Meeting Title: Walkthrough of tables for engagement report Date: 2026-02-04 Meeting participants: Awaish Kumar, Katherine Bayless, Kyle Wandel
WEBVTT
1 00:01:59.150 ⇒ 00:01:59.780 Awaish Kumar: Hello?
2 00:01:59.980 ⇒ 00:02:00.760 Katherine Bayless: How’s it going?
3 00:02:01.730 ⇒ 00:02:03.429 Awaish Kumar: It’s going well. How about you?
4 00:02:03.800 ⇒ 00:02:06.380 Katherine Bayless: Pretty good, yeah. Busy, but good.
5 00:02:06.550 ⇒ 00:02:08.979 Awaish Kumar: Okay, yeah, busy is good.
6 00:02:09.199 ⇒ 00:02:27.339 Katherine Bayless: Yeah, exactly. Yeah. We were just, actually, I’ll show you on the wall behind me. So, some of this data will get used for, I guess it’s kind of blurry, but it’ll get used for journeys and Salesforce Marketing Cloud, and so I was sketching out what the data extensions need to look like in Marketing Cloud.
7 00:02:27.640 ⇒ 00:02:28.940 Awaish Kumar: Oh, yeah, okay.
8 00:02:29.420 ⇒ 00:02:35.670 Awaish Kumar: Like, what is it going to look like in Marketing Cloud? What is that, like…
9 00:02:36.510 ⇒ 00:02:40.039 Awaish Kumar: You are looking to reverse ETL data there?
10 00:02:40.590 ⇒ 00:03:00.249 Katherine Bayless: Yeah, so we’re waiting for the Polyatomic connector to come through on the AWS Marketplace, but then they’re putting together a connector for Marketing Cloud for us, and so right now we’re doing this kind of manually, but we’ve been focused on, like, basic contact details and CES details so far, but now that we have the membership stuff.
11 00:03:00.250 ⇒ 00:03:05.209 Katherine Bayless: we can start pushing membership data into Marketing Cloud for those journeys.
12 00:03:05.210 ⇒ 00:03:05.770 Awaish Kumar: Okay.
13 00:03:05.920 ⇒ 00:03:16.230 Awaish Kumar: And so it’ll… it’ll be a different, probably, like, shape than, like, what we were doing for the engagement report, but it’ll use a lot of the same, underlying, tables, so… Okay, okay.
14 00:03:16.990 ⇒ 00:03:20.019 Awaish Kumar: Yeah. Are we waiting for anyone, or…
15 00:03:20.550 ⇒ 00:03:24.069 Katherine Bayless: Oh yeah, let me see if Kyle, can join us.
16 00:03:26.010 ⇒ 00:03:28.420 Katherine Bayless: And I wasn’t sure if Utan was joining…
17 00:03:28.420 ⇒ 00:03:34.400 Awaish Kumar: No, he might not be joining. I will just walk through with what we have done so far.
18 00:03:46.080 ⇒ 00:03:50.260 Awaish Kumar: So, this thing, like, PDF report that we…
19 00:03:50.410 ⇒ 00:03:53.279 Awaish Kumar: We received, for the engagement report.
20 00:03:53.820 ⇒ 00:04:01.109 Awaish Kumar: Like, so this goes out, like, every day, every week, how that works?
21 00:04:01.580 ⇒ 00:04:06.000 Katherine Bayless: So… and I sent a note to Kyle, so he’ll join if he can.
22 00:04:06.000 ⇒ 00:04:30.650 Katherine Bayless: So right now, I think they’ve been generating it, like, on a monthly basis, but that’s mostly because it was so difficult to pull the data together. Now that we have the, you know, data coming through automatically in Snowflake, I think the remembers data refreshes every 4 hours, and so they’ll probably start looking at this just kind of, like, all day, every day. Like, I have a feeling there will be some people that’ll just have the final sort of
23 00:04:30.650 ⇒ 00:04:33.130 Katherine Bayless: Board rebuild, just kind of open all day.
24 00:04:33.870 ⇒ 00:04:39.029 Awaish Kumar: Okay, so it’s going to be in Power BI, and someone will be looking at that.
25 00:04:39.340 ⇒ 00:04:45.399 Katherine Bayless: Oh, I see. No, we’re gonna actually… we’re gonna bring them into Snowflake. Yeah, we’re gonna have them look at the data in Snowflake, yeah.
26 00:04:45.820 ⇒ 00:04:55.030 Awaish Kumar: Okay, like, Like, that means they will be directly connecting to those models.
27 00:04:55.840 ⇒ 00:05:14.440 Katherine Bayless: Yeah, exactly. So I think we’re either gonna do, like, a Snowflake, like, dashboard, which are, you know, limited, but they get the job done, or a Streamlit app around it. But yeah, so it’ll look similar to the way the Power BI report did, but we’re gonna have them get used to just coming straight into Snowflake to look at those dashboards and Streamlit apps.
28 00:05:14.440 ⇒ 00:05:15.160 Awaish Kumar: Okay.
29 00:05:15.430 ⇒ 00:05:17.340 Awaish Kumar: Yeah. Okay, yeah, got it.
30 00:05:17.990 ⇒ 00:05:18.760 Awaish Kumar: I’m gonna… Yeah.
31 00:05:18.760 ⇒ 00:05:22.380 Katherine Bayless: to get rid of Power BI. I just can’t. I just don’t like it.
32 00:05:22.690 ⇒ 00:05:31.579 Awaish Kumar: Yeah, like… I think now that a lot of tools coming out with AI features and…
33 00:05:32.380 ⇒ 00:05:41.360 Awaish Kumar: things like that. Like, all… all… all legacy tools, like Power BI, Tableau, seem to, yeah, like…
34 00:05:42.350 ⇒ 00:05:57.029 Awaish Kumar: Like, they seem like legacy now, that, like, we have been using it, for our clients, but now, in this last one or two years where we see it, like, all the new tools coming in, they look pretty, like, old, outdated.
35 00:05:57.260 ⇒ 00:05:57.980 Katherine Bayless: Right?
36 00:05:58.220 ⇒ 00:05:58.550 Awaish Kumar: Beautiful.
37 00:05:58.550 ⇒ 00:06:01.940 Katherine Bayless: The other thing in my brain, too, is like…
38 00:06:02.120 ⇒ 00:06:15.570 Katherine Bayless: increasingly, I think people are just going to want to, like, query the data in natural language, right? And so it’s like, if they’re already comfortable in Snowflake to just look at dashboards, then once we roll out Cortex and some of those things, like, they’ll be there already. I think we can tie.
39 00:06:15.570 ⇒ 00:06:16.060 Awaish Kumar: Yeah.
40 00:06:16.060 ⇒ 00:06:33.110 Katherine Bayless: Pretty easily, yeah. We’ll probably eventually need a visualization tool for certain things that are, you know, they go to the board, or they’re, like, you know, once a year they get refreshed, but I’m hoping for most of the operational data usage to happen directly in Spufflake.
41 00:06:33.450 ⇒ 00:06:52.899 Awaish Kumar: Yeah, I have already… I tried to use on the data we have in Snowflake, the Cortex Analyst. I’ve tried it a little bit. It’s not that mature. Like, BI tools are actually much more mature at that moment, in terms of natural language processing, but yeah.
42 00:06:53.500 ⇒ 00:06:54.630 Katherine Bayless: I think it’s changing so fast.
43 00:06:54.630 ⇒ 00:06:54.990 Awaish Kumar: Yeah.
44 00:06:55.080 ⇒ 00:06:55.760 Katherine Bayless: Yeah.
45 00:06:55.760 ⇒ 00:06:56.150 Awaish Kumar: Yeah.
46 00:06:56.150 ⇒ 00:07:05.619 Katherine Bayless: sent me this link from just the other day, where Snowflake was announcing, like, major upgrades to, like, Cortex, I forget what it was, like, Cortex mode or something like that, but yeah.
47 00:07:06.700 ⇒ 00:07:26.329 Awaish Kumar: Okay, yeah. Yeah, I will just quickly walk through over what we have worked so far. Like, based on the given PDF, we have tried our best to come up with the models needed, but obviously, once it is in QA and people start to use it, we might get more feedback, and…
48 00:07:26.470 ⇒ 00:07:32.530 Awaish Kumar: We might need to join here and there, but yeah, so far, we have… we have built the base.
49 00:07:32.660 ⇒ 00:07:34.059 Awaish Kumar: For, for, for that report.
50 00:07:34.600 ⇒ 00:07:36.750 Awaish Kumar: I can share my screen.
51 00:07:37.380 ⇒ 00:07:41.370 Awaish Kumar: Oh, basically, let me just… Shit it.
52 00:07:42.080 ⇒ 00:07:49.589 Awaish Kumar: Okay so I will start with this Google Sheet. I have put some documentation in our data platform.
53 00:07:49.760 ⇒ 00:08:00.209 Awaish Kumar: documentation spreadsheet. So, basically, here you can see the table names, where it lives in the snowflake, and what, basically, the data is about.
54 00:08:00.400 ⇒ 00:08:04.260 Awaish Kumar: Obviously, like, we have…
55 00:08:04.460 ⇒ 00:08:13.259 Awaish Kumar: yeah, we have, like, all of these would start with RPT, they are basically the shortfall report, so…
56 00:08:13.610 ⇒ 00:08:28.370 Awaish Kumar: These report tables are basically in the mods section, which means we want everybody to just use these, just interact with these tables, instead of going into other layers of,
57 00:08:28.450 ⇒ 00:08:47.090 Awaish Kumar: Snowflake, but we… but I have documented some of the tables in SDG. Like, we… for the dbt, we have four different layers, the way we have structured it. I’m not sure how much my team already communicated that, but yeah.
58 00:08:48.560 ⇒ 00:09:00.140 Awaish Kumar: we have raw layer, we have SDG layer, and we have intermediate layer, and we have then, finally, the March layer. So, I have listed here the tables we have in SDG, and the…
59 00:09:00.400 ⇒ 00:09:08.729 Awaish Kumar: Then, on top of it, we have built these reports. So, majority of our transformation happens between either
60 00:09:08.770 ⇒ 00:09:28.099 Awaish Kumar: SDG, or in the intermediate layer. These are, like… in the SDG layer, we do some kind of, what you say, light transformation, changing data types, expanding JSON, and then when it goes into the intermediate, we do some more transformation, include business logic, and all of that.
61 00:09:28.290 ⇒ 00:09:38.010 Awaish Kumar: Then finally, once it is in the March… March is also basically the light transformation, which, like, just join these four tables to make… to make this, or things like that.
62 00:09:38.460 ⇒ 00:09:50.349 Awaish Kumar: So this is basically the documentation. Whoever wants to work on this report, they can come here and figure out which table to use. And then, if you look at this,
63 00:09:50.820 ⇒ 00:09:51.870 Awaish Kumar: GitHub.
64 00:09:53.700 ⇒ 00:09:59.589 Awaish Kumar: it’s structured in this pretty same way. In the dbt project models.
65 00:09:59.690 ⇒ 00:10:03.540 Awaish Kumar: March, if you go to the reports, we have all these
66 00:10:04.080 ⇒ 00:10:06.940 Awaish Kumar: tables which I just showcased here.
67 00:10:08.440 ⇒ 00:10:11.610 Awaish Kumar: And here you can see, like, some of it is…
68 00:10:12.030 ⇒ 00:10:28.740 Awaish Kumar: There are some… just some joints to bring it all together in one place, because in the report section, we just do normally that, but if you go to maybe some intermediate layer, there you would see maybe a little bit more, cleanups, and…
69 00:10:29.030 ⇒ 00:10:30.560 Awaish Kumar: transformations.
70 00:10:30.740 ⇒ 00:10:36.150 Awaish Kumar: Like, with the session schedule, like, we are trying to play with this.
71 00:10:37.980 ⇒ 00:10:42.940 Awaish Kumar: So, yeah, so, like, here you can see a little bit of more of a kind of transformation.
72 00:10:43.610 ⇒ 00:10:47.429 Awaish Kumar: And the business logic is embedded here. And…
73 00:10:48.160 ⇒ 00:10:52.950 Awaish Kumar: Once, yeah, at the inter… in the intermediate layer, only the tables.
74 00:10:53.200 ⇒ 00:11:09.189 Awaish Kumar: which require a lot of transformation live here. So, like, there are some tables which might not require, like, very hard transformations. You can just do some light transformation in SDG layer, and bring in directly to the mods, the folder.
75 00:11:09.190 ⇒ 00:11:13.580 Awaish Kumar: So, that’s how this approach works, and
76 00:11:14.360 ⇒ 00:11:32.910 Awaish Kumar: Basically, yeah, but what we want is that, like, everybody, the end users just should use what is… whatever is in the mods. So, and if it is not there, if they are missing some field, if… if you think, there needs to be some log… update of the logic, or…
77 00:11:33.020 ⇒ 00:11:41.270 Awaish Kumar: bring much more table, like, yeah, then we will be supporting them to bring that data into the mods, so they can basically use it.
78 00:11:41.490 ⇒ 00:11:47.459 Awaish Kumar: And… If you… now, if we go to the snowflake, is it visible?
79 00:11:49.020 ⇒ 00:11:57.989 Awaish Kumar: Okay, if you look at this, you can see, again, in the broad marks, reports, and you can see all the tables here.
80 00:11:58.450 ⇒ 00:12:08.180 Awaish Kumar: So our dbt structure is kind of very, similar to how we structure it snowflake.
81 00:12:09.430 ⇒ 00:12:18.179 Awaish Kumar: So whatever, like, a folder becomes a database, like Mark’s is a database, then inside of that, each folder, like, CRM,
82 00:12:18.340 ⇒ 00:12:20.250 Awaish Kumar: becomes,
83 00:12:20.550 ⇒ 00:12:28.599 Awaish Kumar: a schema, and then reports also becomes a schema. If there are multiple other… if we try to include more,
84 00:12:29.020 ⇒ 00:12:47.239 Awaish Kumar: like, the folders there, if we have some sales smart, some finance mod. So we try… we will be including all those here, and you will see separate schema for each of these. So now, we have this reports folder, which is translated to reports schema and stuff like.
85 00:12:47.240 ⇒ 00:12:53.779 Awaish Kumar: And then there are all of these tables. So if we just look at one of the tables, I have opened it here.
86 00:12:54.580 ⇒ 00:12:59.259 Awaish Kumar: Basically, one section of our…
87 00:12:59.400 ⇒ 00:13:05.340 Awaish Kumar: if I just pull up the report. So, one section of this is…
88 00:13:05.920 ⇒ 00:13:08.670 Awaish Kumar: kind of EB bill, right?
89 00:13:10.630 ⇒ 00:13:17.770 Awaish Kumar: And this EV build section, like, if we see, we have a year, we have a first name, last name, title, and a status.
90 00:13:18.010 ⇒ 00:13:22.400 Awaish Kumar: where you can get this data from is if you go to the Snowflake.
91 00:13:22.910 ⇒ 00:13:26.819 Awaish Kumar: And if you go to the table called,
92 00:13:26.950 ⇒ 00:13:40.919 Awaish Kumar: report, member engagement is the name of the report, and then it’s called underscore, underscore EBBL. So, right now, because, because how, like, we have…
93 00:13:41.040 ⇒ 00:13:44.049 Awaish Kumar: Try to create these tables.
94 00:13:44.360 ⇒ 00:13:48.660 Awaish Kumar: Which basically… which can help you create the exact PDF.
95 00:13:49.510 ⇒ 00:14:01.620 Awaish Kumar: So we have took each segment of it, and tried to create a table for it. But, yeah, as we try to QA, and we identify what the new requirements look like, maybe now.
96 00:14:01.620 ⇒ 00:14:12.069 Awaish Kumar: if we are working… somebody’s working on the streamlet, or someone is going to use, Surflet dashboards or whatever, they might need some more joins.
97 00:14:12.070 ⇒ 00:14:23.370 Awaish Kumar: Some more fields, so we can bring that in. So, but right now, like, using that, if I can see, you can see the exact thing. We have a year, we have first name, last name, title, and a status.
98 00:14:23.420 ⇒ 00:14:25.579 Awaish Kumar: Right? First, it is…
99 00:14:25.650 ⇒ 00:14:44.620 Awaish Kumar: there is a… in the raw EB Bill table, there is a field called, it, EB Bill Period Ends, right? Which includes when it is ending, based on that, and based on the file, where… where the data entered, like…
100 00:14:44.730 ⇒ 00:14:50.690 Awaish Kumar: So, the file was for, for example, December 2025. We see these are the
101 00:14:50.930 ⇒ 00:14:58.570 Awaish Kumar: names and titles for this year, and then if you compare it with year-end.
102 00:14:58.680 ⇒ 00:15:06.670 Awaish Kumar: If it is in 2027, that means it’s a selective. So we have tried to come up with some logic, to do that.
103 00:15:06.730 ⇒ 00:15:24.829 Awaish Kumar: that might not be exactly, the way you want it, or maybe you need a few more filters, we can easily, like, bring that in, like, that’s… yeah. Similarly, we have a few more tables here, like count summary, for example, if you preview this.
104 00:15:24.940 ⇒ 00:15:29.180 Awaish Kumar: If I can just preview… Okay.
105 00:15:30.760 ⇒ 00:15:40.049 Awaish Kumar: It says we don’t have a… Basically… default warehouse? I don’t know.
106 00:15:48.800 ⇒ 00:15:49.640 Awaish Kumar: Okay.
107 00:15:53.790 ⇒ 00:15:57.309 Awaish Kumar: I don’t know what’s going on with this,
108 00:15:57.310 ⇒ 00:15:59.719 Katherine Bayless: It could be uploaded fine, yeah.
109 00:15:59.720 ⇒ 00:16:06.079 Awaish Kumar: Yeah, so… Basically, When you are previewing, we have to set the default
110 00:16:06.460 ⇒ 00:16:09.960 Awaish Kumar: Trust, like, the virtual warehouse.
111 00:16:10.250 ⇒ 00:16:21.180 Awaish Kumar: And I… although I have selected it here, that I want to use account admin role and a warehouse transform, but maybe it’s not… it’s not loading it yet, but I can just query…
112 00:16:21.430 ⇒ 00:16:23.080 Awaish Kumar: This table here.
113 00:16:47.450 ⇒ 00:16:54.819 Awaish Kumar: Yeah, so you can see we have organization ID, record number, name.
114 00:16:55.350 ⇒ 00:17:00.140 Awaish Kumar: And all of the information which you see at the top level.
115 00:17:02.180 ⇒ 00:17:05.829 Awaish Kumar: Which should represent here, right? So…
116 00:17:06.280 ⇒ 00:17:10.220 Awaish Kumar: Like, if there are numbers, so this table is a, you know, kind of a…
117 00:17:10.359 ⇒ 00:17:18.939 Awaish Kumar: at a, like, person… at a granularity of a person, like, in an account, if you see, for example, it says…
118 00:17:20.520 ⇒ 00:17:27.819 Awaish Kumar: Like, dues level, or number of valid contacts, so that does just mean that we have to group by this.
119 00:17:28.630 ⇒ 00:17:38.669 Awaish Kumar: So, in this table, we have organization ID, we have names, so we want to keep our models at some granular level, so we can basically aggregate it.
120 00:17:39.060 ⇒ 00:17:50.729 Awaish Kumar: Then this can be done. Normally, it has to be done in a… in a, like, the final BI tool, or whatever we want to use, Streamlit, or whatever.
121 00:17:50.840 ⇒ 00:17:52.850 Awaish Kumar: Right? So…
122 00:17:52.960 ⇒ 00:18:00.469 Awaish Kumar: Yeah, so this is how we have created these tables. If you… if you want to look at one more here.
123 00:18:00.700 ⇒ 00:18:04.880 Awaish Kumar: Which is gone… I can copy the name.
124 00:18:06.100 ⇒ 00:18:07.450 Awaish Kumar: I put it here.
125 00:18:07.790 ⇒ 00:18:20.009 Awaish Kumar: Yeah, and like, if you see here, if I give an example, like, I just have an organizational ID here right now, in this table, right? If we need more data for an organization, it can be brought in, right?
126 00:18:21.300 ⇒ 00:18:25.240 Awaish Kumar: For an organization, we need name, we need…
127 00:18:25.370 ⇒ 00:18:36.240 Awaish Kumar: maybe there is some description, whatever. If you need a few more columns, you can bring that in, but it is, like, in the… looking at the report, whatever…
128 00:18:36.600 ⇒ 00:18:44.760 Awaish Kumar: how we thought could be… could make sense. We have to just bring… bring that in here, because, we have here, if you look at the…
129 00:18:46.680 ⇒ 00:18:52.379 Awaish Kumar: also the CRM, we have a DIM organization. So, normally, we don’t do…
130 00:18:52.810 ⇒ 00:19:08.719 Awaish Kumar: joins, right? We keep it at, like, organization ID level, and we have a dev organization also here, so you can join both these tables to come up with names. And normally, we… the data analyst does that in a PI tool.
131 00:19:08.890 ⇒ 00:19:21.020 Awaish Kumar: Right? But we also can help, right? If you want, okay, okay, I don’t want to join these in a BI tool, let me create a summary model for me, which basically joins this and brings all these fields. We can do that.
132 00:19:21.180 ⇒ 00:19:39.060 Awaish Kumar: But yeah, that happens when you QA, you look at it, okay, now somebody says, okay, now stream it when I try to join this, yeah, it becomes really slow, it doesn’t work for me, create a, like, table for me in Snowflake. So, we will just create a summary table which will join that.
133 00:19:40.460 ⇒ 00:19:41.260 Katherine Bayless: That makes sense.
134 00:19:41.380 ⇒ 00:19:45.209 Kyle Wandel: And that’s how they did it previously, so, that would make sense.
135 00:19:45.840 ⇒ 00:19:46.460 Katherine Bayless: Yeah.
136 00:19:50.030 ⇒ 00:20:05.710 Awaish Kumar: And I have an example here for exactly that thing. So, in this, we have actually went ahead to do that, to show that, like, basically it is doable. So, we have an organization ID, we have a year, registration type.
137 00:20:05.850 ⇒ 00:20:20.160 Awaish Kumar: And we have events status. So using this table, basically, you can get the registration and attendance, grouped by per organization, right? And by year.
138 00:20:20.580 ⇒ 00:20:34.160 Awaish Kumar: And, for an organization ID, normally you just join, as I explained earlier, but if you want, we can join. I try to join it here. I have… I brought in name, and the branch name, and created by.
139 00:20:34.680 ⇒ 00:20:36.629 Awaish Kumar: To this table.
140 00:20:37.120 ⇒ 00:20:43.950 Awaish Kumar: As well. So, basically, these are the name of the… organizations.
141 00:20:45.400 ⇒ 00:20:49.280 Kyle Wandel: We’ll definitely need to update it based on our own logic, Catherine, but I mean, this is…
142 00:20:49.400 ⇒ 00:20:53.770 Kyle Wandel: What we’re… what we’re looking for, in terms of… the structure.
143 00:20:54.690 ⇒ 00:21:11.270 Katherine Bayless: Yeah, yeah, totally. I mean, I think, yeah, like, exactly what you guys are saying, right? Like, we want to, as much as possible, stick to just kind of creating the, you know, the primitives that we then can, yeah, join downstream, but I think for, yeah, for these purposes, it does make sense to have brought some of it all together,
144 00:21:12.770 ⇒ 00:21:21.499 Katherine Bayless: So, curious, so for, like, for the EBBIL, how,
145 00:21:22.260 ⇒ 00:21:35.210 Katherine Bayless: So in the member engagement report, right, so, like, if we build the Streamlit app, and we’re filtering it to Samsung, how will we associate the EBBIL contacts to Samsung?
146 00:21:38.080 ⇒ 00:21:44.530 Awaish Kumar: Okay, yeah, it’s… okay. Right now, I think it does not have an organization ID,
147 00:21:44.700 ⇒ 00:21:55.389 Awaish Kumar: Right now, here. But we, yeah, we have this IMPEX organization ID, which we’ll just bring in here, so you can basically… if you look at the.
148 00:21:55.640 ⇒ 00:21:58.150 Kyle Wandel: Demo organization table.
149 00:21:58.760 ⇒ 00:22:06.140 Awaish Kumar: So there are two IDs. One is called Organization Acts ID, the other one is called Record Number, and any one of them
150 00:22:07.020 ⇒ 00:22:11.029 Awaish Kumar: can be used to identify the organization. So, basically.
151 00:22:11.060 ⇒ 00:22:28.400 Awaish Kumar: In this table, registration and utterance, I’m using this record number to basically tie it to the name organization and bring that actual, like, join, alright? So, basically, we just have to put this number into the EBPIL table.
152 00:22:28.530 ⇒ 00:22:29.640 Awaish Kumar: And that’s all.
153 00:22:30.470 ⇒ 00:22:41.779 Katherine Bayless: Yeah, that’ll be the challenge, yeah. So this is, I mean, this is where, like, we run into the tricks, with the… and why the entity resolution stuff is gonna be so clutch is, like.
154 00:22:41.780 ⇒ 00:22:53.759 Katherine Bayless: the EBBAL data, it doesn’t… it doesn’t have record number in it by default, and then the company names that are in that data may or may not match exactly to what’s in the remembers, data.
155 00:22:53.820 ⇒ 00:23:02.130 Katherine Bayless: The same with the CES stuff, like, I think the CES reports that we’ve been using, they do have a column for member ID, but it’s…
156 00:23:02.900 ⇒ 00:23:03.610 Awaish Kumar: Yeah, we…
157 00:23:03.610 ⇒ 00:23:11.640 Katherine Bayless: That’s the full picture of the members that need, or the contacts that need to get associated to the member, and that kind of thing, so… yeah. Yeah.
158 00:23:11.640 ⇒ 00:23:24.860 Awaish Kumar: Yeah, yeah, I will take that feedback, and we are going to work on that identity stitching. Like, we have been, like, internally, we have been talking about this, identity stitching thing, but,
159 00:23:25.020 ⇒ 00:23:28.659 Awaish Kumar: I think our, like, the…
160 00:23:28.910 ⇒ 00:23:46.909 Awaish Kumar: like, the colleague, my colleague, Aishwani, is basically, working on that, so he will, like, kind of, like, we either review organization name, or contact name, or something to basically tie it with some… with some organization, and then we are going to share, like.
161 00:23:46.920 ⇒ 00:23:53.009 Awaish Kumar: how successful we were, like, if we could match 80% or 70% or whatever.
162 00:23:53.120 ⇒ 00:23:54.419 Awaish Kumar: That we will bring in.
163 00:23:55.760 ⇒ 00:23:59.290 Katherine Bayless: Yeah. Yes, I think, I guess, for…
164 00:23:59.650 ⇒ 00:24:15.939 Katherine Bayless: for the moment, for the EBBIL stuff, I mean, honestly, it’s high stakes enough, it might make the most sense for one of us, Kyle, to just, like, zip through and maybe add it as a column to that CSV where it is an S3, so that it could be pulled through.
165 00:24:16.140 ⇒ 00:24:30.009 Katherine Bayless: Like, the full way. But if… if not, then it would be, like, company name matching. Normally, I would not say we will hand edit data for anything, but when it’s the… it’s our governing board, so it’s like, alright, well, we gotta get that one right.
166 00:24:30.010 ⇒ 00:24:41.700 Kyle Wandel: And depending upon who owns that, Catherine, name probably does match, honestly. If it’s membership who owns it especially, it definitely matches. That doesn’t? Okay, so then probably… maybe it doesn’t match that.
167 00:24:41.960 ⇒ 00:24:59.469 Katherine Bayless: And it doesn’t come… like, literally, like, Kaylee’s spreadsheet is the single source of truth for the board. So yeah, to your point, hopefully the company names do match, and that’ll make our lives easier, but I wouldn’t be overly surprised if we ran into scenarios where they didn’t. But at least it’s a quarter list.
168 00:24:59.920 ⇒ 00:25:13.870 Awaish Kumar: I think I can come up with an answer tomorrow, like, if I can match it, how much we can match it, and then we can see, like, if the gap is big, like, we can maybe try to solve it some other way, if that’s…
169 00:25:14.160 ⇒ 00:25:19.289 Awaish Kumar: If that is really small, then, yeah, up to you. How do you want to handle that?
170 00:25:19.650 ⇒ 00:25:36.130 Katherine Bayless: Yeah, I mean, I think it’s a small group overall, and it doesn’t change, except once a year. Like, the contact information for the people will change in the spreadsheet, but the, like, the companies and the associations, like, we’d only really need to, once a year after elections, look up a handful of records, so… I mean, yeah.
171 00:25:36.130 ⇒ 00:25:40.849 Katherine Bayless: This is the one exception I will make to the hand editing data, is the board list, but…
172 00:25:41.380 ⇒ 00:25:55.160 Kyle Wandel: There’s no… there’s no custom field in… there might be a custom field, and remember, I don’t… maybe there’s not, but, I would think that we have it… I would think that it’s similar to committee data, because it is technically a committee.
173 00:25:55.160 ⇒ 00:26:01.850 Katherine Bayless: Yeah… I mean, honestly, it’s true, like, it really… it ought to be in numbers, but, I don’t know if it is.
174 00:26:03.870 ⇒ 00:26:07.590 Kyle Wandel: I can probably try to look in the committee stuff, but I think there is…
175 00:26:08.440 ⇒ 00:26:11.369 Kyle Wandel: a committee associated with the executive board.
176 00:26:11.780 ⇒ 00:26:12.530 Katherine Bayless: Okay.
177 00:26:12.800 ⇒ 00:26:15.779 Kyle Wandel: I could be… maybe I’m thinking of some… maybe just the…
178 00:26:16.180 ⇒ 00:26:18.760 Kyle Wandel: You don’t think there’s something else, but I think there is.
179 00:26:19.290 ⇒ 00:26:23.250 Katherine Bayless: Yeah, I mean, if it’s in there and it’s up-to-date, then we could actually…
180 00:26:23.400 ⇒ 00:26:26.140 Katherine Bayless: We could use that instead, but.
181 00:26:26.540 ⇒ 00:26:29.149 Awaish Kumar: Yeah, like, if you have some kind of CSV,
182 00:26:29.340 ⇒ 00:26:32.600 Awaish Kumar: We can easily seed it into Snowflake.
183 00:26:32.950 ⇒ 00:26:38.290 Awaish Kumar: Which will kind of reduce, like, the… reduce the work, what we have already done.
184 00:26:38.480 ⇒ 00:26:41.340 Awaish Kumar: And we can come up with some solution for the future.
185 00:26:42.100 ⇒ 00:26:55.219 Katherine Bayless: Yeah. Yeah, the reason I’m kind of skeptical, Kyle, that it’s, like, truly in remembers is only because, at least in so far as I’ve been here, the authoritative source for the board list was always Kaylee’s spreadsheet, rather than, like, you know, all the old data team.
186 00:26:55.220 ⇒ 00:26:55.760 Kyle Wandel: Yeah.
187 00:26:55.760 ⇒ 00:27:05.359 Katherine Bayless: None of them said, pull it from Impexium, they all said, like, ask Kaylee for latest spreadsheet kind of thing. So that’s why I’m like, chances feel low, but I’ll be pleasantly surprised.
188 00:27:07.720 ⇒ 00:27:10.029 Awaish Kumar: Okay, yeah, so…
189 00:27:10.160 ⇒ 00:27:20.180 Awaish Kumar: I think, like, these… these are all the tables which basically show different segments in the… In the PDF.
190 00:27:20.390 ⇒ 00:27:22.410 Awaish Kumar: So, any questions?
191 00:27:22.760 ⇒ 00:27:26.260 Awaish Kumar: Apart from, you have already raised.
192 00:27:28.610 ⇒ 00:27:36.400 Katherine Bayless: Well, I guess it’s not technically a different question, but same question, really, for, like, the session scans. Like, are those…
193 00:27:37.100 ⇒ 00:27:41.499 Katherine Bayless: joinable to… the other data…
194 00:27:45.250 ⇒ 00:27:53.130 Kyle Wandel: Those are joined on ID to registrant ID for CES. That’s the… that’s really the only way.
195 00:27:53.540 ⇒ 00:28:08.099 Katherine Bayless: Right, so if we wanted to, for tomorrow, if we want to show the, like, session attendance engagement piece for Samsung, we would do the filtering based on company and, email domain, I guess.
196 00:28:08.100 ⇒ 00:28:10.200 Kyle Wandel: Still, yeah. Basically, yep.
197 00:28:10.200 ⇒ 00:28:11.490 Katherine Bayless: Okay, okay.
198 00:28:12.090 ⇒ 00:28:13.969 Kyle Wandel: Yeah, do we have to attend the… go ahead.
199 00:28:13.970 ⇒ 00:28:14.600 Awaish Kumar: Oh, hot.
200 00:28:14.850 ⇒ 00:28:18.789 Awaish Kumar: I think we have already joined here, the company.
201 00:28:20.160 ⇒ 00:28:21.050 Katherine Bayless: Yeah.
202 00:28:21.510 ⇒ 00:28:26.119 Katherine Bayless: Yeah, the company names in this session, Scan Data, it’s just, it…
203 00:28:27.550 ⇒ 00:28:39.930 Katherine Bayless: it is very inconsistently the same as what’s in remembers. Like, there are some that match, but for the most… I mean, I would say it’s probably less than 50% that actually match.
204 00:28:40.120 ⇒ 00:28:58.259 Kyle Wandel: Yeah, for the batch scan data, we’ll have to go from registrant ID to exhibitor slash Impexium ID that’s stored in registration, and then go to remembers, basically. Until we have that identity management table, it’s… that’s where the pain point comes from.
205 00:28:58.670 ⇒ 00:28:59.440 Katherine Bayless: Right.
206 00:28:59.820 ⇒ 00:29:11.360 Katherine Bayless: And even, like, that column with the member ID that’s in the registration and exhibitor data, I trust it more in the exhibitor data, but on the registration side.
207 00:29:11.850 ⇒ 00:29:17.480 Katherine Bayless: That column only got populated if they came through my process, and…
208 00:29:17.480 ⇒ 00:29:32.860 Katherine Bayless: registered just in that category, I think, so, like, I have a feeling… I think it’s a good start, obviously, but yeah, I’m like, I have a feeling it’s gonna very quickly show cracks of, like, you know, people who were in Reg, but didn’t have that member ID follow them through.
209 00:29:33.020 ⇒ 00:29:34.530 Katherine Bayless: But, yeah.
210 00:29:35.540 ⇒ 00:29:39.780 Katherine Bayless: room for improvement, once we have the… the identity stitching. Okay.
211 00:29:40.820 ⇒ 00:29:46.329 Awaish Kumar: Okay, yeah, I think we are going to go back and see…
212 00:29:46.730 ⇒ 00:29:53.070 Awaish Kumar: that, like, for this table, I think, Kyle, what you mentioned, are you talking about this ID?
213 00:29:54.610 ⇒ 00:30:17.209 Katherine Bayless: Yeah, so that’s the, registrant ID, which would then, like Kyle said, that would key out to the CES registration data set, and then we would be able to, in their CES registration data, if they came through the correct channels, there’ll be a member ID column in there, which would key out to the remembers data.
214 00:30:17.210 ⇒ 00:30:17.859 Awaish Kumar: Okay, okay.
215 00:30:17.860 ⇒ 00:30:29.070 Katherine Bayless: we usually can’t rely just on that column. We usually have to say, like, okay, we checked the column, we checked the company name, and we check the email domain for the person to determine a match, so…
216 00:30:29.070 ⇒ 00:30:29.810 Awaish Kumar: Okay, okay.
217 00:30:29.810 ⇒ 00:30:32.960 Kyle Wandel: Yeah, okay. It’s a lovely three-parter.
218 00:30:33.330 ⇒ 00:30:37.069 Katherine Bayless: Yeah, it’s gonna be so much better with the identity stitching.
219 00:30:37.070 ⇒ 00:30:37.460 Kyle Wandel: Oh, I know.
220 00:30:37.460 ⇒ 00:30:38.200 Awaish Kumar: Okay.
221 00:30:38.440 ⇒ 00:30:46.170 Awaish Kumar: So it seems… it seems like we have to do multiple levels of identity stitching for… each table?
222 00:30:47.840 ⇒ 00:30:58.779 Awaish Kumar: Yeah, I mean, I guess, well, the levels are pretty consistent in that, like, if there is a formal link, we can use that, so if there’s a column with the impexium or remembers IDs.
223 00:30:58.900 ⇒ 00:30:59.520 Awaish Kumar: Copy it.
224 00:30:59.520 ⇒ 00:31:23.970 Katherine Bayless: Otherwise, then we fall back to company name, and then to email domain. And that’s where, in that remembers data, there are the company alias and, like, company link tables that have those values so that, you know, the company name in the one field where it’s, like, organization name might be, you know, Samsung, but then that alias table will have Samsung LLC, Samsung Inc, Samsung, you know, lab.
225 00:31:23.970 ⇒ 00:31:37.929 Katherine Bayless: Samsung, you know, UK, whatever, right? And so then we’ll be able to make more matches using those aliases, and then the same with the links. It’ll give us all of the known email domains that we can consider for matching.
226 00:31:38.040 ⇒ 00:31:38.790 Katherine Bayless: Hmm.
227 00:31:38.790 ⇒ 00:31:40.570 Awaish Kumar: Okay, so do we want to keep…
228 00:31:41.920 ⇒ 00:31:49.459 Awaish Kumar: Yeah, my question here would be that, do we want to keep at organizational level, like, subsung, or do we want to, like, have…
229 00:31:49.740 ⇒ 00:31:53.230 Awaish Kumar: subset UK, or, like, all these different
230 00:31:53.690 ⇒ 00:31:57.400 Awaish Kumar: Segment, like, the one-layer board.
231 00:31:58.050 ⇒ 00:32:00.960 Katherine Bayless: I think… so, like, at this stage.
232 00:32:02.890 ⇒ 00:32:22.749 Katherine Bayless: Actually, yeah, no, sorry, let me think about this. Yeah, what we would like to do is, once it is joined back to the correct record from remembers, use the organization name field from remembers, rather than the kind of all-over-the-place stuff that might have been entered, so that way we see a consistent company name in the reports.
233 00:32:22.780 ⇒ 00:32:29.610 Katherine Bayless: And we, you know, we kind of obscure all of the chaotic, raw versions of it through the matching process.
234 00:32:29.990 ⇒ 00:32:31.580 Awaish Kumar: Okay, okay, got it.
235 00:32:31.880 ⇒ 00:32:32.540 Katherine Bayless: Yeah.
236 00:32:35.050 ⇒ 00:32:39.719 Awaish Kumar: Okay, yeah, I think, yeah, we can work on identity switching.
237 00:32:40.350 ⇒ 00:32:40.710 Katherine Bayless: Yeah.
238 00:32:40.710 ⇒ 00:32:47.410 Awaish Kumar: Yeah, apart from that, I think that’s all I had to show, like, so we have a catalog here.
239 00:32:47.780 ⇒ 00:32:53.610 Awaish Kumar: that is a GitHub, the data, and then Snowflake. This is how it lives.
240 00:32:54.880 ⇒ 00:32:56.440 Awaish Kumar: I think the…
241 00:32:57.060 ⇒ 00:33:03.940 Awaish Kumar: major piece which is missing right now is identity strategic, that I think we try… we’ll try to work
242 00:33:04.430 ⇒ 00:33:09.099 Awaish Kumar: This week, come up with some solution this week, yeah.
243 00:33:10.140 ⇒ 00:33:28.660 Katherine Bayless: Yeah, yeah, I mean, this is awesome, like, I mean, it’s incredible, like, how much you guys did from just, like, Friday to today, and I think it gives us a lot of stuff we can talk to the membership team about tomorrow. I think, yeah, the identity stitching is the next big thing that’s gonna really, like, give us the ability to start using a lot of this stuff, but…
244 00:33:28.660 ⇒ 00:33:35.680 Katherine Bayless: But yeah, this is awesome. I don’t want to, cut Kyle short too, though. Like, if you’ve got other, like, more nitty-gritty questions, by all means.
245 00:33:38.440 ⇒ 00:33:57.090 Kyle Wandel: No, I think it’s good. I think it really does, like, just come down to, like, what is the logic for a lot of this stuff. Catherine, I did send you… it does look like EV bill is in the community data, so we can easily just pull that straight from it, if that count is correct, but, it looks like it’s liaisoned by Kaylee, and as long as that’s updated, it seems like it’s good.
246 00:33:57.140 ⇒ 00:34:03.829 Kyle Wandel: So that’s gonna be good. I think it really is just kind of hammering out, some of the main logic. I’ve told…
247 00:34:04.320 ⇒ 00:34:16.229 Kyle Wandel: detail on this, and I’m definitely willing to help out. Once we start reviewing the code, I can start updating some of the logic, and then once we start doing the README, I can even start developing pipelines even easier.
248 00:34:18.590 ⇒ 00:34:30.389 Kyle Wandel: So yeah, I think this is exactly what we want, it’s just more so of now we gotta join all… everything together, and I probably know… I… I probably know the best logic for that, but, I’m working with membership and Catherine and everybody else to try and figure that out, so…
249 00:34:31.290 ⇒ 00:34:36.230 Awaish Kumar: Yeah. So, obviously, while we work on this, we might have questions.
250 00:34:36.699 ⇒ 00:34:39.339 Awaish Kumar: We are going to post that in Slack.
251 00:34:40.580 ⇒ 00:34:50.339 Awaish Kumar: Yeah, and… and, like, that’s what I said, like, we will be working on that piece this week, and at least try to come up with something.
252 00:34:50.449 ⇒ 00:34:51.550 Awaish Kumar: if we’d…
253 00:34:51.670 ⇒ 00:34:58.180 Awaish Kumar: yeah, I don’t know how it goes, but yeah, we will come up with some kind of solution there.
254 00:34:59.490 ⇒ 00:35:02.450 Awaish Kumar: Like, I also had one question.
255 00:35:02.650 ⇒ 00:35:12.209 Awaish Kumar: Regarding, this media opportunities and engagement group, like, although, Kyle, you have answered that Slack yesterday.
256 00:35:13.570 ⇒ 00:35:17.920 Awaish Kumar: But, like, what we are… when we are talking about engagement here.
257 00:35:18.050 ⇒ 00:35:30.930 Awaish Kumar: what it looks like, like, it is a C… we have a table called CES Events History, which gives us all the registered and attended, like, data.
258 00:35:31.440 ⇒ 00:35:36.709 Awaish Kumar: So, is the… is, like, is that what we are calling it, engagement, or is that something different?
259 00:35:37.530 ⇒ 00:35:39.190 Kyle Wandel: Go ahead, Devin.
260 00:35:39.400 ⇒ 00:35:40.520 Katherine Bayless: No, go ahead, go ahead.
261 00:35:40.520 ⇒ 00:35:49.840 Kyle Wandel: I would just say, I really do think that this last thing is just more of, like, a… I wouldn’t say data dump, but it’s, like, an aggregation of all the other, kind of.
262 00:35:50.080 ⇒ 00:35:57.479 Kyle Wandel: opportunities or engagements they did before. So I would say that this… this other engagement stuff, which is all… I mean, I…
263 00:35:58.060 ⇒ 00:36:09.749 Kyle Wandel: it doesn’t even look like that’s what it’s doing, you know what I mean? It looks like it’s more of, like, an aggregation or summary of all of the stuff from above. It doesn’t even look like they are including, like.
264 00:36:10.980 ⇒ 00:36:15.779 Kyle Wandel: I mean, they probably do have it somewhere, but, like, member… it’s not like it’s just member lounges or promotional purchases.
265 00:36:15.810 ⇒ 00:36:33.549 Kyle Wandel: It’s literally just, like, everything it looks like. And that’s how they created it. Like, they… they created this file called Company Engagement, based on all these other engagements, and then they just tallied them up and aggregated them all up, and then used that as, like, the starting file, I guess, base file.
266 00:36:33.610 ⇒ 00:36:36.910 Kyle Wandel: And then put everything into it.
267 00:36:37.570 ⇒ 00:36:37.890 Katherine Bayless: Yeah.
268 00:36:38.230 ⇒ 00:36:41.660 Katherine Bayless: I think for what we’ll want to do with it is, like.
269 00:36:41.790 ⇒ 00:36:45.569 Katherine Bayless: We will probably not wind up rebuilding, like.
270 00:36:45.800 ⇒ 00:36:54.759 Katherine Bayless: this… necessarily, like, this approach, this logic, this list, this code, because, yeah, like, Kyle’s dug in under the hood, and it’s just like, whoa, this gets…
271 00:36:54.760 ⇒ 00:37:10.449 Katherine Bayless: deep, fast. Like, so it’s like, I’m… instead, I’m like, okay, the spirit of this data, right? Like, what are they trying to do with it? Is like, they just want to basically, like Kyle said, see everything that might have occurred that involves this company, and so I think…
272 00:37:10.680 ⇒ 00:37:31.209 Katherine Bayless: the way I’ve approached this in the past with data pipelining is, like, basically kind of coming up with a way that, you know, we’re… there’s, like, a… like, a dimension around, like, engagement opportunity, right? Or something like that, or activity, or touchpoint, right? And that’s where we’ll keep kind of the, like, canonical list of, like, all the things we consider to be an engagement.
273 00:37:31.210 ⇒ 00:37:48.829 Katherine Bayless: And then a fact table that’s, like, you know, kind of joining the individual to the company to the engagement, to the time for it. And so we would start just, yeah, creating basically a massively tall table of everything people do that interacts with our systems that we’re interested in knowing about.
274 00:37:48.830 ⇒ 00:38:04.149 Katherine Bayless: And so, yeah, it’s like, I think we’ll be rebuilding this in spirit, even though I don’t think we’re gonna go back to whatever this was. Because a lot of it, like Kyle said, it was really, like, hand-done, in Excel, like, manual review, logic that only existed in somebody’s head, that kind of stuff.
275 00:38:05.700 ⇒ 00:38:13.779 Awaish Kumar: Yeah, like, I… I think the best approach then would be to, like, if we can document or kind of create a catalog.
276 00:38:14.010 ⇒ 00:38:16.130 Awaish Kumar: For the raw data we have.
277 00:38:16.390 ⇒ 00:38:17.450 Katherine Bayless: And…
278 00:38:17.500 ⇒ 00:38:21.960 Awaish Kumar: Share it with someone who’s in the team, like, the… who’s looking at that.
279 00:38:22.320 ⇒ 00:38:40.710 Awaish Kumar: So, basically, if they know what data is available there, like, they can come up with, okay, I need this kind of report, or… like, either there could be two ways. One is we hand over them the catalog, and they look at that and come up with something, or they just come…
280 00:38:41.180 ⇒ 00:38:44.159 Awaish Kumar: We… we do a discovery with them.
281 00:38:44.270 ⇒ 00:39:00.859 Awaish Kumar: And they have some questions, right? Okay, these are my pain points, I want to look at that, but I’m not able to. We want to see that. Like, maybe they have some more sheets they are trying to use. So if we look at that, like, what they’re trying to do right now.
282 00:39:00.890 ⇒ 00:39:05.789 Awaish Kumar: Maybe we can create a better version of this.
283 00:39:05.960 ⇒ 00:39:06.630 Katherine Bayless: Yeah.
284 00:39:06.890 ⇒ 00:39:23.769 Katherine Bayless: I think, yeah, exactly. The other thing, too, though, is, like, again, with kind of, like, knowing that we’ve had such, you know, inadequate access, and data in the past, like, some of the things that people are used to listing in those conversations are, like.
285 00:39:25.740 ⇒ 00:39:27.060 Katherine Bayless: It’s not that they’re, like.
286 00:39:27.060 ⇒ 00:39:51.919 Katherine Bayless: invalid. It’s just like, okay, well, like, we can actually, now that we’re glowing up all of our data practices, like, we can do a little better. And so, like, I think there’s a kind of a collaborative effort here, where it’s like, yes, we do want to work with the teams to figure out, like, what are the engagement points that you are interested in, but then we also need to kind of help them think a little bit bigger and deeper about, like, okay, what are the things we really should be tracking, like, meaningfully? Because eventually, of course, what we want to do is figure
287 00:39:51.920 ⇒ 00:40:16.230 Katherine Bayless: out, like, which of these engagements has the most predictive capability for renewal, for attendance, for engaging on a committee, you know, whatever it might be, because once we get a lot of these basic reports done, the next big thing people are asking for is attribution, right? Like, who gets credit for this sale, kind of thing? Where did this contact come from? Who’s prospecting them? We have two sales teams, so we need to keep them singing off the same sheet of music, so, like.
288 00:40:16.370 ⇒ 00:40:30.010 Katherine Bayless: I think we’ll have to drive the conversation a little bit to get people to think about it the right way, so that it’s not just this, like, laundry list of things, but rather something intended to become predictive around behaviors.
289 00:40:31.780 ⇒ 00:40:36.100 Awaish Kumar: Yep, I… Okay, yeah, I think.
290 00:40:36.100 ⇒ 00:40:36.639 Katherine Bayless: I dream big.
291 00:40:36.640 ⇒ 00:40:37.420 Awaish Kumar: I get it.
292 00:40:37.420 ⇒ 00:40:54.470 Kyle Wandel: Well, the way they use data here in, like, forever is the most basic use of data you’ve ever heard in your entire life, literally just list building and prospecting. So, like, I think Catherine were trying to go to the next level of that, and I literally showed a executive member, like.
293 00:40:54.510 ⇒ 00:41:01.800 Kyle Wandel: Year-over-year growth of different product interests, basically, and he was blown away by that. Like, simple stuff like that is just, like.
294 00:41:02.320 ⇒ 00:41:22.070 Kyle Wandel: ripe for picking, and, like, I don’t think, like Catherine said, I don’t think they even really truly know what data can do quite yet. But this first part, because it’s visual, and it has all of the engagements of CTAs, they love this, this is their favorite report, like, there’s so many ways we can make this better, and I think to build off what Catherine said.
295 00:41:22.070 ⇒ 00:41:27.230 Kyle Wandel: The end goal is for them to go into Snowflake and then ask Snowflake’s UI or agent, hey.
296 00:41:27.300 ⇒ 00:41:40.970 Kyle Wandel: Samsung… what was Samsung’s engagement last year? Or, like, what is Samsung’s engagement points, basically? So we can build that, like, the dashboard is great, it’s more so the institutional knowledge behind the dashboard that matters.
297 00:41:41.970 ⇒ 00:41:51.800 Katherine Bayless: Yeah, exactly, exactly. Like, there’s a bunch of descriptive analytics we need to deliver because it’s fair to want them, but then we want to kind of really rapidly go towards, like.
298 00:41:51.800 ⇒ 00:41:52.830 Awaish Kumar: Yeah.
299 00:41:53.590 ⇒ 00:42:01.689 Awaish Kumar: Yeah, I get it. I think we, we have some base models when we are done with idle tree stitching.
300 00:42:01.840 ⇒ 00:42:04.450 Awaish Kumar: That would be the next step, and after…
301 00:42:04.600 ⇒ 00:42:14.229 Awaish Kumar: After that, once they see this data, they will have more questions that maybe could be something we consider going forward.
302 00:42:14.550 ⇒ 00:42:38.850 Katherine Bayless: Yeah, no, no, I’m super excited, honestly. Like, my style, generally, around, like, you know, introducing data to teams is, like, the sooner you can get something in front of them for them to say, like, yes, no, oh, hmm, right? Like, I just… I love getting the conversation going, rather than trying to, like, build everything in a vacuum and hope they like it, right? And so, yeah, so I think tomorrow’s gonna be a lot of fun to just start really…
303 00:42:38.850 ⇒ 00:42:39.720 Katherine Bayless: Get those.
304 00:42:40.140 ⇒ 00:42:49.319 Kyle Wandel: all of the badge skin stuff that I’ve posted in the DataOps people, or DataOps, whatever, Slack, is because other people have contacted me and be like, hey, this is wrong, can you fix this? I’m like, yep, sounds good.
305 00:42:49.720 ⇒ 00:42:53.359 Katherine Bayless: yeah, a data man.
306 00:42:54.060 ⇒ 00:42:57.629 Kyle Wandel: Yeah, it’s bad. It’s probably… I didn’t realize it was that bad, but it’s really bad.
307 00:42:58.250 ⇒ 00:43:11.360 Katherine Bayless: I mean, yeah, like, the fact that… well, I mean, getting in the weeds, but yeah, I mean, there’s, like, a legal consideration around the, like, you know, we’re saying that these were people that went to a thing that now you can market to them, and it’s like, that might not be very true.
308 00:43:12.690 ⇒ 00:43:18.699 Kyle Wandel: It is, I mean, you can figure it out, but it is definitely a, not easy to figure out.
309 00:43:19.180 ⇒ 00:43:20.430 Katherine Bayless: Yeah. Yeah.
310 00:43:21.210 ⇒ 00:43:22.939 Katherine Bayless: But…
311 00:43:22.940 ⇒ 00:43:23.940 Kyle Wandel: It’s great, yeah.
312 00:43:24.500 ⇒ 00:43:26.069 Katherine Bayless: Go ahead.
313 00:43:26.070 ⇒ 00:43:26.700 Kyle Wandel: dying.
314 00:43:26.830 ⇒ 00:43:41.249 Kyle Wandel: This is great, like I said, you, Tom, and, to Ashwini, I would love to get my hands dirty, and if anything, you want me to QA something, or look at any of the PRs, that’d be great. So yeah, just let me know if I can… how I can help.
315 00:43:41.610 ⇒ 00:43:46.950 Kyle Wandel: I mean, I’m busy with my own stuff, but I am definitely… I’m definitely willing to help.
316 00:43:47.210 ⇒ 00:44:02.900 Awaish Kumar: Yeah, like, what… I think what we might need help with is… is just when you… when we don’t know the, like, the business context yet. Like, we are new to the team, so we might not know that, like, what this data means, what this field means, or…
317 00:44:02.940 ⇒ 00:44:10.940 Awaish Kumar: how you are stitching it together. So, we will be throwing some questions into Slack if we come across anything, but yeah.
318 00:44:11.720 ⇒ 00:44:31.419 Kyle Wandel: One… one… one tool that I… I don’t… I can’t remember if I told you guys I created it, but one tool I’ve actually been using is, if you go to, my, my DevMart, so under WandelDev, there’s a view… a couple views for data… for Remember’s Data Dictionary, that might help a little bit, but it’s just because, like, DevMarts, and then Wandel…
319 00:44:31.480 ⇒ 00:44:35.060 Kyle Wandel: Deb, I believe, is where it’s under.
320 00:44:35.060 ⇒ 00:44:35.540 Awaish Kumar: moon.
321 00:44:35.540 ⇒ 00:44:37.030 Kyle Wandel: We’re Dev at the bottom.
322 00:44:37.440 ⇒ 00:44:37.980 Awaish Kumar: This word?
323 00:44:37.980 ⇒ 00:44:39.290 Kyle Wandel: And, yeah…
324 00:44:39.610 ⇒ 00:44:55.470 Kyle Wandel: And then views and tables, they’re just a bunch of dictionaries and stuff, and how they relate, and the joins of each table. This is all remembers. It’s not necessarily… it’s not all remembers, it’s just CRM, but CRM is obviously a big portion of the data.
325 00:44:56.070 ⇒ 00:44:57.280 Awaish Kumar: Okay, yeah.
326 00:44:58.090 ⇒ 00:45:00.660 Katherine Bayless: Yeah, thank you. I think that will be…
327 00:45:00.700 ⇒ 00:45:06.929 Awaish Kumar: That will come in handy. There’s a lot of information in there that will be useful. Yeah, thank you for that.
328 00:45:06.930 ⇒ 00:45:26.610 Kyle Wandel: Yeah, and I mean, and honestly, the coolest thing is, maybe this is good and bad, like, that’s almost all Python slash AI generated. So that’s… that’s obviously good and bad. Like, you have to verify stuff, and I have to verify stuff here and there, but, it’s a good starting point. Like, what I’ve been doing is using that as a starting point, and then I’ll reach out to membership and be like, hey.
329 00:45:26.660 ⇒ 00:45:40.189 Kyle Wandel: what do you think of this? And then ask them how… how… what X is, like, Samsung dues. Like, I figured that out before it ever even told me, and I was like, oh, is there dues 65? And they’re like, yep. I’m like, okay, I think I got it. So…
330 00:45:41.480 ⇒ 00:45:43.349 Katherine Bayless: Yeah, it’s definitely an adventure.
331 00:45:45.320 ⇒ 00:45:49.680 Awaish Kumar: Yep. Okay. Yeah, I think that’s it from… for me.
332 00:45:50.840 ⇒ 00:45:58.760 Katherine Bayless: Cool. Well, thank you so much for the time, and like I said, all the work. I mean, that’s a lot of models put together in a short window, so,
333 00:45:58.760 ⇒ 00:46:13.660 Katherine Bayless: Like, yeah. And after we’re… we have our meeting with the membership team tomorrow, I think I have a couple calls, like, back-to-back, but we’ll definitely let you know, like, how it goes, and we’ll capture all of the, like, feedback and requests and stuff like that, and put them into Asana so that we can work on them, and yeah.
334 00:46:13.730 ⇒ 00:46:14.310 Katherine Bayless: Hell yeah.
335 00:46:14.310 ⇒ 00:46:25.349 Awaish Kumar: Yeah, yeah, I’m also taking notes from this meeting, and whatever next steps we have discussed, and we will be working on that, and looking forward to the feedback from the
336 00:46:25.480 ⇒ 00:46:27.330 Awaish Kumar: From the team, yeah. Thank you.
337 00:46:27.830 ⇒ 00:46:28.799 Katherine Bayless: Yeah, thank you.
338 00:46:28.800 ⇒ 00:46:29.650 Kyle Wandel: Thank you.
339 00:46:29.860 ⇒ 00:46:30.310 Katherine Bayless: See you later.
340 00:46:30.310 ⇒ 00:46:31.820 Awaish Kumar: I… when I see you.