Meeting Title: Ellie <> Brainforge Check-In Date: 2025-09-16 Meeting participants: Shreya Chowdhury, acromie, Robert Tseng, Zoran Selinger, Adam Kittleson
WEBVTT
1 00:00:26.040 ⇒ 00:00:27.510 acromie: Hello!
2 00:00:27.510 ⇒ 00:00:29.079 Shreya Chowdhury: Hey, how’s it going?
3 00:00:29.280 ⇒ 00:00:30.610 acromie: Good, how are you?
4 00:00:31.190 ⇒ 00:00:32.009 Shreya Chowdhury: Good.
5 00:00:32.460 ⇒ 00:00:33.530 acromie: Good, good.
6 00:00:34.300 ⇒ 00:00:38.979 acromie: I know I’m ready for lunch, hopefully we’re not too long.
7 00:00:41.630 ⇒ 00:00:44.180 acromie: How’s everybody else doing?
8 00:00:44.650 ⇒ 00:00:45.600 Robert Tseng: Doing well.
9 00:00:45.840 ⇒ 00:00:46.820 acromie: Good.
10 00:00:48.100 ⇒ 00:00:50.510 acromie: Good, good. All right.
11 00:00:52.200 ⇒ 00:00:54.000 acromie: Well, where do we want to start?
12 00:00:55.490 ⇒ 00:00:57.310 Robert Tseng: We’re gonna wait for Adam, or we’ll just get started.
13 00:00:57.310 ⇒ 00:01:04.760 acromie: Oh, is Adam… Adam is just back after a long vacation, so…
14 00:01:04.769 ⇒ 00:01:05.379 Robert Tseng: Okay.
15 00:01:05.379 ⇒ 00:01:09.839 acromie: Me? I don’t care if he even knows.
16 00:01:15.549 ⇒ 00:01:23.709 acromie: Come on, OP is in there. Let me… I’ll just ping him, and I would say, let’s get going.
17 00:01:24.290 ⇒ 00:01:24.910 Robert Tseng: Okay.
18 00:01:25.470 ⇒ 00:01:29.910 Robert Tseng: Alright, yeah, I can get… I can kick us off. So…
19 00:01:30.010 ⇒ 00:01:36.099 Robert Tseng: I know there’s a couple outstanding things that we’ll get through, but I think just to kind of frame it in the greater context of, like.
20 00:01:36.500 ⇒ 00:01:42.929 Robert Tseng: what we’ve done, where we’re headed, and then… so I’ll just kind of use the deck as, like, the, like, some of the guidepost here.
21 00:01:43.560 ⇒ 00:01:45.479 Robert Tseng: Oh, there he is. Hey, hey, Adam.
22 00:01:46.680 ⇒ 00:01:49.789 Robert Tseng: Welcome back from your trip with us.
23 00:01:50.080 ⇒ 00:01:51.190 Adam Kittleson: Thank you, thank you.
24 00:01:51.860 ⇒ 00:01:52.520 Robert Tseng: Yeah.
25 00:01:52.700 ⇒ 00:01:59.550 Robert Tseng: Let’s see… hopefully this will… Oh, Lori.
26 00:02:04.110 ⇒ 00:02:09.890 Robert Tseng: Cool. So, I have here… I have a widescreen, so I just kind of have the deck here, and then this is just, like, our Slack channel, so…
27 00:02:10.360 ⇒ 00:02:17.389 Robert Tseng: Yeah, so… I guess at a high level, kind of what we’re… kind of where we’re at now,
28 00:02:17.680 ⇒ 00:02:30.190 Robert Tseng: So, I’m kind of just referencing the old… I kind of took pieces from that old Notion doc that we first started with, where we asked you a bunch of questions, and then, Allison, you kind of just went in and just, like, responded to a bunch of things, then that kind of kicked us off.
29 00:02:30.190 ⇒ 00:02:39.000 Robert Tseng: So I’ve just kind of distilled kind of our work into a couple objectives that, are probably just continuation of things that we’ve already done.
30 00:02:39.090 ⇒ 00:02:54.209 Robert Tseng: So, we’re still kind of tackling that longer-term attribution setup, you know, and I think, you know, we’ve dug into the short-term solution, which we’ve kind of… we’ve stood up, there may be some issues with the rollout, and so maybe the custom solution that
31 00:02:54.210 ⇒ 00:03:03.639 Robert Tseng: Zoran has kind of proposed, again, a direction we’re going to head in, but we’ll kind of hopefully get to that decision, or kind of how we want to approach that later on, later in this meeting.
32 00:03:03.640 ⇒ 00:03:18.819 Robert Tseng: And then, from, yeah, just, like, the work that Shrey had already kicked off, so just, like, kind of getting a handle of, like, what’s going on in amplitude, things that we need to track, getting a sense of the different milestones and the patient journey so that we have, just clearer funnel reporting.
33 00:03:18.820 ⇒ 00:03:30.190 Robert Tseng: Especially in that kind of initial stage before, a patient ends up, kind of setting their first appointment. I think there’s more to do that, because, you know, the life cycle doesn’t end there, it’s just the beginning.
34 00:03:30.190 ⇒ 00:03:39.659 Robert Tseng: So, being able to kind of complete that life cycle, get through that initial booking, you know, and then any other subsequent patient behavior that we haven’t really dug too much into.
35 00:03:39.660 ⇒ 00:03:52.419 Robert Tseng: You know, like, how many points or… appointments are they… are they… are they kind of, going through? What’s the… what’s the break-even point after, you know, a patient, you know, with the budget that we’re using to acquire customer… acquire patients?
36 00:03:52.420 ⇒ 00:04:04.179 Robert Tseng: at what point do they break even? Like, how many appointments do they need to get through? Just any kind of those, things that we haven’t really kind of dived too much into, I think that would help us to kind of round out
37 00:04:04.180 ⇒ 00:04:17.079 Robert Tseng: this, this patient, life cycle intelligence initiative, and I think there’s, there’s, there’s some conclusions, there are a couple deliverables that I would kind of pitch here that, you know, we could… we could drive towards there.
38 00:04:17.560 ⇒ 00:04:18.110 acromie: Great.
39 00:04:18.600 ⇒ 00:04:27.139 Robert Tseng: But yeah, so, you know, I think the pain points are pretty clear, so I’m not going to run through this again. I think we’ll just kind of park here for a second, because…
40 00:04:27.140 ⇒ 00:04:39.549 Robert Tseng: I think as we’re kind of talking through, yeah, these are kind of the initiatives that, you know, Zoran and I kind of talked through. Zoran, this might look a little different from kind of the format that you sent to me, but I just kind of fleshed it out a bit more.
41 00:04:39.630 ⇒ 00:04:46.869 Robert Tseng: And I tried to put some… some, like, kind of timelines here on, like, what we expect, like, kind of the work… to… how long the work to take.
42 00:04:47.250 ⇒ 00:04:47.640 acromie: and kind.
43 00:04:47.640 ⇒ 00:04:51.509 Robert Tseng: Kind of, like, the specific results that we’re hoping to achieve with each of these initiatives.
44 00:04:52.010 ⇒ 00:04:56.859 Robert Tseng: So, with, like, this account setup stuff, you know, we’ve…
45 00:04:56.870 ⇒ 00:05:09.339 Robert Tseng: we’ve already kind of evaluated the risk of the kind of short-term solution. Obviously, we saw that there’s kind of some issue here. Maybe I’ll just kind of pause here, Zoran, if you want to kind of… I know you reconnected it to kind of see, like, how
46 00:05:09.340 ⇒ 00:05:23.239 Robert Tseng: how, you know, if… how long these would… these would stay… stay alive before they get… they got revoked again? Kind of where… where are we with this? Like, is this really just kind of a dead end, and do we really need to kind of explore one of those other solutions?
47 00:05:24.960 ⇒ 00:05:37.880 Zoran Selinger: Yeah, so… you might remember that I’ve said this before, that before we even connected the first one, that basically we are going to attempt to send every single Google Click ID.
48 00:05:37.880 ⇒ 00:05:38.240 acromie: Yeah.
49 00:05:38.240 ⇒ 00:05:39.410 Zoran Selinger: every connector.
50 00:05:39.470 ⇒ 00:05:49.990 Zoran Selinger: And it should… it would be great to have some… some way to filter. I understand that that’s really not easy, so…
51 00:05:50.030 ⇒ 00:06:06.909 Zoran Selinger: there… I mean, with this native, native connector, it’s… it’s not easy, but if we use the API directly, there is a click ID report, where we can grab, we can know, okay, this… this… this…
52 00:06:07.150 ⇒ 00:06:20.539 Zoran Selinger: this click is actually from this particular account, so using API directly, it’s an easy thing to do. Here, I was hoping this wouldn’t happen,
53 00:06:20.850 ⇒ 00:06:26.239 Zoran Selinger: This kind of behavior happens with a lot of different tools, like Zapier as well, if you’re…
54 00:06:26.820 ⇒ 00:06:42.239 Zoran Selinger: your Zaps fail pretty regularly, they will disable them. This is the case here. Today, we only got one disabled automatically, even though I reconnected 3 of them yesterday.
55 00:06:42.350 ⇒ 00:06:48.340 Zoran Selinger: So we, we, yesterday, it was 3 of, or, of 4.
56 00:06:48.470 ⇒ 00:06:54.809 Zoran Selinger: And today, only another one got, I immediately enabled it again.
57 00:06:55.480 ⇒ 00:06:56.709 Zoran Selinger: I mean…
58 00:06:57.290 ⇒ 00:06:59.550 acromie: Yeah, I mean, it’s obviously…
59 00:06:59.550 ⇒ 00:07:11.630 Zoran Selinger: We’re missing a lot when we, you know, we can go in and manually enable them, it’s not a… it’s not a lot of work, but obviously, in that space where they are paused, we will maybe miss a few.
60 00:07:11.820 ⇒ 00:07:24.579 acromie: I have one question on this one. If we had a source connected also to the Google Ad account, is there a possibility that then that…
61 00:07:24.800 ⇒ 00:07:27.680 acromie: ClickID would be available.
62 00:07:27.910 ⇒ 00:07:30.870 acromie: To… do a filter.
63 00:07:31.110 ⇒ 00:07:32.650 acromie: Does that make sense?
64 00:07:32.850 ⇒ 00:07:36.390 Zoran Selinger: So… I’m not sure what you mean by source.
65 00:07:36.390 ⇒ 00:07:43.540 acromie: So, in Amplitude, we’re connecting this as a destination, right? Because we’re trying to send the conversion out.
66 00:07:44.120 ⇒ 00:07:55.290 acromie: But if we connected each account also as a source, is it possible that we might then have the ability to feature or filter?
67 00:07:56.960 ⇒ 00:08:01.959 Zoran Selinger: it is possible. I… I… let me look into that. Let me look into that.
68 00:08:01.960 ⇒ 00:08:02.559 acromie: That would just be…
69 00:08:02.560 ⇒ 00:08:11.660 Zoran Selinger: What do we get by connecting it as a source? I understand what you mean. Let me have a look, and I’ll get back to you.
70 00:08:14.100 ⇒ 00:08:22.449 Robert Tseng: Okay, so I guess we have a couple… we’ll try, like, one more thing here, but otherwise, I think it seems like we’ll… might have to go with one of these two solutions, in which case…
71 00:08:22.770 ⇒ 00:08:24.280 Robert Tseng: Kind of, these are the steps, yeah.
72 00:08:24.280 ⇒ 00:08:32.269 acromie: I think there’s one more potential solution here that maybe was the way we were trying not to go, but might be viable, is that
73 00:08:32.270 ⇒ 00:08:51.660 acromie: What if we went back to trying to send these conversions to a GA4, and then instead of connecting directly to Amplitude, they’re connected to GA4 like they always have been, and we’re continuing to use Tag Manager to, you know, say.
74 00:08:51.660 ⇒ 00:08:59.920 acromie: fire on this page, and this is, you know, the connection, or whatever. But, I’m not 100% sure on that, but…
75 00:09:01.050 ⇒ 00:09:03.339 acromie: Wanted to throw that in there as well.
76 00:09:05.530 ⇒ 00:09:06.350 Robert Tseng: Yeah, well, I guess…
77 00:09:06.350 ⇒ 00:09:08.270 Zoran Selinger: So if we can…
78 00:09:08.270 ⇒ 00:09:24.250 acromie: I know in the end, the custom solution is going to be probably the best way to go. My concern is not knowing what the cost is going to be, and knowing that our budget for the rest of this year is pretty, tied up. I want to make sure that we have either a…
79 00:09:24.720 ⇒ 00:09:39.850 acromie: interim plan and a where we’re gonna go as soon as I absolutely can, or if there’s really nothing we can do, and we have to go forward now, I need to know if I need to go find the budget to make it happen. So…
80 00:09:40.810 ⇒ 00:09:41.470 acromie: Hmm.
81 00:09:42.580 ⇒ 00:09:45.960 Robert Tseng: Yeah, I guess this is kind of like your…
82 00:09:46.530 ⇒ 00:09:53.159 Robert Tseng: this is, like, the server… this is the custom solution, if we were to actually build it out, right? So these are the different steps that we take here.
83 00:09:53.260 ⇒ 00:10:12.629 Robert Tseng: Yeah, I mean, cost-wise, like, it’s just, like, it’s a function of our time, right? So, I think we could do it within a month, so it’d probably just be another month extension, kind of, to what we’ve had, so we can kind of talk about the budget details, like, afterwards. But if, you know, if you don’t feel like you can get that, we need to do the GTM approach.
84 00:10:12.780 ⇒ 00:10:20.849 Robert Tseng: I mean, I guess that was kind of, like, one of our initial kind of conversations on how we could approach it. Yeah, it’s probably less hands-on than this, like.
85 00:10:21.030 ⇒ 00:10:26.209 Robert Tseng: Timing-wise, probably, I mean, it would be shorter, I would say.
86 00:10:26.750 ⇒ 00:10:30.140 Robert Tseng: Yeah, so… but then it’s, like, kind of, we’d probably have to…
87 00:10:31.000 ⇒ 00:10:42.000 Robert Tseng: I forgot exactly what we discussed the risks were, maybe it was, like, on the server-side GTM… or no, this wasn’t even server-side GTM, this was just, like, using GTM itself, so… Yeah.
88 00:10:42.270 ⇒ 00:10:50.430 Robert Tseng: Yeah, I think… I think we said that there were some issues, possibly, with replicating GA4 property settings into the GTM.
89 00:10:50.430 ⇒ 00:11:09.620 Robert Tseng: And then, like, a lot of our attention would go into the stitching piece as far on the QA, right? So, yeah, I mean, we could kind of create another slide and just make this more clear on, like, if we went the GTM route, so we’re on, I guess, so maybe that’s the action item we could take here. So, this is kind of like the…
90 00:11:09.780 ⇒ 00:11:15.439 Robert Tseng: the custom, like, approach, and then we can do the… we can have another one for the GTM approach.
91 00:11:15.650 ⇒ 00:11:19.120 Robert Tseng: But any other kind of thoughts there before I move on?
92 00:11:23.620 ⇒ 00:11:24.949 acromie: Not for me, nope.
93 00:11:27.160 ⇒ 00:11:28.140 Robert Tseng: Okay, cool.
94 00:11:28.280 ⇒ 00:11:40.950 Robert Tseng: So yeah, so that’s that. I think, you know, just to kind of, like, align on the outcome here, so, you know, obviously, this has been honestly slower than all of us all wanted it to be, so I think just kind of figuring out, like.
95 00:11:40.950 ⇒ 00:11:41.270 acromie: Yeah.
96 00:11:41.270 ⇒ 00:11:46.460 Robert Tseng: Okay, yeah, like, let’s work… you know, what…
97 00:11:46.730 ⇒ 00:11:59.380 Robert Tseng: if we’re gonna rule this out, I know there are a lot of ad accounts that we haven’t connected. We’ve kind of received 4. I remember hearing some that there were, like, 15 plus, and then possibly, you know, up to 30 in the kind of, like, potential ones, so…
98 00:11:59.380 ⇒ 00:12:17.610 Robert Tseng: I just want to make sure that we’re making, like, a big enough dent in this, you know, that it’s, like, actually going to help you move the needle. Like, I know you’ve been giving us a lot of re… like, you know, leeway to kind of, like, try different options here, but I just want to make sure you feel good about, kind of, the pace that we’re moving at on this, because I feel like.
99 00:12:17.610 ⇒ 00:12:18.580 acromie: Yeah, and I know.
100 00:12:18.580 ⇒ 00:12:19.130 Robert Tseng: chat.
101 00:12:19.420 ⇒ 00:12:29.400 acromie: A lot of pay stuff has 100% been on me. With Adam being out, with just other things we’ve been doing, like, I have not been able to give as much time as I would have liked to this.
102 00:12:29.400 ⇒ 00:12:39.600 acromie: And so yeah, nothing… I have no concerns whatsoever about how you have been reactive and been able to move forward at all. This has been wonderful.
103 00:12:41.000 ⇒ 00:13:02.700 acromie: I, and then… oh, what was I gonna say? As far as the number of accounts we’ve connected, I am okay with where we’re at, given that we’ve now found an issue. I don’t want to have so many, you know, be working on adding in so many before we know we have the right solution, because eventually we’re going to have to replicate this up to, like.
104 00:13:02.770 ⇒ 00:13:04.970 acromie: 200-some accounts.
105 00:13:04.970 ⇒ 00:13:27.870 acromie: So, I would rather keep this initial group that we’re trying to connect smaller. I am slowly getting more… I’ve had quite a few agencies who have been, like, clamoring for this, taking really long time to get back to me, so, like… I see, yeah. So they’re slowly coming in. I do have maybe another one I could send over, but, like…
106 00:13:28.120 ⇒ 00:13:32.130 acromie: It’s been kind of like pulling teeth on that side, so…
107 00:13:32.210 ⇒ 00:13:41.770 Robert Tseng: Yeah. But yeah, like I said, like, I don’t want to spend all the time connecting a million accounts only to find out that we have to do something completely different. Like, that’s not a good use of anybody’s time.
108 00:13:41.810 ⇒ 00:13:47.420 acromie: Yeah. So that’s… I feel like we’re in a good place there at the moment, so…
109 00:13:47.420 ⇒ 00:13:57.559 Robert Tseng: Okay. Yeah, that makes sense. I mean, obviously, from the agencies, I can see it from their perspective, too, if it’s kind of like, hey, we have something that we want you to test out, they’re maybe not so eager to jump at it, so…
110 00:13:57.560 ⇒ 00:14:09.819 acromie: Well, they were excited. They sent me over, like, yes, right away, but then when I asked for them to do this X, Y, and Z in order to get us the access, it’s been a little slower, so…
111 00:14:09.820 ⇒ 00:14:11.240 Robert Tseng: Yeah, for sure.
112 00:14:12.470 ⇒ 00:14:27.830 Robert Tseng: Okay, well, I mean, I think we have some viable options, line of sight, like, this isn’t, like, we’re not… we’re not going in blind at this point, so we… we know… we know, kind of, like, the… the path ahead of us. We do think that there… it’s… yeah, we do think this… we… we could get this… we could get this very quickly, so…
113 00:14:27.830 ⇒ 00:14:34.509 Robert Tseng: I think that’s, that’s all I want to, you know, reassure you that, you know, we’re on track there.
114 00:14:34.610 ⇒ 00:14:50.499 Robert Tseng: The other objective, I think, is maybe a bit more, kind of, like, longer term, and I want to give us some time to discuss this and make sure that we’re, you know, given, kind of, your constraints and, like, how much attention you want to give to it, like, really delivering something, that’s… that’s gonna…
115 00:14:50.770 ⇒ 00:14:54.940 Robert Tseng: Have we been able to kind of get some momentum on the product analytics side.
116 00:14:55.050 ⇒ 00:15:13.749 Robert Tseng: So, obviously, with Amplitude, like, one of the big purposes is to really create really robust cohorts, and then also, like, on the reporting side, we know that you have some initial reports, but we want to get deeper and flush out some real, kind of, analysis to help you make decisions there. So, just from, like, kind of this is gathering.
117 00:15:13.750 ⇒ 00:15:30.979 Robert Tseng: some context from what we’ve observed, what’s already there, kind of the gaps in the tracking that Shreya’s already pointed out, and then kind of, like, you know, kind of next stages that I’ve typically seen, with teams that we’ve worked with. And so I just want to kind of spend some time walking through this, because this might.
118 00:15:30.980 ⇒ 00:15:31.420 acromie: You a little bit.
119 00:15:31.420 ⇒ 00:15:33.170 Robert Tseng: Unfamiliar for your team?
120 00:15:33.280 ⇒ 00:15:41.989 Robert Tseng: So yeah, we were focused on, kind of, appointment… appointment setting, kind of just, like, what that funnel really looks like, because data is being passed through a few systems.
121 00:15:42.080 ⇒ 00:15:53.139 Robert Tseng: Sorry, this is not really… this is not zoomed in. This is just a screenshot from what you had sent me before. And so, you know, everything, like, down here, we’re familiar with. We’ve even seen the health fully data here.
122 00:15:53.140 ⇒ 00:16:11.059 Robert Tseng: you know, the patient profiles were actually stitching pretty well, like, we have all this data that maybe you weren’t privy to before. You know, maybe this stuff around, kind of, like, CS data, like, valiant, like, EHR data, we haven’t seen yet, and I don’t really think that’s coming to amplitude anyway. Correct.
123 00:16:11.060 ⇒ 00:16:26.700 acromie: Not set up yet. We… we lost our developer internally, so it’s, like, the… the pipeline is set up in Azure for that to be… get started, but, like, I don’t have anyone to code it right now, is the… is the issue there.
124 00:16:28.740 ⇒ 00:16:33.899 Robert Tseng: Meaning, like, what part of development are you, like, are you missing?
125 00:16:33.900 ⇒ 00:16:42.079 acromie: Like, actually making sure that we’re syncing between the two. Got it. So, writing the code that’s gonna do that…
126 00:16:42.080 ⇒ 00:16:42.470 Robert Tseng: Yeah.
127 00:16:42.470 ⇒ 00:16:43.150 acromie: sync.
128 00:16:43.900 ⇒ 00:16:56.850 Robert Tseng: Okay, I guess that can… that’s another can of worms we can open. I know we haven’t really gotten too much into kind of seeing how we can partner with your internal data team, but, I mean, that seems like a data orchestration kind of problem, and.
129 00:16:56.850 ⇒ 00:17:14.899 acromie: Yeah, and I think… I do have someone in another department that kind of sits next to us that could potentially do some of this. I think what I’d like to do is make sure that we’re really clear on, like, what we want coming in and where it’s going, so that when I bring them in on this, like…
130 00:17:15.530 ⇒ 00:17:21.989 acromie: it’s just really clear, so I know we’ve kind of been talking over the data and what we need and where to go, so…
131 00:17:21.990 ⇒ 00:17:22.560 Robert Tseng: Yeah.
132 00:17:22.560 ⇒ 00:17:30.079 acromie: like, I kind of want to take what we’ve been working on, and then make sure that that’s solid, too, so they can partner with us here.
133 00:17:30.650 ⇒ 00:17:31.320 Robert Tseng: Okay.
134 00:17:31.510 ⇒ 00:17:36.230 Robert Tseng: Yeah, I mean, from the work that we’ve done so far, just to kind of reference, like, this, like,
135 00:17:36.480 ⇒ 00:17:44.810 Robert Tseng: Yeah, I mean, we’re… yeah, like, this event stream data that’s kind of from Health Emily, plus amplitude, like, web interactions, very clear.
136 00:17:45.090 ⇒ 00:17:55.030 Robert Tseng: maybe some adjustments on, kind of, like, tracking additional events or whatever, but, that’s, you know, generally from Healthily, they’re giving you the whole kitchen sink, so there’s plenty of data to comb through.
137 00:17:55.030 ⇒ 00:17:55.980 acromie: Yup.
138 00:17:56.320 ⇒ 00:18:00.500 Robert Tseng: I don’t know, it seems like you’re sending a lot of that into the warehouse, and then.
139 00:18:00.500 ⇒ 00:18:01.170 acromie: Yeah, so…
140 00:18:01.170 ⇒ 00:18:01.780 Robert Tseng: there.
141 00:18:02.160 ⇒ 00:18:10.649 acromie: Right, and so, yeah, Healthfully also goes into the data ware… you know, into Valent. Valent is a, a daily dump into the data warehouse.
142 00:18:10.650 ⇒ 00:18:11.130 Robert Tseng: Always.
143 00:18:11.130 ⇒ 00:18:24.819 acromie: be after the fact. But that’s where we’re going to get more of the long-term data. Did they attend the appointment? How many appointments have they attended? Like, that’s… that’s the data we’re missing at this point.
144 00:18:25.350 ⇒ 00:18:26.000 Robert Tseng: Got it.
145 00:18:26.120 ⇒ 00:18:40.829 Robert Tseng: Yeah, so, I mean, there seems to be, like, an opportunity for bi-directional syncing here, where you get those longer-term behavioral signals within the warehouse, because Valent is able to pass all that in there. Yep. And then, you know, this is kind of part of the roadmap as well that we put together.
146 00:18:40.830 ⇒ 00:18:46.459 Robert Tseng: But the second initiative, the idea of building, like, a golden data set, it’s kind of just, like.
147 00:18:46.460 ⇒ 00:18:51.959 Robert Tseng: That master, like, patient profiles, like, it’s a long, it’s a really wide, flat table.
148 00:18:51.960 ⇒ 00:19:00.029 Robert Tseng: That captures all the demographics, captures appointments, you know, life cycle stages, revenue, LTV, and just, like, everything that you would want to know about the patient.
149 00:19:00.030 ⇒ 00:19:11.749 Robert Tseng: Obviously, you can strip out the PII, but if you put that back into amplitude, that gives you a lot more, you know, options to be able to create really robust segments, that you can, you know, do for further reporting.
150 00:19:11.750 ⇒ 00:19:22.420 Robert Tseng: or if you want to, since you’re using Amplitude as your CDP, if you want to reverse ETL that into other kind of, like, tools that you use. I didn’t see, like, a CRM platform or anything here.
151 00:19:22.420 ⇒ 00:19:31.220 acromie: We don’t have it yet, but that’s, like, next steps in… in tech, you know, our… architecture here.
152 00:19:31.220 ⇒ 00:19:34.989 Robert Tseng: Cool. Yeah, so I mean, there’s even whatever you end up using for emails.
153 00:19:34.990 ⇒ 00:19:35.520 acromie: So, basically.
154 00:19:35.520 ⇒ 00:19:43.920 Robert Tseng: kind of, like, distribution channel for messaging like that, you know, that’s kind of what this work would really be preparing you for. So.
155 00:19:43.920 ⇒ 00:19:44.450 acromie: Yeah.
156 00:19:44.450 ⇒ 00:19:45.140 Robert Tseng: That’s a big one.
157 00:19:45.140 ⇒ 00:19:48.530 acromie: How do we get that back out to the ad channels? So…
158 00:19:48.530 ⇒ 00:19:49.080 Robert Tseng: Yep.
159 00:19:49.290 ⇒ 00:19:55.230 Robert Tseng: Yeah, so ads definitely seems like it’s the main thing first to go after. Yeah, that’ll help with retargeting, for sure.
160 00:19:55.230 ⇒ 00:19:57.970 acromie: Yeah. So yeah, we can… we can talk about, like.
161 00:19:57.970 ⇒ 00:20:08.190 Robert Tseng: Yeah, and this probably goes hand-in-hand with what you’re saying. Want to be able to specifically describe, like, well, what do you need from these systems so that we can actually get what we need for a targeting campaign?
162 00:20:08.190 ⇒ 00:20:15.829 acromie: Right. Now, when you say retargeting, are you familiar with a way to retarget that is HIPAA compliant? Because I am not. I am under the.
163 00:20:15.830 ⇒ 00:20:16.260 Robert Tseng: Yeah.
164 00:20:16.260 ⇒ 00:20:20.780 acromie: impression that you… we can send, like.
165 00:20:21.010 ⇒ 00:20:30.819 acromie: We can send demographics in general, like, hey, here’s an audience profile out, but we can’t send, like, specific, like, we can’t…
166 00:20:30.980 ⇒ 00:20:33.230 acromie: retarget someone.
167 00:20:33.230 ⇒ 00:20:44.130 Robert Tseng: Yeah, we can’t tell them, hey, you booked this type of appointment before here, like, why don’t you try… we know you did this, why don’t you try that? Like, yeah, it’s not really… it’s not truly retargeting. I guess that’s just kind of…
168 00:20:44.130 ⇒ 00:20:45.920 acromie: Cool, I just want to make sure we are on the same page.
169 00:20:45.920 ⇒ 00:20:50.759 Robert Tseng: Yeah, yeah. Yeah, I understand that you can’t, you can’t…
170 00:20:51.200 ⇒ 00:21:01.530 Robert Tseng: I mean, products and services… specific products and services to sign up for, you can’t use that in your messaging. Yeah. Yeah, yeah, anyway, so, it is mostly demographic and, yeah.
171 00:21:01.530 ⇒ 00:21:07.140 acromie: Once we have a CRM, then we can do this kind of stuff, because we’re sending direct, we’re not advertising.
172 00:21:07.140 ⇒ 00:21:08.400 Robert Tseng: And you’re not using the ads, right?
173 00:21:08.400 ⇒ 00:21:10.429 acromie: Using Facebook, right? Yes.
174 00:21:10.790 ⇒ 00:21:24.580 Robert Tseng: Yeah, so I do think the ads enablement is an easier lift, because, you know, you don’t really need that much, you probably already have most of what you need, so there wouldn’t really be any, necessarily, you know, more stuff to do there. But,
175 00:21:24.900 ⇒ 00:21:38.450 Robert Tseng: Yeah, okay, and so… and then, obviously, this is just, like, from a measurement perspective. Once we push data into, these platforms, like, what… what is that lift going to look like? Well, it’s going to take some time to track that.
176 00:21:38.800 ⇒ 00:21:39.310 acromie: get it.
177 00:21:39.310 ⇒ 00:21:43.460 Robert Tseng: So, I’ll pass… we’ll pause there for a second.
178 00:21:43.790 ⇒ 00:21:45.470 Robert Tseng: And,
179 00:21:46.160 ⇒ 00:21:53.610 Robert Tseng: yeah, just kind of… actually, I… I think the package delivery person just rang my bell, so I’m gonna open the door for him.
180 00:21:53.610 ⇒ 00:21:58.730 acromie: I’ll be back in, like, 10 seconds. Okay.
181 00:22:00.600 ⇒ 00:22:01.450 acromie: Hmm.
182 00:22:02.890 ⇒ 00:22:21.889 Zoran Selinger: Alright, so, while we wait, I’m just reading about, about the source. It’s, it’s, it’s very simple, so the, the, what source actually does is just, gives you one daily event called Daily Admetrics with your clicks.
183 00:22:22.680 ⇒ 00:22:33.779 Zoran Selinger: impressions, campaign ID, ad group ID, and it’s… it’s on the ad level, so there’s… there’s… there is no click ID or anything in there, so… Okay, bummer.
184 00:22:33.780 ⇒ 00:22:34.150 acromie: Oh.
185 00:22:34.150 ⇒ 00:22:37.810 Zoran Selinger: No use there. But a good idea.
186 00:22:37.970 ⇒ 00:22:38.710 acromie: Yeah.
187 00:22:38.710 ⇒ 00:22:40.019 Zoran Selinger: from what I’ve seen.
188 00:22:40.230 ⇒ 00:22:55.210 Zoran Selinger: their native… native implementations are super simple, here, so I would be surprised if, it was kind of intelligent enough to… to.
189 00:22:55.210 ⇒ 00:22:57.170 acromie: connectos, to lingos. Right.
190 00:22:57.170 ⇒ 00:23:04.370 Zoran Selinger: Right. So, yeah, no surprise there, maybe. I mean, the conversation is open there,
191 00:23:05.300 ⇒ 00:23:08.570 Zoran Selinger: So, yeah, we’ll see.
192 00:23:09.720 ⇒ 00:23:21.830 Zoran Selinger: It would be… I mean, I can ask for that, that we… basically, they could… they could add a simple call, API call, on their side.
193 00:23:21.830 ⇒ 00:23:35.270 Zoran Selinger: to look up the account ID for the click ID, and just not, not send the invo… not send the request, to upload the click, if it’s not the correct account, but…
194 00:23:36.240 ⇒ 00:23:37.060 acromie: Yeah.
195 00:23:37.060 ⇒ 00:23:41.310 Zoran Selinger: I don’t think that they will do that. There’s probably other priorities.
196 00:23:41.480 ⇒ 00:23:49.170 Zoran Selinger: Right, right. Because it’s simply a nice-to-have, right? Yep, yep. It’s never going to be a priority.
197 00:23:49.370 ⇒ 00:23:56.120 Zoran Selinger: Yeah. So, yeah, we… but yeah, we do have those two options, enabling
198 00:23:56.320 ⇒ 00:24:05.559 Zoran Selinger: GA4 properly, and then we do have that option of a custom solution. Yeah. Okay, so Robert’s back.
199 00:24:05.960 ⇒ 00:24:06.630 Robert Tseng: Shutzu.
200 00:24:06.760 ⇒ 00:24:17.729 Robert Tseng: Yeah, no, thanks for… thanks for just kind of jumping on that, on this call. Yeah, so I guess just to kind of make sure we have some kind of, like, adjustments that we can go back and make, so…
201 00:24:17.730 ⇒ 00:24:23.959 acromie: I mean, I shared this link with you as well, so you guys can kind of pour over. There’s some other slides, just appendix things about, like.
202 00:24:23.980 ⇒ 00:24:28.130 Robert Tseng: And a lot of it’s just kind of, you know, other samples of…
203 00:24:28.480 ⇒ 00:24:43.289 Robert Tseng: things that we feel, how we think about this problem, whatever. But yeah, I guess, like, how… how is this, how’s this hitting? Like, how do we, like, what… what… what adjustments do you feel like we need to come back to you with? Like, what… what more clarity do you need so that we can.
204 00:24:43.290 ⇒ 00:24:43.870 acromie: Yeah.
205 00:24:43.870 ⇒ 00:24:45.369 Robert Tseng: Gotta come back into normal.
206 00:24:45.370 ⇒ 00:24:56.759 acromie: I think what I want to know is just, out of all the things we’ve talked about, like, being really, really clear about what is in scope with what we’ve started, that maybe we… if there’s anything we haven’t done yet.
207 00:24:56.760 ⇒ 00:25:07.179 acromie: Versus what is going to be incremental, you know, another phase, etc. So that’s… that’s the only place where I’m just a little on, like, unsure.
208 00:25:07.390 ⇒ 00:25:23.340 Robert Tseng: Okay. Yeah, I mean, I can go back to that Notion doc when we’ve originally started. We’ve kind of answered every question on there, so that’s… that’s kind of, like, the before, like, current state, before the current state of what we’ve done. And then this is mostly incremental. I think there’s a couple things that have carried over, like.
209 00:25:23.340 ⇒ 00:25:37.929 Robert Tseng: this we fully expect it to be done before at this point, but, you know, whatever, I guess things just drag out a little longer. So, most of this is net new, it’s incremental in terms of that it’s building upon what we’ve started, but, I think…
210 00:25:38.500 ⇒ 00:25:54.420 Robert Tseng: I feel like the value of this is pretty kind of clear. If anything, we’ll adjust to make the, like, well, what if we just ended up doing the GTM setup instead? Like, how would that impact the timeline? And then, Treya, maybe just kind of on our end, like, for this.
211 00:25:54.520 ⇒ 00:26:06.079 Robert Tseng: Well, yeah, I think we… this is gonna be a bit more of a coordinated effort. Like, if you have folks internally that you want to leverage for this, like, we can do data orchestration, data engineering, everything, like, that’s kind of…
212 00:26:06.080 ⇒ 00:26:06.460 acromie: What the hell.
213 00:26:06.460 ⇒ 00:26:15.070 Robert Tseng: like the bread and butter of, like, my background as well, so I would love to work with you on this if you would get that opportunity, like, I think this would be a big win.
214 00:26:15.070 ⇒ 00:26:24.269 acromie: give me the numbers, and I will… we’ll look into it, right? Like, I’m not saying no, because I don’t know how much time that other department has right now, so…
215 00:26:24.270 ⇒ 00:26:25.040 Robert Tseng: Sure.
216 00:26:25.160 ⇒ 00:26:31.820 Robert Tseng: Okay, and then for this, I think, Shreya, like, just in terms of, like, where we left off with reporting, like.
217 00:26:32.030 ⇒ 00:26:47.040 Robert Tseng: okay, well, what do these enhancements, like, look like, and kind of what, you know, what recommendations do we feel like are going to come out of, like, the next, next phase? I think is important. This is, you know, a bit longer term, but,
218 00:26:47.500 ⇒ 00:26:51.290 Robert Tseng: I think this is more like, okay, once we have these… once…
219 00:26:51.620 ⇒ 00:26:58.879 Robert Tseng: This is assuming that data is all flowing into the data warehouse, and we have controls there to be able to, like.
220 00:26:59.030 ⇒ 00:27:08.369 Robert Tseng: have that golden data set, we’re able to move data wherever we please. Like, what… what platform should we push it to first? It’s probably ads, and we can kind of,
221 00:27:08.370 ⇒ 00:27:21.190 Robert Tseng: Like, that’s kind of what this is really trying to tee up, and then, like, figuring out, like, well, what specific campaigns do you want to run on those app… ad platforms, so that we can really, model out, like, what exactly we need to send to them.
222 00:27:21.190 ⇒ 00:27:22.460 acromie: Like.
223 00:27:22.460 ⇒ 00:27:31.710 Robert Tseng: that’s… yeah, I think that’ll… the actual development time doesn’t take that long, but I think we’re a couple steps away from that, so that’s why that’s kind of projected out as a bit longer… more longer term.
224 00:27:31.710 ⇒ 00:27:33.019 acromie: For sure. So, yeah, and I think…
225 00:27:33.350 ⇒ 00:27:41.500 acromie: Right now, like, our number one priority is just getting those conversion events to be able to sync back, like…
226 00:27:41.620 ⇒ 00:27:49.589 acromie: that is the thing I need to solve, and as quickly as possible, because the whole system is missing out on that, so…
227 00:27:49.870 ⇒ 00:27:50.500 Robert Tseng: Sure.
228 00:27:50.600 ⇒ 00:27:59.440 Robert Tseng: Yeah, so that’s definitely still, like, one of our primary objectives here, like, that’s what we want to… this is the… this is still, like, kind of the focus area.
229 00:27:59.440 ⇒ 00:28:12.380 Robert Tseng: I think this work, it just takes a bit longer to kind of get the ROI that you want, so, you know, we’ve already started on some stuff. I do think it’ll take kind of a little bit more time to realize the full benefit of it, but…
230 00:28:12.380 ⇒ 00:28:22.639 Robert Tseng: If we can get involved in helping, kind of, build some of these initial connections, make sure that, you know, you know, just being able to be in the warehouse and have some
231 00:28:22.730 ⇒ 00:28:32.949 Robert Tseng: we can advise on, like, what architecture decisions you need to make in order… is nightly exports enough? Do we need more refreshes? Maybe this is, you know, I think it…
232 00:28:32.950 ⇒ 00:28:44.649 Robert Tseng: I think the answer is, for some data, you want it more frequently. It’s probably… you would want it hourly. Maybe, like, other data is… it’s fine if it just, kind of refreshes on a daily basis, and so…
233 00:28:44.650 ⇒ 00:28:51.979 Robert Tseng: Could just kind of get ahead of that. So, yeah, like, you know, a month from now, like, when you’re ready to…
234 00:28:52.300 ⇒ 00:29:04.850 Robert Tseng: you have your… you’re ready to focus on sending that data into the ads platform, or maybe next… or yeah, if you… whenever you bring in that CRM system, we’ll have something ready to push into it immediately. Yeah.
235 00:29:04.850 ⇒ 00:29:05.400 acromie: Yep.
236 00:29:05.650 ⇒ 00:29:06.860 acromie: Exactly.
237 00:29:07.530 ⇒ 00:29:08.230 acromie: Okay.
238 00:29:08.230 ⇒ 00:29:08.840 Robert Tseng: Okay.
239 00:29:09.010 ⇒ 00:29:19.199 Robert Tseng: I guess… oh, we’re at… we’re at time. So, but, I mean, Adam, if you… I don’t know, we haven’t really talked too much about your roadmap and what you’re working on, but we’d love to know, kind of, kind of how we could better
240 00:29:19.230 ⇒ 00:29:33.279 Robert Tseng: partner in supporting you as well, like, you’re obviously the primary analyst with the more… with more, context. We’re kind of trying to build the scaffolding to make, you know, your… enable your work, so, would… would love to kind of get your thoughts on this as well.
241 00:29:33.450 ⇒ 00:29:35.450 Robert Tseng: If not on this call, like, later on.
242 00:29:35.450 ⇒ 00:29:53.759 Adam Kittleson: Yeah, yeah. I mean, I’ve got, like, a couple initially. I mean, everything that is being worked on is, like, kind of the last pieces of what I think I need for my role to, like, really have that foundation, so I think we’ve got good coverage there. But, yeah, if anything… if I think of anything else, you know, in the next…
243 00:29:53.890 ⇒ 00:30:05.649 Adam Kittleson: couple weeks or whatever, I’ll definitely reach out. I’m just… today, it’s a little hectic, because I’m just, like, digging out of my inbox and everything, so I… It’s tough to, like, get my mind back into this mindset. Sure.
244 00:30:05.650 ⇒ 00:30:06.389 Robert Tseng: All good.
245 00:30:07.490 ⇒ 00:30:21.620 Robert Tseng: Okay, cool. Well, that was very productive. I think we got through a lot. We’ll make some of these adjustments. We’ll probably send it out the next day. Hopefully, we can kind of have a… have a path moving forward. And, you know, yeah, like, I think,
246 00:30:22.050 ⇒ 00:30:31.170 Robert Tseng: either this or last week is technically our last, kind of, the kind of expiration of the contract from the previous one, so I want to make sure that we’re in a good place when we forward, so…
247 00:30:31.170 ⇒ 00:30:41.240 acromie: And Robert, you mentioned answering questions in Notion. Were you just mentioning the initial, like, questions doc that you sent us, or was there one I’m not seeing?
248 00:30:41.240 ⇒ 00:30:42.899 Robert Tseng: Yep, yeah, that’s… that’s the one.
249 00:30:43.240 ⇒ 00:30:44.130 acromie: Okay, cool.
250 00:30:45.440 ⇒ 00:30:47.409 acromie: Yeah, get numbers in front of me.
251 00:30:48.190 ⇒ 00:30:53.139 acromie: Okay, sure. As soon as you can.
252 00:30:53.140 ⇒ 00:30:57.109 Robert Tseng: Yeah, I will, I will send, I’ll… we’ll send that soon. Okay, thank you.
253 00:30:57.110 ⇒ 00:30:57.820 acromie: Thank you.
254 00:30:57.820 ⇒ 00:30:58.510 Robert Tseng: Alright.
255 00:30:58.510 ⇒ 00:31:00.730 acromie: Alright, bye. Bye.