s going to be, you know… so I think what we would need to do is look at it from a…
175 00:18:08.480 ⇒ 00:18:22.219 acromie: Like, maybe it’s a range. Or, you know, or like, what do you… where do you think this is gonna come in if we’re doing it hourly, so that I can get the top of that range? You know, just make that expectation that if we’re gonna do it that way.
176 00:18:22.220 ⇒ 00:18:27.290 acromie: That, you know, there’s gonna be some variability in there.
177 00:18:27.460 ⇒ 00:18:31.330 acromie: And so that I’m not, like, all of a sudden going back and be like.
178 00:18:31.690 ⇒ 00:18:36.329 acromie: Yeah. We actually need another couple grand, you know, like…
179 00:18:36.330 ⇒ 00:18:36.720 Robert Tseng: Yeah.
180 00:18:36.720 ⇒ 00:18:38.959 acromie: I want to avoid that if we can.
181 00:18:39.320 ⇒ 00:18:59.239 acromie: even if we know… I mean, I’m a project manager, I totally understand that, you know, time and materials is time and materials, but if we can get really clear on where we think that might land, even, again, even if it’s a range, that’ll help me make the case for it and put us in a good position for any future approvals needed.
182 00:18:59.630 ⇒ 00:19:07.549 Robert Tseng: Yeah. Okay, yeah, well, I mean, like, the… I think it should be pretty… pretty predict… pretty predictable. I mean, the… the… the,
183 00:19:07.930 ⇒ 00:19:19.760 Robert Tseng: like, the way that we do it, when we build by hour… I mean, internally, we run weekly sprints, so, like, everything is going to be scoped week to week, and so, like, even when… assuming… even if you did that, we would tell you, like.
184 00:19:19.900 ⇒ 00:19:27.860 Robert Tseng: Yeah, we would… we would at least give you, like, some lead time two weeks in advance, like, where, where we, where, like, what, what we’re… what’s coming up, and, you know, and
185 00:19:28.050 ⇒ 00:19:36.489 Robert Tseng: Yeah, so, like, that’s how I… that’s how we operate internally, and I think that means that, you know, whether it’s…
186 00:19:36.640 ⇒ 00:19:37.730 Robert Tseng: like…
187 00:19:37.950 ⇒ 00:19:53.520 Robert Tseng: you know, if you… if it’s, like, 15… if it’s 15 grand a month at, like, $150 an hour, I think we could, you know, then, like, that’s… that’s how many hours you can get. I mean, I think, like, we’ll have… I’ll have to check with, like, I mean, I have… we have other…
188 00:19:55.340 ⇒ 00:20:01.729 Robert Tseng: like, there’s some people on the bench right now, so I have to go and, like, kind of confirm, like, what we can do, obviously, because things have kind of…
189 00:20:01.750 ⇒ 00:20:16.829 Robert Tseng: shaken up. We’re expecting a couple other clients to start next month, so, like, I am trying to, like, swap this in, but, like, that could end up… that ends up impacting, like, what we… what my final number will be, but, but yeah, I think that’s… that’s kind of how I would handle it.
190 00:20:17.210 ⇒ 00:20:27.349 acromie: Okay, yeah, so yeah, as long as we can be really clear on what we think that’s gonna sit, I mean, obviously, this project originally was put forward, I think, what would we end up? Like, 18?
191 00:20:27.350 ⇒ 00:20:27.930 Robert Tseng: Yeah.
192 00:20:28.140 ⇒ 00:20:34.819 acromie: So, like, what I would need to know is, like, is that, you know, is there a…
193 00:20:35.070 ⇒ 00:20:39.860 acromie: plus or minus on that, right? Based on, you know, an hourly service.
194 00:20:39.860 ⇒ 00:20:40.340 Robert Tseng: Yeah.
195 00:20:40.340 ⇒ 00:20:49.029 acromie: that’s gonna, you know, fall in a one month, or maybe across two, so that I’m… yeah.
196 00:20:49.740 ⇒ 00:20:52.170 Robert Tseng: Okay, sure, I can, I can,
197 00:20:52.650 ⇒ 00:20:54.119 Robert Tseng: I can give you that option.
198 00:20:54.120 ⇒ 00:20:54.620 acromie: Cool.
199 00:20:55.840 ⇒ 00:20:56.580 acromie: Cool.
200 00:20:58.070 ⇒ 00:20:59.330 acromie: Awesome.
201 00:21:02.070 ⇒ 00:21:06.930 Robert Tseng: Are there any other, kind of, like… I know you guys are really insistent on trying to do the…
202 00:21:07.140 ⇒ 00:21:10.650 Robert Tseng: server-side GTM, I mean, like, we can…
203 00:21:10.650 ⇒ 00:21:20.400 acromie: It’s… it’s not so much that I’m insistent on it, it’s just that I had someone bring up it again in terms of how we could potentially work with Facebook in a compliant way.
204 00:21:20.400 ⇒ 00:21:32.029 acromie: Yeah. And so that’s really the goal of how do we… how do we solve that in the right way? And I, you know, I thought, oh, well, you know, with seat… with call tracking metrics, we have to set
205 00:21:32.550 ⇒ 00:21:41.149 acromie: tracking codes for each account on the website anyway, so we’re probably gonna be using GTM to do that. Yeah.
206 00:21:42.920 ⇒ 00:21:59.250 acromie: And… so if we’re already in there, does it make sense to bring anything else in there at the same time, is kind of where I was going. Doesn’t mean it has to be, it was more of a, how do we utilize this in the way so that, like, we can…
207 00:21:59.580 ⇒ 00:22:06.289 acromie: Have some control, but also, like, just keep things consolidated as much as possible.
208 00:22:06.470 ⇒ 00:22:07.800 acromie: So…
209 00:22:08.770 ⇒ 00:22:09.160 Robert Tseng: Yeah.
210 00:22:09.160 ⇒ 00:22:09.760 acromie: Yeah.
211 00:22:10.080 ⇒ 00:22:13.609 Robert Tseng: I mean, I hear you, like, that makes… intuitively that, like.
212 00:22:13.870 ⇒ 00:22:21.640 Robert Tseng: forwarding events to different ad platforms, like, I think that totally makes sense. Like, I… I think we… it’s… it’s…
213 00:22:21.640 ⇒ 00:22:37.120 Robert Tseng: it’s not that it can’t be done, I think it’s just the reliability that you want may not be there, and so… I mean, typically when we’re building things, if we had all the time in the world, we would always build a redundant solution. Like, we will do… we will do, like, both… we would test both things, and it’s like…
214 00:22:37.960 ⇒ 00:22:55.210 Robert Tseng: obviously, like, one is easier for your team to also kind of jump into than the other, but we always have a fail-safe. So, like, I think it’s just that when we try to, like, move at a certain speed, or we’re trying to enter budget constraints, like, we default to picking the one that we know is more reliable, rather than, like, saying, like.
215 00:22:55.210 ⇒ 00:23:02.920 Robert Tseng: we’ll… we’ll try both, which… but, like, if that’s… I’m just saying, if that’s what you… if that’s what you want to try, like, we… we can. I just, like, I…
216 00:23:02.920 ⇒ 00:23:07.179 Robert Tseng: I think that those are… those are some of the bets that, like, I feel like, you know, generally we don’t…
217 00:23:07.180 ⇒ 00:23:14.280 Robert Tseng: have the luxury of taking being an external vendor, because, like, you know, our clients expect us to deploy the thing that works, and if we fail.
218 00:23:14.280 ⇒ 00:23:14.800 acromie: Oh, yeah.
219 00:23:14.800 ⇒ 00:23:16.220 Robert Tseng: You know, that’s, that’s.
220 00:23:16.220 ⇒ 00:23:17.620 acromie: Totally understandable.
221 00:23:17.620 ⇒ 00:23:23.749 Robert Tseng: But yeah, I mean, I do think that, you know, we could try, if that’s really what you want, yeah.
222 00:23:23.750 ⇒ 00:23:36.110 acromie: Let’s, let’s, let’s continue going down the path we’re going down, again, with the reliability, and this might be something that we, you know, try to work into next year’s budget as a, hey, how can we…
223 00:23:36.260 ⇒ 00:23:43.280 acromie: you know, Get this a little more… Contained.
224 00:23:43.280 ⇒ 00:23:43.940 Robert Tseng: Yeah.
225 00:23:43.940 ⇒ 00:23:53.130 acromie: you know, whatever. But yeah, no, I mean, that’s where that was. It’s not a requirement at all, it’s just… it’s me trying to figure out how to solve
226 00:23:53.350 ⇒ 00:23:58.179 acromie: You know, 5 problems at the same time, and where are the redundancies?
227 00:23:58.550 ⇒ 00:24:15.299 Robert Tseng: Yeah. So, yeah, I mean, we will consolidate wherever we can. I definitely, like, don’t like to over-engineer things. You know, I, you know, so, like, I… you know, we will… I think we will know better once we’re actually running it. I think, just based on what we know, like, that’s… that’s the recommendation. So, yeah.
228 00:24:15.840 ⇒ 00:24:16.490 acromie: Cool.
229 00:24:16.750 ⇒ 00:24:17.290 Robert Tseng: Yeah.
230 00:24:17.650 ⇒ 00:24:18.350 acromie: Cool, cool.
231 00:24:19.480 ⇒ 00:24:34.570 Robert Tseng: Okay, cool, so I’ll send over, yeah, just some kind of updated options, I guess, and then, yeah, I got some good notes from this call, so, you know, we’re gonna aim for option one, in terms of, like, the path forward, but, you know, knowing that, hey, like, maybe, like.
232 00:24:34.580 ⇒ 00:24:41.960 Robert Tseng: icing it differently might be better, because it still needs to be more… be more flexible, and this may… we might not actually be able to do option one.
233 00:24:42.170 ⇒ 00:25:00.069 acromie: Yeah, right. So, I mean, my thought is, like, we’re either gonna do option one, or we’re gonna be relegated down to do option three. Like, that’s what I see happening here. And so knowing, like, what the difference is between those, what we’re looking at, and then,
234 00:25:00.260 ⇒ 00:25:13.360 acromie: Yeah, no, and then I… as soon as I hear back from this other group, I will bring you into the loop on that as well. You know, hopefully it doesn’t change anything, it’ll probably just make our decision on which way we go. So…
235 00:25:14.310 ⇒ 00:25:15.810 Robert Tseng: Okay, sounds good.
236 00:25:16.060 ⇒ 00:25:16.850 acromie: Okay.
237 00:25:18.110 ⇒ 00:25:19.680 Robert Tseng: Alright, I will get you a response.
238 00:25:20.400 ⇒ 00:25:28.299 Robert Tseng: Yeah. Yeah, I’m, yeah. I’ll be here for a little bit, I have another call after this.
239 00:25:28.300 ⇒ 00:25:32.879 acromie: Oh my gosh. Alright, well, good luck, and yeah, look forward to that email.
240 00:25:33.030 ⇒ 00:25:34.290 Robert Tseng: Okay, thanks. Thanks, Bob.
241 00:25:34.290 ⇒ 00:25:36.190 acromie: Thanks so much. Alright, bye.