Meeting Title: Brainforge <> Ellie Check-in Date: 2025-08-27 Meeting participants: Zoran Selinger, acromie, Shreya Chowdhury, Adam Kittleson, Robert Tseng
WEBVTT
1 00:01:08.080 ⇒ 00:01:09.500 acromie: Hello!
2 00:01:12.500 ⇒ 00:01:13.590 Zoran Selinger: Sometime.
3 00:01:14.170 ⇒ 00:01:14.820 acromie: real.
4 00:01:15.200 ⇒ 00:01:17.139 Zoran Selinger: Yeah, good, good.
5 00:01:17.420 ⇒ 00:01:21.650 acromie: Good. Good, good. I assume Adam will be jumping on shortly here, too.
6 00:01:21.650 ⇒ 00:01:30.419 Zoran Selinger: Yeah, I think Robert will also join us for an intro, I think, and then he has to join. Great! Yeah.
7 00:01:30.930 ⇒ 00:01:31.760 Zoran Selinger: ….
8 00:02:15.150 ⇒ 00:02:16.060 Robert Tseng: Hey, everyone.
9 00:02:17.520 ⇒ 00:02:18.510 acromie: Hey, hey!
10 00:02:22.860 ⇒ 00:02:31.800 Robert Tseng: I don’t know if you guys have introduced… made intros yet, but I just wanted to… I kind of mentioned on Slack, but Trey has been helping out on this project, too, so…
11 00:02:32.550 ⇒ 00:02:33.750 Robert Tseng: later on to this check-in.
12 00:02:34.250 ⇒ 00:02:39.049 acromie: Yeah, I hadn’t seen Slack yet. We’ve been… I’ve been in back-to-backs all morning. Okay.
13 00:02:40.080 ⇒ 00:02:40.810 acromie: But.
14 00:02:41.860 ⇒ 00:02:53.989 Robert Tseng: Well, I guess for today’s check-in, I think we were just gonna give you updates. I know that you’ve been talking to Zoran and Slack, we’ve been kind of going through a couple different options for what we can do with the ads conversion.
15 00:02:54.030 ⇒ 00:03:11.839 Robert Tseng: you know, one is we do all these many, many different kind of connections, or we… I mean, Zoran was tried exploring a different… different solution. I’ll let him talk through that more… in more detail. Hopefully, we just get, like, a green light on what the short term is, and then we can tell you, like, how things can change, you know, if we wanted to
16 00:03:12.130 ⇒ 00:03:29.380 Robert Tseng: maybe we just do plug in everything, and then we, we go and build out something else, like, later down the line. And then Shreya will kind of walk you through kind of where we’re at, from a product analytics perspective. We’ve gone through the, you know, the core workflows for everything.
17 00:03:29.520 ⇒ 00:03:34.230 Robert Tseng: … up to when, I guess, a user, kind of.
18 00:03:34.650 ⇒ 00:03:50.759 Robert Tseng: makes a purchase or kind of converts and starts scheduling in. So, I think she’s got… she kind of organized everything and has some updates there on tracking changes that we would recommend, maybe not in the short term, but also, like, reporting things that we could do in the short term.
19 00:03:51.200 ⇒ 00:03:52.219 acromie: Okay, great.
20 00:03:53.260 ⇒ 00:04:01.769 Robert Tseng: Okay, cool. And then I think I’ll just, apologize, but I will have to drop off a bit early. It was… the reschedule wasn’t, like, that ideal for me, so I….
21 00:04:01.770 ⇒ 00:04:02.540 acromie: No worries.
22 00:04:02.540 ⇒ 00:04:04.759 Robert Tseng: for… until that happens. Yeah.
23 00:04:05.440 ⇒ 00:04:10.790 Robert Tseng: Okay, cool. Zoran, you want to go first? Just because I think there’s more to cover on that side.
24 00:04:10.790 ⇒ 00:04:11.110 Zoran Selinger: Sure.
25 00:04:11.110 ⇒ 00:04:13.399 Robert Tseng: That needs immediate attention.
26 00:04:13.590 ⇒ 00:04:22.490 Zoran Selinger: Sure, of course. Well, we finally got a successful connection test on the… yeah.
27 00:04:22.490 ⇒ 00:04:22.930 acromie: Awesome!
28 00:04:22.930 ⇒ 00:04:25.220 Zoran Selinger: Yeah, we got there. …
29 00:04:25.940 ⇒ 00:04:45.799 Zoran Selinger: For… it’s… it’s an undocumented behavior, so their documentation, basically, is… that’s not how it is at all. Because in their documentation, it clearly says that you put an email that created the service account, not the email of the service account itself.
30 00:04:45.950 ⇒ 00:04:50.619 Zoran Selinger: It is the email of the service account, it’s the long one.
31 00:04:50.730 ⇒ 00:05:10.089 Zoran Selinger: One has to have access to the account. Good thing there is, we don’t have to… there’s no process of sending an invite, then an invite has to have… needs to be accepted, there’s nothing like that. We just add a user with a standard level of access, and everything… everything will work.
32 00:05:10.370 ⇒ 00:05:14.910 Zoran Selinger: Short term, that’s… that’s all there is to it.
33 00:05:14.960 ⇒ 00:05:38.099 Zoran Selinger: So, add a user, and then you can configure… you can configure it the way… the way you did. You select your events and filters, you map GCLID to GCL ID, that’s fine. You obviously know how to create a conversion action ID and all that, so that’s… that’s all good and well there.
34 00:05:38.810 ⇒ 00:05:40.540 Zoran Selinger: So, the short term.
35 00:05:40.880 ⇒ 00:05:48.750 Zoran Selinger: We can do this, of course. I don’t know how many, how many really active partners you have at the moment that you’re.
36 00:05:48.750 ⇒ 00:05:49.770 acromie: So…
37 00:05:49.770 ⇒ 00:06:10.220 acromie: That’s a really good question that we don’t know. So part of this, now that we know we can do this, my goal is to now go back to the system and say, hey, if you want to get access to these conversion points, which we know everybody does, we need to get access to your account, because we don’t have access to everyone’s ads account right now.
38 00:06:10.220 ⇒ 00:06:27.589 acromie: And so, I’m not sure… I mean, I would imagine that every location likely has one. The question is whether or not, you know, do they have one for each location, or are they running all ads off of, you know, one account for multiple locations, and just doing the…
39 00:06:27.590 ⇒ 00:06:38.540 acromie: targeting, you know, so that’s… those are the things I don’t know, because in general, this has been run by outside agency partners, and we are at a point where,
40 00:06:38.820 ⇒ 00:06:40.340 acromie: we’re…
41 00:06:40.340 ⇒ 00:07:00.419 acromie: we have far more franchisees going off and doing their own thing than we have in the past, and so we’ve kind of lost sight. So our goal is to go out and say, hey, we can connect you, but we need this level of access so we can do X, Y, and Z, and then we will make sure you get, you know, plugged into the system. So…
42 00:07:00.420 ⇒ 00:07:08.310 acromie: That is 100% something we can, you know, we’ll take on and go forward with, as long as we know it’s working, so….
43 00:07:08.310 ⇒ 00:07:08.840 Zoran Selinger: Yeah.
44 00:07:09.780 ⇒ 00:07:13.070 Zoran Selinger: Yeah, that is fine. The…
45 00:07:14.430 ⇒ 00:07:19.950 Zoran Selinger: This native, this native integration that we see there, ….
46 00:07:21.860 ⇒ 00:07:26.300 Zoran Selinger: is… For one account, and one conversion action.
47 00:07:27.420 ⇒ 00:07:28.360 Zoran Selinger: Can you imagine?
48 00:07:29.320 ⇒ 00:07:29.900 Zoran Selinger: How many….
49 00:07:29.900 ⇒ 00:07:32.400 acromie: Did you order every conversion action?
50 00:07:33.630 ⇒ 00:07:40.159 Zoran Selinger: the way… you see the options there, it’s just you put one conversion action ID into the connector.
51 00:07:41.630 ⇒ 00:07:56.740 Zoran Selinger: So we… we have a… even, like, in short term, if we currently want to do this for 5 accounts, fine, that is doable, but you can imagine… you can imagine having to create 5
52 00:07:57.120 ⇒ 00:08:02.840 Zoran Selinger: Five connections for one For one account. Right. Sub-account.
53 00:08:02.840 ⇒ 00:08:21.880 acromie: Oh, that is annoying. So, I mean, the good news is, the one conversion action that they don’t have access to right now is when someone schedules online. That’s what they’re saying, it’s the primary conversion action. Like, we could go forward and just say, this is it, but we are gonna want more.
54 00:08:21.880 ⇒ 00:08:22.920 acromie: Overtime.
55 00:08:22.920 ⇒ 00:08:26.030 acromie: Right? Like… Yeah.
56 00:08:27.320 ⇒ 00:08:47.149 Zoran Selinger: So this is, … I mean, the process of doing this isn’t, like, the standard operating procedure, 3 or 4 steps there, in terms of, like, you add… you add that email, you create a conversion action, you do… and you… you do… you upload… you upload the service account key file.
57 00:08:47.480 ⇒ 00:08:59.920 Zoran Selinger: you know, configure that connection, and you’re good to go. You don’t really need anyone, anyone to do it, as long as you have access to the account, and you have your manager, so you can add a user, that’s fine.
58 00:09:00.070 ⇒ 00:09:03.450 Zoran Selinger: Yep. So, that is… that is doable. I…
59 00:09:03.850 ⇒ 00:09:13.050 Zoran Selinger: So, I don’t know if Robert maybe knows the limitations of how… how many destinations can we have in the…
60 00:09:13.240 ⇒ 00:09:14.080 Zoran Selinger: ….
61 00:09:14.470 ⇒ 00:09:18.790 acromie: I’ve been told there’s no, like, limit, but… Yeah.
62 00:09:19.220 ⇒ 00:09:19.990 Zoran Selinger: Excellent.
63 00:09:19.990 ⇒ 00:09:29.679 Robert Tseng: Yeah, it’s not the limit on the number of destinations, it’s probably more on, like, how much data we push through it, which… I don’t think the volume is that much, but… I mean….
64 00:09:29.680 ⇒ 00:09:30.240 acromie: Right.
65 00:09:30.510 ⇒ 00:09:35.419 Robert Tseng: with more connectors, more things could break, so, like, I mean, I don’t know, there’s just…
66 00:09:35.610 ⇒ 00:09:36.200 Robert Tseng: I don’t really know.
67 00:09:36.200 ⇒ 00:09:51.969 acromie: I do also know that this… the ability to connect to multiple ad accounts is brand new for them, or at least it was this spring. It was a feature they brought us in on on this beta. I don’t even know if it’s still in beta or not, because we didn’t get to test it immediately.
68 00:09:51.970 ⇒ 00:10:07.780 acromie: And so, I do have connections to that product team, and could go back and say, hey, here are the things that we know, like, one, why can’t we do this at an MCC level? And two, even if we can’t, like.
69 00:10:07.780 ⇒ 00:10:12.390 acromie: We need to be able to do multiple conversion events per connection, because…
70 00:10:12.550 ⇒ 00:10:18.410 acromie: Otherwise, we’re setting up thousands of these over time, right? Like… That sounds awful.
71 00:10:18.410 ⇒ 00:10:23.450 Zoran Selinger: Yeah, so if we do have access to them, that’s a great conversation to start.
72 00:10:23.450 ⇒ 00:10:27.470 acromie: Cool. And could I copy you all, maybe, on an email to them, just to….
73 00:10:27.470 ⇒ 00:10:30.470 Zoran Selinger: Of course, I would love to talk about it.
74 00:10:30.470 ⇒ 00:10:31.740 acromie: You’re great.
75 00:10:31.740 ⇒ 00:10:33.640 Zoran Selinger: Of course. …
76 00:10:34.080 ⇒ 00:10:42.049 Zoran Selinger: So, if… if we have a system like this, there is… even if we, even if we do have, …
77 00:10:43.190 ⇒ 00:10:53.830 Zoran Selinger: an option to have multiple conversion action IDs. So, let’s say we map one event to one conversion action ID in that connection point.
78 00:10:53.950 ⇒ 00:11:01.570 Zoran Selinger: we are still gonna run into a problem of attempting to upload a click ID,
79 00:11:01.680 ⇒ 00:11:14.709 Zoran Selinger: to one subaccount that is actually from another sub-account. Yeah. So you will still trigger… trigger events, so one event will trigger multiple destinations, and only one of them will work.
80 00:11:14.830 ⇒ 00:11:25.480 Zoran Selinger: all the other ones will basically fail, because it’s a click ID from a different account. That can be sold by… if we can add, a campaign name parameter.
81 00:11:26.200 ⇒ 00:11:27.959 Zoran Selinger: to amplitude. We can feel.
82 00:11:27.960 ⇒ 00:11:28.320 acromie: Right.
83 00:11:28.320 ⇒ 00:11:29.599 Zoran Selinger: Based on that as well.
84 00:11:29.710 ⇒ 00:11:36.990 Zoran Selinger: So that could be added in, like, as one of the new UTMs, or query parameters.
85 00:11:36.990 ⇒ 00:11:48.109 acromie: Yeah, the only… the only thing is, there’s… there’s… yeah, that’s the problem. There’s so many hands in the pot. I don’t… as much as we could put something out like that, I don’t know if we’d…
86 00:11:48.370 ⇒ 00:11:52.900 acromie: be successful. I mean, it definitely wouldn’t be 100%.
87 00:11:52.970 ⇒ 00:11:54.209 Zoran Selinger: Yeah, I don’t know if we can.
88 00:11:54.210 ⇒ 00:12:05.899 acromie: Because we’re not controlling it, right? Each individual location is running their own ads, so they could, you know, we want to be able to… and even outside… well, no, this would only be ads, so…
89 00:12:07.040 ⇒ 00:12:08.820 acromie: Blah, okay.
90 00:12:10.640 ⇒ 00:12:12.510 acromie: I mean, there’s potential here.
91 00:12:13.450 ⇒ 00:12:18.030 Zoran Selinger: Yeah. So if we can in any way, shape, or form, infer
92 00:12:18.530 ⇒ 00:12:23.360 Zoran Selinger: the name or the ID of the account, That is actually…
93 00:12:23.570 ⇒ 00:12:33.209 Zoran Selinger: the campaign is coming from. I was looking into a few accounts, there’s really no campaign naming consistency anywhere, and we can’t really enforce it.
94 00:12:34.170 ⇒ 00:12:34.570 acromie: app.
95 00:12:34.570 ⇒ 00:12:37.389 Zoran Selinger: We can’t really enforce it, so, …
96 00:12:37.980 ⇒ 00:12:44.139 Zoran Selinger: I understand, so maybe there’s something we could do there.
97 00:12:44.140 ⇒ 00:12:44.520 acromie: Yeah.
98 00:12:44.520 ⇒ 00:13:01.599 Zoran Selinger: actually filter, so it’s not just, okay, this event happened there, when there is GCLID that it exists, but also it has to come from this and this account. Nothing will… nothing different will happen, it’s just the number of errors will be, like.
99 00:13:01.600 ⇒ 00:13:02.090 acromie: Yeah.
100 00:13:02.090 ⇒ 00:13:14.470 Zoran Selinger: We will… we will remove the errors from… Yeah, it seems very inefficient. Yeah, yeah, it is. I mean, this is something… I mean, it’s probably going against some kind of limit.
101 00:13:15.020 ⇒ 00:13:15.970 Zoran Selinger: Right, so….
102 00:13:15.970 ⇒ 00:13:17.759 acromie: Right, that would, yeah, be other worry.
103 00:13:17.760 ⇒ 00:13:25.469 Zoran Selinger: We can. But I don’t think that the volume here is getting us any close to there, but who knows, maybe in the future.
104 00:13:26.220 ⇒ 00:13:34.529 Zoran Selinger: So that is… that’s, like, this solution that we have right now has some… some problems.
105 00:13:35.580 ⇒ 00:13:37.740 Zoran Selinger: In the… in the longer term.
106 00:13:37.850 ⇒ 00:13:41.930 Zoran Selinger: We can do this… We can do this on…
107 00:13:43.120 ⇒ 00:13:45.330 Zoran Selinger: We can have a custom solution for this.
108 00:13:45.840 ⇒ 00:13:46.500 acromie: Okay.
109 00:13:46.500 ⇒ 00:13:49.270 Zoran Selinger: We can ingest amplitude.
110 00:13:49.450 ⇒ 00:13:53.749 Zoran Selinger: events, and based on that, do… we will run
111 00:13:53.960 ⇒ 00:13:58.430 Zoran Selinger: Our workload on the backend that will do this.
112 00:14:00.430 ⇒ 00:14:03.610 Zoran Selinger: That will do this much more easily, right?
113 00:14:04.320 ⇒ 00:14:07.789 Zoran Selinger: We can, you know, that workflow can….
114 00:14:07.980 ⇒ 00:14:16.330 acromie: if I get a click ID, I can have a lookup on the account, for example. Got it. I don’t have to have many, many different things in there.
115 00:14:16.330 ⇒ 00:14:18.220 Zoran Selinger: And …
116 00:14:18.530 ⇒ 00:14:27.619 Zoran Selinger: I mean, uploading clicks for… through the API, that, like I said before, this is a very simple implementation.
117 00:14:27.990 ⇒ 00:14:33.830 Zoran Selinger: two pieces of data, and that’s the event time and the company itself, and we’re good.
118 00:14:33.980 ⇒ 00:14:52.530 Zoran Selinger: So, that’s a short workflow, but we can do it very tidily and quickly in the backend. You just have to figure out how to ingest the… how to get the data from Apple. I would prefer to have it in real time, basically.
119 00:14:52.860 ⇒ 00:15:08.139 Zoran Selinger: When the event happens, we… we get that… we get a single, event on the worker, and it gets processed. But maybe, if not, we… we do… we do pull… batch pulls every, I don’t know.
120 00:15:08.430 ⇒ 00:15:09.280 Zoran Selinger: even…
121 00:15:10.460 ⇒ 00:15:20.479 Zoran Selinger: two times a day, or every 5 minutes, I don’t know what, what makes sense at that point. So, that’s kind of longer-term solution if… Yep.
122 00:15:20.950 ⇒ 00:15:27.260 Zoran Selinger: But if you say that we have, we have a contact from the product team.
123 00:15:27.740 ⇒ 00:15:29.670 Zoran Selinger: have that conversation first. Maybe we don’t have.
124 00:15:29.670 ⇒ 00:15:30.310 acromie: Yeah.
125 00:15:30.310 ⇒ 00:15:30.910 Zoran Selinger: ….
126 00:15:30.910 ⇒ 00:15:43.670 acromie: Right, right. Yep, and if we go down that route, yeah, I would just want to know, like, what does that look like? What, you know, what would we need? Is it… would we need another system to do that work, or where would it live? That kind of stuff, so…
127 00:15:43.670 ⇒ 00:15:59.229 acromie: Yeah, I’m open to going down that route if that’s where we need to go in the future. We’ll just have to get it, obviously, in our prioritization and budget and all that stuff, but, yeah, no, for sure, we’ll start with the product team and see if we can’t solve it natively first, but….
128 00:15:59.730 ⇒ 00:16:02.399 Zoran Selinger: And in the meantime,
129 00:16:02.700 ⇒ 00:16:11.650 Zoran Selinger: So, do you want us to enable this for a few accounts, conversion actions? Is that what you want us to do? Can you…
130 00:16:11.960 ⇒ 00:16:13.760 Zoran Selinger: Make a list, and….
131 00:16:13.760 ⇒ 00:16:31.780 acromie: Yeah, we can do that. I will see… there’s a couple that we already have access to, and I’ll… I’ll double check and see if they’re currently doing anything, or I’ll solicit a couple people who are interested in testing. And then I think my goal would be, out of that, to have some sort of, just
132 00:16:32.420 ⇒ 00:16:38.869 acromie: A good set of instructions, even if they’re real simple, for our teams, that we can continue moving forward with it.
133 00:16:38.870 ⇒ 00:16:41.669 Zoran Selinger: All right, I can do that. I can do that.
134 00:16:41.670 ⇒ 00:16:43.080 acromie: Awesome. Great.
135 00:16:46.800 ⇒ 00:16:53.880 Zoran Selinger: I think that that’s everything from my side, and if there’s nothing else, we’ll… we’ll… Sure, we’ll continue.
136 00:16:54.670 ⇒ 00:16:55.180 acromie: Simple.
137 00:16:55.180 ⇒ 00:16:57.070 Zoran Selinger: This sort of part, yeah. Okay.
138 00:16:57.420 ⇒ 00:17:11.660 Shreya Chowdhury: Yeah, cool. First of all, hi, nice to meet you guys. I just onboarded this week, so I’ve spent the last two days kind of, catching up on stuff and digging into,
139 00:17:12.329 ⇒ 00:17:28.480 Shreya Chowdhury: the… the data parts of Ellie, and yeah, so I’ve been working on, getting a data tracking plan set up, so I can kind of go through, what I have so far. Feel free to jump in with any
140 00:17:28.590 ⇒ 00:17:39.779 Shreya Chowdhury: questions, or if anything looks off, like, I went through it, like, I was kind of just clicking through the website, looking at the different events and the different, like, trying to understand the business processes, but…
141 00:17:40.140 ⇒ 00:17:51.719 Shreya Chowdhury: I can also share these, and you guys can go through it, and see if there’s anything, that stands out. And… yeah.
142 00:17:52.520 ⇒ 00:18:03.830 Shreya Chowdhury: Basically, there’s a couple of, different flows that we want to map out. Eventually, …
143 00:18:03.830 ⇒ 00:18:24.320 Shreya Chowdhury: we want to have, like, the purchase flow and the customer events flow mapped out, so I have a few different… like, I segmented it into, like, pre-purchase, purchase, post-purchase, and eventually, like, we also ideally would want to be able to track, like, customer retention, onboarding, how the sessions went, etc.
144 00:18:24.540 ⇒ 00:18:34.430 Shreya Chowdhury: for now, I’m kind of just gonna focus on the purchase flow, and then, eventually we can reiterate and polish up everything on the customer side. …
145 00:18:34.650 ⇒ 00:18:47.070 Shreya Chowdhury: So, the way that I mapped these out was I gave every event, like, a step in the funnel so that we’d be able to track it, and then, I listed, like, what the event was and what
146 00:18:47.070 ⇒ 00:18:58.540 Shreya Chowdhury: action it would be triggered by, and then the properties are event scope, so that’s just what, on the data side, how it translates, so what it would look like in the schema of the data model. …
147 00:18:58.830 ⇒ 00:19:14.800 Shreya Chowdhury: So, I’ll go through these in, like, the high level. So, for events, like, there’s a lot of homepage visits, like, where you find Click My Location, there’s also an option for online there. Then when they click on the request appointment.
148 00:19:14.800 ⇒ 00:19:26.660 Shreya Chowdhury: button, then that, triggers, like, the request form CTA click, and then how we would translate something like that is, like, we can have a button ID, or a page location, or something like that.
149 00:19:26.690 ⇒ 00:19:31.370 Shreya Chowdhury: … Yeah, so… This is…
150 00:19:31.530 ⇒ 00:19:50.859 Shreya Chowdhury: kind of all the events that I chose to include here. I tried to walk a good line between being, like, succinct and also thorough. I don’t want to over-index on every single thing you could click on the page, but, just to capture most of the important events at a high level. …
151 00:19:51.330 ⇒ 00:19:59.980 Shreya Chowdhury: And then, for the purchase flow, I mainly focused on, … How a user would…
152 00:20:00.100 ⇒ 00:20:14.809 Shreya Chowdhury: create an appointment, so they’ll select, like, their date and time, provider, go through their insurance information. If that all checks out, then either they’ll continue, or they’ll abandon the checkout, whichever.
153 00:20:14.810 ⇒ 00:20:21.520 Shreya Chowdhury: And then, if they don’t, then the appointment will be confirmed, and then they’ll get, like, a confirmation email or a text.
154 00:20:21.900 ⇒ 00:20:22.970 Shreya Chowdhury: ….
155 00:20:23.350 ⇒ 00:20:32.790 acromie: And our… our flow is a slightly different from a normal checkout flow. You know, what I… so we might want to make a couple adjustments here to…
156 00:20:34.130 ⇒ 00:20:36.820 acromie: to connect, but yeah, I mean, I’m following.
157 00:20:37.090 ⇒ 00:20:54.449 Shreya Chowdhury: Okay, yeah, sounds good, and feel free to drop, any of the tweaks that would be useful for me to make. I’m also working on, like, a flow chart slash diagram that I’ll present right after, and then we can, polish that up so we know how we want to do the tracking and the analytics there.
158 00:20:54.450 ⇒ 00:21:12.000 Shreya Chowdhury: But yeah, and then for post-purchase, I figured that’s everything that happens after you’ve already confirmed that you’re gonna have a session, so that’ll be, like, you get the reminder notifications, like, if they open the reminder, if there’s a way we can track that, if they click on, like, …
159 00:21:12.290 ⇒ 00:21:24.179 Shreya Chowdhury: The reminder button, if they reschedule or cancel the appointment, and then after that, if they follow through with the appointment, and then whatever follow-up surveys or whatever there is after.
160 00:21:24.180 ⇒ 00:21:46.609 acromie: yeah, so some of this we’re going to be able to do. So, for example, things like, you know, if the appointment is canceled, or they no-show, or they complete, 100%, we’re working on bringing that into the data warehouse so that we can pull that back up to amplitude. The other pieces, though, I don’t know if we have tracking data on. The…
161 00:21:46.870 ⇒ 00:21:54.080 acromie: The EHR we use is a little behind the times technologically.
162 00:21:54.080 ⇒ 00:22:13.260 acromie: And, I mean, they literally just started putting out APIs for some of their services, so, like, in the last year. So, like, I don’t know if they track, like, if the emails have been opened, or when they’ve been sent on the records, but we will 100% go through and ask
163 00:22:13.620 ⇒ 00:22:18.910 acromie: Our internal data folks to see if they have any insights into that.
164 00:22:18.910 ⇒ 00:22:33.690 Shreya Chowdhury: Yeah, and then we can also assess what the complexity there is, or, like, the priority if we want to have that. These are just… like, I tried to create as thorough of a funnel as I could, but we can also some of these fields, if we don’t need them.
165 00:22:34.580 ⇒ 00:22:37.770 Shreya Chowdhury: Yeah, and then, …
166 00:22:38.550 ⇒ 00:22:48.599 Shreya Chowdhury: I kind of mapped out each one for, like, a tracking plan. This is just, like, a little bit more, like, goes a step further, and so, like, for each, like.
167 00:22:48.600 ⇒ 00:23:04.339 Shreya Chowdhury: field in the schema, like, what data type it’s gonna be, the property definition, like, some sample values. Right now, I… I didn’t have a great idea of, like, which ones are implemented, so I just went through and put no for all of them. Robert did share a list with me for
168 00:23:04.340 ⇒ 00:23:09.840 Shreya Chowdhury: some of which ones have been implemented, I just haven’t had a chance to fully cross-reference all of them yet, but …
169 00:23:09.840 ⇒ 00:23:23.159 Shreya Chowdhury: I can… I think maybe a first step would be to, like, once we get aligned on all of the steps that we want to include and can add tracking for, then we can go through and see which ones have been implemented or can be. …
170 00:23:24.680 ⇒ 00:23:32.519 Shreya Chowdhury: Yeah, and then other than that, let me… I have one more tab that I can share.
171 00:23:32.670 ⇒ 00:23:43.640 Shreya Chowdhury: … It’s… Let me see… Sorry, I just have so many links open.
172 00:23:45.700 ⇒ 00:23:47.939 Shreya Chowdhury: Are you guys able to see this?
173 00:23:48.310 ⇒ 00:23:56.140 Shreya Chowdhury: Yeah. Okay, yeah. So basically, this… it’s pretty similar to the tracking sheet, but it’s just, like, a little bit more, ….
174 00:23:57.170 ⇒ 00:23:57.650 acromie: visual.
175 00:23:57.650 ⇒ 00:23:59.649 Shreya Chowdhury: To look at visually, and then.
176 00:23:59.650 ⇒ 00:24:02.789 acromie: Good, I’m a visual person, this is helpful.
177 00:24:03.310 ⇒ 00:24:19.060 Shreya Chowdhury: Yeah, so I just added, like, the funnel steps, the events, what they’re triggered by, and the properties, and then I’m planning on just breaking this down into, like, a separate table so that we can just have an idea of, like, exactly what each data model would look like, so if we’re trying to look at it in amplitude, it’s, like.
178 00:24:19.090 ⇒ 00:24:32.870 Shreya Chowdhury: you would literally be able to select those specific fields, or, like, calculate whatever sample conversions. I added, in the table here, too, I think for these, I also added,
179 00:24:33.510 ⇒ 00:24:48.980 Shreya Chowdhury: using my best judgment, like, what KPIs we’d want to track for each of these metrics here, and that’s, also subject to change. That’s mostly, like, your guys’ prerogative, but my suggestions, for the KPIs are there.
180 00:24:49.100 ⇒ 00:25:00.479 Shreya Chowdhury: So, yeah, I’m planning on iterating on this, on all three of the flows, and then I can just put, like, a wireframe visual for, like, what the funnel would look like and how we’d want it to look.
181 00:25:00.480 ⇒ 00:25:20.120 acromie: Awesome. Yeah, I think the biggest change to the… to the funnels is that we don’t have a payment structure in our funnel at all, because it’s all done through insurance, and I mean, that’s like a months-long thing, like, and so for us, it’s really about getting into that scheduling an appointment through that online scheduling tool.
182 00:25:20.200 ⇒ 00:25:20.820 acromie: Okay.
183 00:25:21.150 ⇒ 00:25:25.540 acromie: And then afterwards, did they attend, or did they cancel, etc.
184 00:25:25.800 ⇒ 00:25:43.339 Shreya Chowdhury: Okay, sounds good. So yeah, then I think we can probably, like, if it’s too difficult to get, or we don’t want to, we can probably eliminate some of the payment fields that I added. Do we have a way to track, like, if, the payment was completed successfully, or is that also….
185 00:25:45.450 ⇒ 00:25:57.090 acromie: Oof, duh. … I don’t know. I would have to go back and look at the tables to see if that data’s in there. I’ve never really paid attention to it, just because it’s not something that, from a corporate standpoint, like…
186 00:25:57.200 ⇒ 00:26:03.739 acromie: We are tracking in our department, and a lot of that happens on…
187 00:26:04.060 ⇒ 00:26:12.910 acromie: the, the location side, so I’m not sure how much of that is recorded in the EHR.
188 00:26:13.080 ⇒ 00:26:20.290 acromie: So, definitely something to… poke around, and Adam, do you want to look through that, that, individual…
189 00:26:20.440 ⇒ 00:26:23.229 acromie: Appointment roster data, or whatever it is.
190 00:26:23.700 ⇒ 00:26:24.689 Adam Kittleson: Yeah, I see….
191 00:26:24.690 ⇒ 00:26:25.850 acromie: around that.
192 00:26:26.070 ⇒ 00:26:34.030 Adam Kittleson: Yeah, I’m sure stuff like that is captured somewhere. Like, there’s a whole tab on… in the data warehouse for billing.
193 00:26:34.030 ⇒ 00:26:34.969 acromie: Yeah. Of course.
194 00:26:35.350 ⇒ 00:26:37.720 Adam Kittleson: But yeah, I could take a look.
195 00:26:38.150 ⇒ 00:26:42.179 Shreya Chowdhury: Yeah, that one just… just more so for, like.
196 00:26:43.000 ⇒ 00:26:53.309 Shreya Chowdhury: like, if, for whatever reason, we wanted to… I’m thinking long-term, like, just for analytics, like, if we want to take a look at the one… the appointments that were paid for, but then no shows,
197 00:26:53.310 ⇒ 00:26:54.710 acromie: Yeah, right.
198 00:26:55.880 ⇒ 00:26:58.730 Shreya Chowdhury: But… Yeah.
199 00:27:00.450 ⇒ 00:27:00.940 Shreya Chowdhury: ….
200 00:27:00.940 ⇒ 00:27:06.730 acromie: Yeah, they only… they pay after the appointment. That’s the other thing. It’s not like they pay and then…
201 00:27:07.310 ⇒ 00:27:16.000 acromie: And then go to the appointment, or don’t show up, or show up. It’s more of a… they go, and then it gets billed after the fact, so….
202 00:27:16.630 ⇒ 00:27:18.769 Shreya Chowdhury: Right, but is there not, …
203 00:27:20.020 ⇒ 00:27:28.390 Shreya Chowdhury: Okay, I’ll take another look, but I… you still have to, update the payment information prior to the appointment, is that correct? Or, like, the insurance information?
204 00:27:28.390 ⇒ 00:27:36.569 acromie: Well, they do collect insurance information, yes. From the standpoint of…
205 00:27:36.820 ⇒ 00:27:41.049 acromie: Like, the systems where they’re scheduling through?
206 00:27:42.180 ⇒ 00:27:46.079 acromie: No, you’re right. I mean, that would all be in a valent, so that would be in the EHR.
207 00:27:46.180 ⇒ 00:27:50.640 acromie: if we have their insurance information. But yeah, whether or not the…
208 00:27:50.790 ⇒ 00:27:58.410 acromie: you know, I don’t know if there’s anything in there about if the insurance has been verified, Or… ….
209 00:27:58.790 ⇒ 00:27:59.310 Shreya Chowdhury: Yeah, just because.
210 00:27:59.310 ⇒ 00:28:00.350 acromie: Once, yeah.
211 00:28:00.680 ⇒ 00:28:06.879 Shreya Chowdhury: Yeah, because I think when they’re connecting you to a provider, they do the insurance check first, prior.
212 00:28:06.880 ⇒ 00:28:31.370 acromie: Yeah, and so right now, there’s a bit of a disconnect in the online scheduling system on that. We just collect it, and then they do the checks and make sure everything was good, because most cases it is, and then they reach out. But that’s something we are building into the system. So that will be there, that insurance check. In fact, we have an event in Helpfully for insurance check, it’s just not coming over because it hasn’t been…
213 00:28:31.480 ⇒ 00:28:36.029 acromie: activated yet. We’re not actually gonna get anything. So….
214 00:28:38.790 ⇒ 00:28:46.500 Shreya Chowdhury: Okay, yeah, … Yeah, well, that’s kind of what I had from a high-level.
215 00:28:46.510 ⇒ 00:28:47.430 acromie: Yep.
216 00:28:47.430 ⇒ 00:28:53.479 Shreya Chowdhury: View of the tracking planning, and then, obviously, there’s a lot of adjustments to be made there, but, …
217 00:28:54.600 ⇒ 00:28:56.430 Shreya Chowdhury: Yeah, of course.
218 00:28:57.670 ⇒ 00:28:58.580 Shreya Chowdhury: That was pretty much it.
219 00:28:58.580 ⇒ 00:29:15.770 acromie: we just need to adjust some of the, like, maybe rethink how we’re labeling things, just to make them make more sense for our use case. But, yeah, no, that looks good. Do you need anything else from us to fill in or finish any of those sections?
220 00:29:15.770 ⇒ 00:29:24.369 Shreya Chowdhury: I think… maybe I can, send over the sheet, and then if…
221 00:29:24.470 ⇒ 00:29:35.940 Shreya Chowdhury: you guys have bandwidth, you can look through and just, like, X out whichever ones don’t make sense to you, or, like, we don’t have those, or we’re not able to get those fields, whatever. For sure. Yeah, because I think it might be…
222 00:29:36.040 ⇒ 00:29:45.590 Shreya Chowdhury: better to do that before, like, visually drawing out the funnel, just so I don’t end up including a bunch of extra stuff to take it out later.
223 00:29:45.840 ⇒ 00:29:46.869 acromie: Yeah, I agree.
224 00:29:46.870 ⇒ 00:29:54.300 Shreya Chowdhury: So yeah, I can probably send that over, and then once we have the spreadsheet polished up, I can finish the wireframe.
225 00:29:54.630 ⇒ 00:29:55.779 acromie: Cool, great.
226 00:30:00.480 ⇒ 00:30:01.580 acromie: Alright.
227 00:30:07.250 ⇒ 00:30:14.600 acromie: So I think… I think we’re all good from here, then. Adam, do you have any questions? I think just being able to get into Google is gonna be great.
228 00:30:15.300 ⇒ 00:30:16.030 Adam Kittleson: Yeah.
229 00:30:16.220 ⇒ 00:30:25.399 acromie: What about, as we kind of finish off this Google Ads section, have you started thinking at all about the Google Analytics?
230 00:30:25.670 ⇒ 00:30:28.330 acromie: the GA4 connection at all yet?
231 00:30:29.250 ⇒ 00:30:32.700 acromie: That was the other one that’s being wonky on us.
232 00:30:32.880 ⇒ 00:30:37.429 Zoran Selinger: Yeah, yeah, yeah, so we, we are, we are discussing that.
233 00:30:37.430 ⇒ 00:30:38.240 acromie: Okay, cool.
234 00:30:38.240 ⇒ 00:30:44.830 Zoran Selinger: Yeah, we are discussing that. Generally speaking, still kind of… What’s, …
235 00:30:45.090 ⇒ 00:30:51.680 Zoran Selinger: What we’re not sure about is, do we… do we need it? Do we actually need it?
236 00:30:51.890 ⇒ 00:30:52.370 Zoran Selinger: ….
237 00:30:52.370 ⇒ 00:30:53.000 acromie: Right.
238 00:30:53.000 ⇒ 00:30:54.250 Zoran Selinger: Yeah, ….
239 00:30:54.250 ⇒ 00:30:56.769 acromie: Yep, nope, that’s 100% a valid question.
240 00:30:57.310 ⇒ 00:31:07.719 acromie: And we have some things happening internally that may provide a lot more insights into that coming up here. So, …
241 00:31:07.910 ⇒ 00:31:08.770 acromie: Yeah.
242 00:31:09.520 ⇒ 00:31:12.109 acromie: Yeah, because we really only have…
243 00:31:12.830 ⇒ 00:31:23.029 acromie: a couple places that are relying on some sort of Google Analytics connection, and there’s chances that we might just not have that anymore, so….
244 00:31:23.030 ⇒ 00:31:24.200 Zoran Selinger: Yeah, okay, okay.
245 00:31:24.200 ⇒ 00:31:27.539 acromie: If that’s the case, then I don’t want to spend time on it either, so…
246 00:31:28.040 ⇒ 00:31:34.909 acromie: We’ll, we’ll do some… some reconnaissance there, and figure that out, that answer here for you as well.
247 00:31:34.910 ⇒ 00:31:36.769 Zoran Selinger: Yeah, okay, good, good.
248 00:31:36.770 ⇒ 00:31:37.460 acromie: Awesome.
249 00:31:38.680 ⇒ 00:31:39.610 acromie: Okay!
250 00:31:41.070 ⇒ 00:31:49.960 acromie: I’m assuming that’s everything, then, for today, and we’ll… I will get you a couple of accounts that we can connect in, and then…
251 00:31:50.440 ⇒ 00:31:56.819 acromie: And is the best way to get those to you just to add your user to them?
252 00:31:58.710 ⇒ 00:31:59.220 acromie: Sure.
253 00:31:59.220 ⇒ 00:32:03.119 Zoran Selinger: I mean, if the account is under the MCC, I can do everything.
254 00:32:03.120 ⇒ 00:32:08.280 acromie: Okay, so I will get them into the MCC, and then let you know which ones, is what I will….
255 00:32:08.280 ⇒ 00:32:14.739 Zoran Selinger: Yes, and obviously, which events and all that. That’s important. …
256 00:32:15.420 ⇒ 00:32:26.799 Zoran Selinger: In the meantime, tomorrow when you wake up, I’m gonna have the standard operating procedure for creating those yourself. I will write that up for you.
257 00:32:27.300 ⇒ 00:32:33.759 Zoran Selinger: the steps and screenshots and everything else. Awesome, great. It’s really clear. I’ll create a document for you. Perfect.
258 00:32:33.760 ⇒ 00:32:34.430 acromie: Thank you so much!
259 00:32:34.430 ⇒ 00:32:35.330 Zoran Selinger: Yeah, yeah.
260 00:32:35.560 ⇒ 00:32:36.230 acromie: Great.
261 00:32:36.610 ⇒ 00:32:39.730 acromie: All right, well, thanks everybody, and excited to.
262 00:32:40.380 ⇒ 00:32:40.730 Adam Kittleson: Yup.
263 00:32:42.320 ⇒ 00:32:44.489 acromie: Yeah. Alright, we’ll talk soon.
264 00:32:45.600 ⇒ 00:32:47.100 Zoran Selinger: Have a good one, too. Bye.