Meeting Title: Brainforge x CTA: Weekly! Date: 2026-02-27 Meeting participants: Awaish Kumar, Katherine Bayless, Kyle Wandel, Chi Quinn, Uttam Kumaran, Ashwini Sharma
WEBVTT
1 00:01:54.390 ⇒ 00:01:54.920 Katherine Bayless: B.
2 00:01:54.920 ⇒ 00:01:55.470 Awaish Kumar: Right?
3 00:01:55.840 ⇒ 00:01:56.229 Awaish Kumar: Good morning.
4 00:01:56.990 ⇒ 00:01:58.040 Awaish Kumar: I’m running.
5 00:01:58.480 ⇒ 00:02:00.130 Katherine Bayless: Happy Friday!
6 00:02:00.450 ⇒ 00:02:01.610 Awaish Kumar: Oh, that’s funny.
7 00:02:03.160 ⇒ 00:02:06.530 Katherine Bayless: I feel like, it’s been a long, short week.
8 00:02:06.760 ⇒ 00:02:07.650 Awaish Kumar: Yeah.
9 00:02:08.240 ⇒ 00:02:09.699 Awaish Kumar: How’s it going?
10 00:02:10.449 ⇒ 00:02:23.390 Katherine Bayless: It’s going! I think, finally, I kind of got through some of the more, like, just… yeah, the beginning of the week was a lot. It was very intense, there was a lot of, like, different requests kind of going through, and, like, people were very…
11 00:02:24.490 ⇒ 00:02:31.719 Katherine Bayless: anxious. But it seems to have kind of calmed down, so I’m like, I think maybe today I can get some things done and, you know, actually clean out my inbox.
12 00:02:32.860 ⇒ 00:02:36.649 Awaish Kumar: And how… What… do you have any plans for the weekend?
13 00:02:37.450 ⇒ 00:02:46.440 Katherine Bayless: Well, fingers crossed. It’s supposed to be nice weather, tomorrow, and so I’m gonna try and get the car out and go for a nice drive.
14 00:02:47.690 ⇒ 00:02:50.400 Kyle Wandel: Is this the first time you’ve been out since the winter?
15 00:02:50.860 ⇒ 00:03:00.030 Katherine Bayless: Yeah, essentially, yeah. I managed to get it out, like, twice, just to, like, keep the battery going, you know? Just, like, run it up and down the road, but this would be my first, like, real drive.
16 00:03:00.290 ⇒ 00:03:04.719 Katherine Bayless: I’m realizing, though, Awash does not know why this is a thing, let me…
17 00:03:04.720 ⇒ 00:03:05.900 Kyle Wandel: Big deal, yeah.
18 00:03:05.900 ⇒ 00:03:08.569 Katherine Bayless: Let me find a picture for you.
19 00:03:08.570 ⇒ 00:03:15.430 Awaish Kumar: I have lived in Canada for, like, 7-8 months, so I know a little bit in the winters.
20 00:03:16.640 ⇒ 00:03:18.689 Kyle Wandel: Yeah, it gets real bad in the winters.
21 00:03:19.350 ⇒ 00:03:20.060 Awaish Kumar: Yeah.
22 00:03:20.900 ⇒ 00:03:25.249 Kyle Wandel: But what Kathy’s gonna show you is why she can’t take her car out.
23 00:03:26.890 ⇒ 00:03:28.769 Kyle Wandel: Cause I can take my car out, just fine.
24 00:03:28.770 ⇒ 00:03:29.310 Awaish Kumar: No.
25 00:03:30.030 ⇒ 00:03:35.440 Katherine Bayless: Let’s see, right, yeah, exactly. No, no further, issues there. Can I put the image in the chat?
26 00:03:36.640 ⇒ 00:03:40.580 Katherine Bayless: There you go. That’s… That’s why the car hasn’t been out in a while.
27 00:03:50.010 ⇒ 00:03:53.169 Kyle Wandel: So, Courtney, you are not going… wow. Catherine, you’re in the office today?
28 00:03:53.810 ⇒ 00:04:04.419 Katherine Bayless: Yeah, I came in, the Snowflake rep wanted to stop by in the afternoon to, like, chit-chat and, I think bring me a cupcake, apparently? So, apparently, I will take an entire metro ride for a cupcake.
29 00:04:05.010 ⇒ 00:04:17.820 Katherine Bayless: But yeah, and I’m… in theory, if I can manage enough youthful energy, I’m gonna go watch my old boss, he has a show tonight in DC, but it doesn’t start until 9.30, and I’m old. It’s late.
30 00:04:18.870 ⇒ 00:04:20.770 Katherine Bayless: We’ll see what happens.
31 00:04:21.450 ⇒ 00:04:26.660 Katherine Bayless: But yeah, there’s a few people here. JTK is here, I think Melissa might be here. I saw Erica.
32 00:04:28.180 ⇒ 00:04:30.710 Kyle Wandel: All the executive people.
33 00:04:31.000 ⇒ 00:04:32.539 Katherine Bayless: Yeah, exactly.
34 00:04:33.790 ⇒ 00:04:40.870 Awaish Kumar: Yeah, like, I think Utam will maybe join a little, later, during the call.
35 00:04:41.220 ⇒ 00:04:47.390 Awaish Kumar: So, yeah, I, like, we started working on the CE, Marveline.
36 00:04:47.980 ⇒ 00:05:01.499 Awaish Kumar: And I think, maybe we can literally talk about that. I have a document ready, we can use that to, like, jive our conversation, but yeah, I think that will be the… something I want to discuss.
37 00:05:01.830 ⇒ 00:05:02.590 Katherine Bayless: Yeah, yeah, no, I think that’.
38 00:05:02.590 ⇒ 00:05:04.859 Awaish Kumar: Yes, there’s anything, you know, you want to add.
39 00:05:05.350 ⇒ 00:05:28.269 Katherine Bayless: No, I think, honestly, I think it makes sense for you guys to kind of drive today. I know we have, I have a few small updates, and we have a few questions about how we want to kind of coordinate, but I think definitely first, let’s kind of talk through, what you guys have got, done and planned. I did take a look through the Word doc, that you, or the Google Doc that you shared, and I mean, it looks good to me, so I feel like,
40 00:05:28.400 ⇒ 00:05:30.440 Katherine Bayless: Though, oh, well, all parking lots.
41 00:05:30.440 ⇒ 00:05:32.509 Awaish Kumar: Yeah, that was one day switching, yeah.
42 00:05:32.800 ⇒ 00:05:36.610 Katherine Bayless: Yeah, yeah, yeah. Oh, the identity stitching. I didn’t see a dog for that one.
43 00:05:38.190 ⇒ 00:05:43.299 Awaish Kumar: I think you reached… the one we shared is… is for identity switching, and the…
44 00:05:43.550 ⇒ 00:05:43.970 Katherine Bayless: I don’t.
45 00:05:43.970 ⇒ 00:05:47.209 Awaish Kumar: Today is for CES modeling.
46 00:05:47.500 ⇒ 00:05:53.919 Katherine Bayless: Gotcha, gotcha. Okay, I think I went through the CES pre-audit doc. I didn’t catch the identity stitching one, but I.
47 00:05:53.920 ⇒ 00:05:59.390 Kyle Wandel: Identity Station 1 was last week, I think, and then the pre-audit one was yesterday.
48 00:05:59.950 ⇒ 00:06:01.680 Katherine Bayless: Gotcha, gotcha, gotcha. Okay.
49 00:06:04.080 ⇒ 00:06:05.659 Katherine Bayless: I’ll dig it up out of here somewhere.
50 00:06:06.490 ⇒ 00:06:08.890 Awaish Kumar: Okay, I can maybe share my screen.
51 00:06:09.110 ⇒ 00:06:10.060 Katherine Bayless: Oh yeah, perfect.
52 00:06:10.500 ⇒ 00:06:11.130 Awaish Kumar: Okay.
53 00:06:13.380 ⇒ 00:06:21.590 Awaish Kumar: So this, this is what we started modeling. So, like, using that pre-audit report, we tried to…
54 00:06:21.700 ⇒ 00:06:29.179 Awaish Kumar: Gather all the tables that we had in that report and classify them into some categories.
55 00:06:29.300 ⇒ 00:06:33.530 Awaish Kumar: And, and then maybe try… we actually went in
56 00:06:34.270 ⇒ 00:06:40.209 Awaish Kumar: Did exercise to how… what models could we build, to answer those questions?
57 00:06:40.310 ⇒ 00:06:41.650 Awaish Kumar: And,
58 00:06:42.070 ⇒ 00:06:49.230 Awaish Kumar: by… like, and I’ve started with some… some of it, and, like, we have… there are some initial, like,
59 00:06:49.520 ⇒ 00:06:53.800 Awaish Kumar: 15, 16 tables, which we can basically use this… basically this white…
60 00:06:53.850 ⇒ 00:07:12.760 Awaish Kumar: table to answer that, but yeah, I’ve, took a first pass, but there are obviously some edge cases, year by… year by year, like, the variations that we do not want to, like, the update the model for, and then…
61 00:07:12.760 ⇒ 00:07:19.830 Awaish Kumar: that’ll be good to go. So… these are two models, you can see that in Snowflake.
62 00:07:19.910 ⇒ 00:07:27.060 Awaish Kumar: One is called NSYS Registration Auditwide, other one is called… other one’s in… other one is in the reports.
63 00:07:28.140 ⇒ 00:07:33.259 Awaish Kumar: And… These are the filters that I have applied.
64 00:07:33.750 ⇒ 00:07:37.120 Awaish Kumar: to ex… like, you… like, I have…
65 00:07:37.290 ⇒ 00:07:48.009 Awaish Kumar: use that report config that you shared in Excel, and try to gather some of the configuration from that, and put it in the model. So, number one thing is that we…
66 00:07:48.620 ⇒ 00:07:55.290 Awaish Kumar: Like, use these fields, cancel flag or isCancel, to basically Filter out the canceled.
67 00:07:56.520 ⇒ 00:08:06.560 Awaish Kumar: registrations, and then this… there was one flag on ST01. I think… I thought it is maybe representing students?
68 00:08:06.560 ⇒ 00:08:08.180 Katherine Bayless: Yep, yep.
69 00:08:09.510 ⇒ 00:08:16.149 Awaish Kumar: Because I didn’t find a field where I can filter for ST01. Instead, there is a field called…
70 00:08:17.070 ⇒ 00:08:29.229 Awaish Kumar: ILD job description, which says student. So, actually, I tried to use that to filter the… filter out the students, and there is a title-coded thing for the past years,
71 00:08:29.410 ⇒ 00:08:38.260 Awaish Kumar: Which gives us the same, values. So, yeah, these are the two fields that I’m using for…
72 00:08:38.380 ⇒ 00:08:41.509 Awaish Kumar: past years, and for 2026, this is the one I’m using.
73 00:08:41.669 ⇒ 00:08:44.809 Awaish Kumar: And then, yeah, for emails, we are just free.
74 00:08:45.070 ⇒ 00:08:53.780 Awaish Kumar: Filtering out if there’s no valid email, or it’s null, or something like fake or something, then it filters out those registrations as well.
75 00:08:56.420 ⇒ 00:08:59.730 Awaish Kumar: Does that sound okay?
76 00:09:00.230 ⇒ 00:09:16.849 Katherine Bayless: Yeah, it does. So, I think I was just making notes for, things. So, one thing I was trying to track down before we hopped on this was that I do have somewhere, I need to find it, like, a document from the registration team that has, like, all of the different codes and mappings.
77 00:09:16.850 ⇒ 00:09:24.470 Katherine Bayless: And so I can share that too, but what you caught here with the individual job description and the ST01 thing is correct, like.
78 00:09:24.470 ⇒ 00:09:37.669 Katherine Bayless: when we get… or the report that we work with has the, like, more of the text values for the fields, whereas I think on the merit side, they see the codes, and so ST01 is student, and we see student, they see ST01.
79 00:09:37.670 ⇒ 00:09:50.440 Katherine Bayless: Eventually, hopefully, we’ll have, you know, more raw data, not this sort of weird report that they generate for us. But yeah, you guessed correctly. And I’ll find the rest of the mappings, and put them in that S3 bucket.
80 00:09:51.870 ⇒ 00:10:09.530 Awaish Kumar: Yeah, and similarly, I… yeah, I found this from, you know, report config, where we are using these for industry, media, but this one, there is one… one extra value, which I’m using eXp, which was not in the report config. Maybe, I think, I assigned it to Exhibitor, I…
81 00:10:09.800 ⇒ 00:10:10.440 Awaish Kumar: I hope…
82 00:10:10.440 ⇒ 00:10:21.489 Kyle Wandel: So it’s, like, yeah, it’s exhibitor paid. I actually, in the Brainforge, like, execution map, I wrote, I think, a little bit of comment of logic in terms of, like, what goes into each one.
83 00:10:22.210 ⇒ 00:10:25.019 Awaish Kumar: For…
84 00:10:25.130 ⇒ 00:10:41.929 Kyle Wandel: registration, like, what is an actual verified registration? What is a verified attendee? So, I can include some of that. I didn’t see this… I guess, yeah, I didn’t see this worksheet that you’re working on. I only had the execution roadmap, so I can look at this as well, if that would help.
85 00:10:43.410 ⇒ 00:10:48.169 Awaish Kumar: Yeah, that was before we dive into this. That is just, like, the…
86 00:10:48.290 ⇒ 00:10:54.950 Awaish Kumar: created a plan how we are going to tackle that. This is the one which I actually took a first pass in doing.
87 00:10:54.950 ⇒ 00:10:55.760 Kyle Wandel: Okay.
88 00:10:56.240 ⇒ 00:10:59.060 Awaish Kumar: This is what came after that.
89 00:11:00.910 ⇒ 00:11:08.770 Awaish Kumar: So, similarly, the… for, like, the data for 20… the past years and the 2026 a little bit varies in terms of…
90 00:11:08.900 ⇒ 00:11:11.300 Awaish Kumar: What comes, from where?
91 00:11:11.830 ⇒ 00:11:17.385 Awaish Kumar: For example, for this, like, 2023 to 2025, do… 2…
92 00:11:18.170 ⇒ 00:11:21.069 Awaish Kumar: To see, like, if it’s an on-site or pre-show.
93 00:11:21.300 ⇒ 00:11:24.890 Awaish Kumar: kind of filter. We actually… I didn’t find any…
94 00:11:25.600 ⇒ 00:11:28.529 Awaish Kumar: a specific field for that, so I just use that logic.
95 00:11:28.980 ⇒ 00:11:42.790 Awaish Kumar: If it is being registered in the same week, maybe then it’s, like, on-site registration or something. And for 2026, I am still figuring out where to find that from.
96 00:11:44.540 ⇒ 00:11:54.520 Katherine Bayless: Yeah, we actually… we are also trying to figure it out. So I did ask, Michael, Brown on the marketing team if he was familiar with what the logic was here, and I think…
97 00:11:54.670 ⇒ 00:12:03.399 Katherine Bayless: As far as Kyle and I were able to tell, and as far as Michael was guessing, I think registration week equals zero is how we have measured this in the past.
98 00:12:03.400 ⇒ 00:12:18.709 Katherine Bayless: However, there is, yeah, in that either it’s registrant interface or reg source code, one of the two does have a value for on-site, but the numbers using that field are significantly lower than, like, what’s been reported in the past, so I’m guessing…
99 00:12:19.320 ⇒ 00:12:38.809 Katherine Bayless: probably what we will end up finding out is that there’s two concepts. There’s, like, on-site reg, like, somebody who, like, straight up came in and had not done anything prior to, registering until they’re showing up at that desk in Vegas, and they’re walking through the whole process in person, because there’s, like, less than a thousand people that did it, which sounds about right. And then, like.
100 00:12:38.810 ⇒ 00:12:40.420 Katherine Bayless: The, you know, sort of…
101 00:12:40.420 ⇒ 00:12:51.149 Katherine Bayless: on-site, in terms of, like, last minute, is more so that registration week equals zero, kind of concept. But yeah, we need to figure out exactly what the logic is for the business on this one.
102 00:12:52.340 ⇒ 00:13:04.260 Awaish Kumar: Okay, so, like, should I just try to use an Excel sheet to whatever we are discussing? Like, open questions for both of us, so we can follow… follow that with a…
103 00:13:04.710 ⇒ 00:13:15.770 Katherine Bayless: Yeah, I mean, actually, or maybe what we could do is put in, like, an Asana ticket for each of the questions, and that way, as we kind of close them, we’ll have a record of, like, what it was and what the answer was right there in the board.
104 00:13:16.280 ⇒ 00:13:16.970 Awaish Kumar: Okay.
105 00:13:17.650 ⇒ 00:13:18.830 Katherine Bayless: Yeah, I like that.
106 00:13:21.370 ⇒ 00:13:27.410 Awaish Kumar: com… Then, moving on,
107 00:13:28.270 ⇒ 00:13:43.600 Awaish Kumar: Yeah, so for… yeah, moving on to using attendance, right? There are multiple ways I found this information. So, multiple places where I find this information. So, one is these registrations.
108 00:13:43.600 ⇒ 00:13:49.429 Awaish Kumar: For past years, it also includes a column itself, which says attended or
109 00:13:49.930 ⇒ 00:13:54.100 Awaish Kumar: I’ll register, whatever. Then the second place is…
110 00:13:54.420 ⇒ 00:14:02.240 Awaish Kumar: For 2026, we actually find this data in badge scans, and… For the…
111 00:14:02.490 ⇒ 00:14:15.500 Awaish Kumar: past years, we also find this data in event patterns, like CS image history table, which basically says attended. But the thing is, when I’m trying to, get the
112 00:14:15.650 ⇒ 00:14:24.849 Awaish Kumar: the final metrics, using these… all three means, I am not able to, get the same numbers, with…
113 00:14:25.080 ⇒ 00:14:27.770 Awaish Kumar: Like, as per your document.
114 00:14:28.050 ⇒ 00:14:36.490 Awaish Kumar: So registration… actually, the registration number matches with that. Almost, like, matches with that, but the attendance doesn’t match.
115 00:14:37.190 ⇒ 00:14:38.010 Katherine Bayless: Yeah, so…
116 00:14:38.010 ⇒ 00:14:39.300 Awaish Kumar: Posing the data, or…
117 00:14:40.110 ⇒ 00:14:53.700 Katherine Bayless: Yeah, so, like, in our… in our case, registration is, yeah, like, did they have a record in the reg system, of course. And then for, attendance, again, we’ve got kind of, like, dueling definitions, so…
118 00:14:53.730 ⇒ 00:15:11.609 Katherine Bayless: In terms of attendance at CES, big picture, it’s verified. So as long as they’re considered a verified attendee, they are considered to have attended CES. Like, there’s no further, sort of, flag that would differentiate between verified and attended. We equate those two as the same.
119 00:15:11.730 ⇒ 00:15:29.869 Katherine Bayless: But then for other events, like the badge scans, which are at the individual session level, or like some of the other things, like Zoom webinars and that kind of thing, we do have the attendance concept being either they scanned into the session, or, you know, they actually attended the webinar and were, you know, checked into the event.
120 00:15:29.870 ⇒ 00:15:35.570 Katherine Bayless: But yeah, in the case of CES, if they’re verified, they are attended as well.
121 00:15:36.940 ⇒ 00:15:41.120 Awaish Kumar: Okay, so we are saying that if I find in the registration table.
122 00:15:41.800 ⇒ 00:15:46.469 Awaish Kumar: If there’s a column which is verified, that means they attended the event.
123 00:15:47.190 ⇒ 00:15:57.100 Katherine Bayless: Well, so there is the column for is verified, but it’s only one component of the logic for an attendee to be considered verified. In the previous…
124 00:15:57.100 ⇒ 00:16:07.350 Uttam Kumaran: like, one is, like, encompassing… one encompasses… one is, like, the baseline, and then the other is, like, if that isn’t there, then the other also counts, so it’s kind of like…
125 00:16:07.800 ⇒ 00:16:09.900 Uttam Kumaran: That’s the Venn diagram a little bit, right?
126 00:16:10.550 ⇒ 00:16:15.119 Katherine Bayless: No, not exactly. It’s more like…
127 00:16:15.170 ⇒ 00:16:34.290 Katherine Bayless: is verified is inadequate to actually determine whether or not they’re verified. Oh yeah, here, Kyle popped it in the chat. So it’s like, for the past years, the data team did just put the verified status in a column using all the different criteria, but for the 2026 data, where we’re still
128 00:16:34.290 ⇒ 00:16:39.799 Katherine Bayless: calculating it. They do have to be verified, but also not canceled.
129 00:16:39.820 ⇒ 00:16:49.970 Katherine Bayless: They can’t have a certain promo code in their market code list, they can’t be a student, and their credentials have to be either approved or…
130 00:16:49.970 ⇒ 00:17:01.880 Katherine Bayless: their registration is incomplete, and it was a soft decline, or something like that. So yeah, so there’s a handful of things that are required to be considered verified that are beyond just the is verified column.
131 00:17:05.680 ⇒ 00:17:07.400 Katherine Bayless: I know it’s confusing.
132 00:17:08.220 ⇒ 00:17:14.280 Awaish Kumar: No, that’s, I just… I was just taking note of the case Kyle shared.
133 00:17:14.400 ⇒ 00:17:17.999 Awaish Kumar: So I can maybe use that.
134 00:17:18.210 ⇒ 00:17:19.310 Awaish Kumar: After that.
135 00:17:19.440 ⇒ 00:17:26.730 Awaish Kumar: Yeah, so… This is for Attendance flag, and
136 00:17:27.369 ⇒ 00:17:35.049 Awaish Kumar: Okay, then, yeah, that’s what I was getting, that numbers were very low, and I’m using these flags for the attendance.
137 00:17:35.470 ⇒ 00:17:43.970 Awaish Kumar: Then there is… these are basically the mapping of the columns that I’m trying to use for each of the…
138 00:17:44.280 ⇒ 00:17:45.250 Awaish Kumar: Years?
139 00:17:51.540 ⇒ 00:17:59.149 Awaish Kumar: Does that sound okay, maybe? Yeah, there’s one more thing on here, on the registrations. We have a…
140 00:17:59.260 ⇒ 00:18:12.260 Awaish Kumar: like, the city, country, regional information, but I think that seems to be tied to the attendee. It does not seem to be tied to the event, where the event is happening.
141 00:18:12.390 ⇒ 00:18:25.459 Awaish Kumar: So, I was looking at the data, like, to see, like, attendees in the Las Vegas, that means the event is happening in Las Vegas, and I was not able to find that info in any of the registration tables.
142 00:18:27.280 ⇒ 00:18:37.560 Katherine Bayless: Yeah, no, so since our events, it’s always in Vegas, and so these are, yeah, the geography we’re interested in is where are the attendees coming from?
143 00:18:37.610 ⇒ 00:18:47.230 Katherine Bayless: And so yeah, it’s the attendee mailing address. With the note that there are also some metrics that use that country HQ field.
144 00:18:47.230 ⇒ 00:18:58.500 Katherine Bayless: Which is not necessarily where the attendee is coming from, but where their company is headquartered. So sometimes we want that as the country value, and other times we want the attendee’s country value.
145 00:18:58.990 ⇒ 00:19:16.520 Awaish Kumar: Okay, yeah, I have that in the model. Okay. Obviously, we can use that, but the other one, which, like, on the report, I… on the top, as a title, I see the Las Vegas. I thought there are maybe different cities where you can do events, and we might need that info, but yeah.
146 00:19:18.020 ⇒ 00:19:19.190 Katherine Bayless: Always Vegas.
147 00:19:19.930 ⇒ 00:19:21.330 Katherine Bayless: We don’t fit anywhere else.
148 00:19:21.970 ⇒ 00:19:23.490 Awaish Kumar: Okay, and then…
149 00:19:23.490 ⇒ 00:19:35.390 Kyle Wandel: The… I don’t know if the CS history, it might be different. I mean, well, it will be different for some of them, but for… I guess for 2014, I guess for our purposes, it’s not different. I guess for the exhibitor booth, it’s different, but that would be it.
150 00:19:36.610 ⇒ 00:19:44.449 Kyle Wandel: Because we don’t have any data… we don’t have any registration data before 2014, so it’s always going to be in Vegas from that standpoint. But for the exhibitor stuff data.
151 00:19:44.570 ⇒ 00:19:54.199 Kyle Wandel: there… we will… maybe we will… could do this on our side, but we… there will be different, like, locations, because there’s one sometimes in Chicago, sometimes it’s in New York.
152 00:19:54.200 ⇒ 00:19:55.580 Katherine Bayless: And then, most recently.
153 00:19:55.580 ⇒ 00:19:59.199 Kyle Wandel: it’s in Las Vegas the entire time. I mean, that’s not…
154 00:19:59.320 ⇒ 00:20:03.530 Kyle Wandel: It’s not the end of the world, but, I mean… Really trivial. Really trivial.
155 00:20:04.030 ⇒ 00:20:19.739 Katherine Bayless: Yeah, and truthfully, it’s more of, like, an… eventually, we’d probably want to have an events table that is, like, just a seed file, and then, yeah, we could put the metadata around the event location there, because, yeah, it’s true, it did used to move around, it just hasn’t in a very long time.
156 00:20:19.740 ⇒ 00:20:25.999 Katherine Bayless: But yeah, for… to Kyle’s point, for the moment, it’s kind of moot, because the data we’re working with is… it’s been in Vegas that whole time.
157 00:20:26.780 ⇒ 00:20:28.270 Awaish Kumar: Okay, yep.
158 00:20:28.440 ⇒ 00:20:32.030 Katherine Bayless: But I’ll make a note for the historical events sort of file.
159 00:20:34.350 ⇒ 00:20:39.879 Awaish Kumar: Okay, for the data before 2014, that the exhibitors were in different cities, or…
160 00:20:40.890 ⇒ 00:20:41.800 Awaish Kumar: Is that okay?
161 00:20:41.800 ⇒ 00:20:49.920 Katherine Bayless: Like, a long time ago, it used to have… we used to have other cities that would, host the show as well, but for the last…
162 00:20:50.120 ⇒ 00:20:52.710 Katherine Bayless: at least decade. It’s just been in Vegas.
163 00:20:53.510 ⇒ 00:20:54.520 Awaish Kumar: Okay, yep.
164 00:20:55.560 ⇒ 00:21:01.210 Awaish Kumar: Then, moving on to this flag for Fortune 500, I think that’s…
165 00:21:01.440 ⇒ 00:21:04.919 Awaish Kumar: Just different ways to calculate for historical.
166 00:21:05.060 ⇒ 00:21:11.569 Awaish Kumar: For historical data, we just have a column, and for the 2026, we are going to get this data from.
167 00:21:13.110 ⇒ 00:21:13.980 Katherine Bayless: Yeah.
168 00:21:14.240 ⇒ 00:21:17.380 Katherine Bayless: So this one, I, I’ll give you the…
169 00:21:17.620 ⇒ 00:21:26.159 Katherine Bayless: Actually, I can’t remember if I put it in the folder yesterday or not, but in any case, finally yesterday, we got to the finalized list of the 2026 Fortune 500 attendants.
170 00:21:26.290 ⇒ 00:21:27.600 Katherine Bayless: I do think…
171 00:21:28.010 ⇒ 00:21:36.210 Katherine Bayless: the… like, what you found for the previous years, like, it was a column in the table where they were putting it against each individual attendee. I think…
172 00:21:37.120 ⇒ 00:21:50.260 Katherine Bayless: we do want to probably split that out so that it’s not sitting at the attendee level, but instead have the Fortune 500 list year over year, and whether or not that company was present at CES.
173 00:21:50.260 ⇒ 00:22:01.979 Katherine Bayless: Like, I think it makes more sense to affiliate the company to CES, to the Fortune 500 status, rather than the person to the company to the Fortune 500, because it’s just going to be a lot messier.
174 00:22:01.980 ⇒ 00:22:08.650 Katherine Bayless: Even though I say that knowing we’re immediately gonna get questions of, like, how many attendees came from Fortune 500 companies?
175 00:22:08.650 ⇒ 00:22:21.870 Katherine Bayless: But it still makes more sense to go through the companies and flags there, and then use the identity stitching tables to get to the attendees, rather than capturing it on the attendee data specifically.
176 00:22:22.100 ⇒ 00:22:25.210 Awaish Kumar: Yeah, in that case, actually, we are,
177 00:22:25.430 ⇒ 00:22:29.640 Awaish Kumar: In the teles, actually, we are just going to have a flag, which says,
178 00:22:30.080 ⇒ 00:22:33.939 Awaish Kumar: if a person belongs to, maybe, Fortune 500 company or not.
179 00:22:34.070 ⇒ 00:22:35.900 Awaish Kumar: But then we can have a…
180 00:22:36.100 ⇒ 00:22:40.639 Awaish Kumar: Like, some reviews on top of it that, okay, what companies actually…
181 00:22:40.770 ⇒ 00:22:47.159 Awaish Kumar: And which year, came, like, were, like, Fortune 500 companies came for the event.
182 00:22:47.560 ⇒ 00:22:49.289 Katherine Bayless: Right, but I think…
183 00:22:49.560 ⇒ 00:23:08.649 Katherine Bayless: we want to remove it from the attendee, because… because we know the attendee company names are much messier, and so we want one clean list of Fortune 500 companies, and just a kind of a Boolean yes-no for CES for the given year, and then we can join that data back out to the attendees.
184 00:23:08.840 ⇒ 00:23:18.719 Katherine Bayless: But if we rely on it at an attendee level, I don’t think we’re gonna get clean answers to the Fortune 500 question, and it is one that leadership gets very, like…
185 00:23:18.830 ⇒ 00:23:29.020 Katherine Bayless: nervous and confused when we give different answers to. So we want to make sure that the answer to how many Fortune 500 companies were at CES is always the same.
186 00:23:30.090 ⇒ 00:23:34.810 Awaish Kumar: Okay, so how we wanna, like,
187 00:23:34.920 ⇒ 00:23:53.259 Awaish Kumar: handle it. The one way is that, like, because I’m asking that because there are two fields, one called company name, and then there is Fortune 500, which also includes company name. So, for example, if we have a final table as a list of, like, year, CES, attendee, and then Fortune 500, yes, no.
188 00:23:53.610 ⇒ 00:24:02.869 Awaish Kumar: And then we are tying it back to the attendee table. Will we be using the company column, or will we be using Fortune 500 column?
189 00:24:03.330 ⇒ 00:24:17.210 Katherine Bayless: So we would want to use… we would want to remove the Fortune 500 column from this data entirely, and then just use the company column via the identity stitching bridge. Yeah, yeah. And we can give you the files of, like.
190 00:24:17.390 ⇒ 00:24:24.779 Katherine Bayless: each year, because we’d want to obviously also have the Fortune 500 rankings and stuff like that, so we’ll probably need one file for, like.
191 00:24:25.240 ⇒ 00:24:35.959 Katherine Bayless: 2023 Fortune 500, 2024, 2025, 2026, and it’ll have the Fortune 500 list, the company ranking, name, and then the… whether or not they were present at CES.
192 00:24:37.220 ⇒ 00:24:38.750 Awaish Kumar: Yeah, yeah.
193 00:24:39.690 ⇒ 00:24:44.380 Awaish Kumar: That would be nice. Currently, we have some… something like that for 2026, maybe?
194 00:24:44.770 ⇒ 00:24:45.710 Awaish Kumar: Yeah.
195 00:24:45.710 ⇒ 00:24:46.960 Kyle Wandel: Just 20 snacks right now.
196 00:24:47.120 ⇒ 00:24:47.860 Katherine Bayless: Yeah.
197 00:24:48.050 ⇒ 00:24:48.990 Katherine Bayless: Yeah.
198 00:24:49.130 ⇒ 00:24:50.400 Katherine Bayless: We’ll create the rest.
199 00:24:50.880 ⇒ 00:24:54.050 Awaish Kumar: Okay, similarly, there are these two flags, which…
200 00:24:54.900 ⇒ 00:25:13.479 Katherine Bayless: Yeah, so this will work the same way. We’ll want to track these, remove it from the attendee data, track it at the company level. For interbrand, we can persist the historical data, that’s fine, but we don’t need to worry about that one going forward, I’ve been told. For twice, we do need to figure out that one, so we can
201 00:25:13.540 ⇒ 00:25:22.199 Katherine Bayless: again, we’ll, like the Fortune 500, we’ll get the historical stuff, we’ll pull it together, and then we’ll figure out what the 2026 list and, matches should be.
202 00:25:23.480 ⇒ 00:25:36.210 Awaish Kumar: Okay, okay, yeah, and then… I think I already asked this question regarding… Image locations, and…
203 00:25:37.110 ⇒ 00:25:40.090 Awaish Kumar: Yeah, then moving on to, basically.
204 00:25:45.220 ⇒ 00:25:49.890 Awaish Kumar: the results I actually got Oh…
205 00:25:50.930 ⇒ 00:25:55.760 Awaish Kumar: pretty much, like, when I was seeing the number and comparing with the doc.
206 00:25:56.470 ⇒ 00:26:00.360 Awaish Kumar: I found this column matching a lot with the
207 00:26:01.830 ⇒ 00:26:05.799 Awaish Kumar: the Word document, the attendees list, like, the attendees was really…
208 00:26:06.200 ⇒ 00:26:10.860 Awaish Kumar: You know, with the branch scanner, or using the… Even stable.
209 00:26:11.210 ⇒ 00:26:11.920 Katherine Bayless: Hmm.
210 00:26:12.120 ⇒ 00:26:19.140 Katherine Bayless: Oh yeah, the events table is probably going to be mostly irrelevant to what the CES stuff. I think that’s all the, like, other…
211 00:26:19.500 ⇒ 00:26:26.800 Kyle Wandel: other events, yeah. The only, yeah, the only CES events is, like, will have CES, on it, basically.
212 00:26:27.010 ⇒ 00:26:27.420 Katherine Bayless: Yeah.
213 00:26:27.420 ⇒ 00:26:27.990 Awaish Kumar: Okay.
214 00:26:30.180 ⇒ 00:26:37.109 Awaish Kumar: Yeah, I think that’s mainly it. I… the rest of the document just includes all the queries that I used, or…
215 00:26:37.600 ⇒ 00:26:38.830 Awaish Kumar: Oh, that’s cool.
216 00:26:39.250 ⇒ 00:26:45.340 Awaish Kumar: And, and I will create the… whatever we have discussed for… so far,
217 00:26:45.670 ⇒ 00:26:49.690 Awaish Kumar: So, maybe I’m… I will just go and create all these tickets in Asana.
218 00:26:50.000 ⇒ 00:26:51.959 Awaish Kumar: So that we can follow up on these.
219 00:26:52.130 ⇒ 00:26:54.570 Awaish Kumar: Fortune 500 flag and all these things.
220 00:26:56.560 ⇒ 00:27:02.050 Awaish Kumar: Yeah, I think that’s… That’s it, Paul, for this.
221 00:27:02.750 ⇒ 00:27:14.149 Katherine Bayless: Okay, so I have one question. So for this CES data, so the file that’s currently in S3 is the pre-audit, data,
222 00:27:14.440 ⇒ 00:27:19.560 Katherine Bayless: I will need to replace it with what’s essentially the same pre-audit data.
223 00:27:19.560 ⇒ 00:27:40.240 Katherine Bayless: it’ll be minus… there were, like, 14 duplicate records the auditors found so far, and so they’re removed already. But it’s also adding a new column, and so… but for the product codes, because Merits realized that the data we were getting was truncated, which is why in that validation queries file, some of the, like.
224 00:27:40.240 ⇒ 00:27:47.089 Katherine Bayless: last handful of queries didn’t have exact matching counts, because the product code data we had was slightly truncated.
225 00:27:47.100 ⇒ 00:28:01.139 Katherine Bayless: And so they created an additional column in the report that gives us just the product codes, not the, like, full name, so that it fits in the field and won’t get truncated. And so when I put that file in S3, it’ll have that additional column in it.
226 00:28:01.140 ⇒ 00:28:08.289 Katherine Bayless: I wasn’t sure what the best way for me to, like, alert you guys that you’ll have to change the, like, the dbt model to bring it in was.
227 00:28:10.030 ⇒ 00:28:14.050 Awaish Kumar: I think, like, did you update it the file, or, like…
228 00:28:14.050 ⇒ 00:28:25.650 Katherine Bayless: I haven’t yet. I didn’t want to, like, do it and just, like, cause chaos, but I can put it in, like, today. I just wanted to figure out whether or not it would cause issues, because it’ll have that new column that we’ll need to ingest.
229 00:28:25.970 ⇒ 00:28:28.950 Awaish Kumar: Maybe if you can share that right now, we can…
230 00:28:29.210 ⇒ 00:28:31.329 Awaish Kumar: See how that changes the data?
231 00:28:31.920 ⇒ 00:28:34.909 Katherine Bayless: Yeah, sure, let me grab it from here.
232 00:28:37.660 ⇒ 00:28:39.760 Katherine Bayless: First, I have to remember where I put it, yes, okay.
233 00:29:04.480 ⇒ 00:29:06.770 Katherine Bayless: That’s a big file, Bloodford Token, there we go.
234 00:29:14.550 ⇒ 00:29:20.019 Katherine Bayless: Okay, let me… Alright, let me share my screen.
235 00:29:21.970 ⇒ 00:29:24.110 Katherine Bayless: Okay, so…
236 00:29:34.900 ⇒ 00:29:39.259 Katherine Bayless: It’s this one. So… actually, let me not expand the other one.
237 00:29:39.420 ⇒ 00:29:42.859 Katherine Bayless: So basically, actually…
238 00:29:45.680 ⇒ 00:30:03.209 Katherine Bayless: basically this column is what we had already, ProdCat Desk, and so most attendees can only select 5, so it’s not a problem. But media, can select as many categories as they want, and so for the media people that were selecting basically everything, it just didn’t all fit in the field, and so I think…
239 00:30:03.510 ⇒ 00:30:12.389 Katherine Bayless: They also fixed it so that this doesn’t get truncated, but regardless, this is now what we would want to use for the product category.
240 00:30:12.390 ⇒ 00:30:27.250 Katherine Bayless: interests is because these codes correspond… well, we’ll give you guys the mappings, obviously. But these codes correspond to the long-form names, and they fit in the cell more reliably. And so, basically, it’s this column getting inserted into the data set, kind of in the middle.
241 00:30:27.250 ⇒ 00:30:33.770 Katherine Bayless: And then this one getting moved out. But the rest of the columns should still be the same. That was the only change they had made.
242 00:30:35.470 ⇒ 00:30:39.790 Awaish Kumar: Okay, so is this the, 2026 registration data, or…
243 00:30:39.790 ⇒ 00:30:41.190 Katherine Bayless: Yes.
244 00:30:41.190 ⇒ 00:30:41.760 Awaish Kumar: Okay.
245 00:30:43.260 ⇒ 00:30:48.849 Katherine Bayless: I suspect we had the same issue in past years, but there’s no way to go back and get that data, so… this is what it is.
246 00:30:50.060 ⇒ 00:30:55.289 Awaish Kumar: Okay, yeah. Yeah, like, if you can upload that to a screen, you can ingest it.
247 00:30:56.050 ⇒ 00:30:57.170 Katherine Bayless: Okay, okay, cool.
248 00:30:57.170 ⇒ 00:30:58.450 Awaish Kumar: Okay.
249 00:31:00.650 ⇒ 00:31:01.220 Katherine Bayless: Awesome.
250 00:31:01.720 ⇒ 00:31:19.559 Katherine Bayless: And then at some point, when the auditors are… or when the audit is complete, we’ll get one more version of that file that shouldn’t have any schema changes, but it should just be the final curated, audited data set. And so we would only need to bring it into S3, load it into Snowflake one more time, and then we’re done. No more.
251 00:31:20.590 ⇒ 00:31:29.200 Awaish Kumar: Okay, yeah, like, right now, as we discussed, like, there are… you mentioned quite a few things on flags and these, like, these all needs to be…
252 00:31:29.320 ⇒ 00:31:36.529 Awaish Kumar: completed to get the correct results. Otherwise, we will be, like, off from the actual report.
253 00:31:37.220 ⇒ 00:31:38.720 Katherine Bayless: Yeah, exactly.
254 00:31:41.550 ⇒ 00:31:46.499 Awaish Kumar: Yeah, is there… do you have any more questions? Yeah, you mentioned you have some…
255 00:31:47.010 ⇒ 00:32:01.309 Katherine Bayless: Well, actually, I think some of them we ended up walking through. So the Fortune 500 was one of the things, that was on my mind, the new file for, the registrations, and then… let me see if I take another look at this…
256 00:32:02.590 ⇒ 00:32:07.839 Katherine Bayless: Oh, and then I was just gonna, actually, I can share my screen one more time.
257 00:32:08.350 ⇒ 00:32:09.110 Katherine Bayless: Excuse me.
258 00:32:11.700 ⇒ 00:32:26.230 Katherine Bayless: So I created, in the, data lake bucket, a directory for data governance. We can totally reorganize this, but my thought was, I went ahead and dropped in some of these mapping tables, so, like.
259 00:32:26.280 ⇒ 00:32:40.290 Katherine Bayless: The market codes will give us, in that market code list, what they corresponded to. This is totally not super, like, high priority to have these in there, but these are the mappings for those. The product interest codes does not
260 00:32:40.290 ⇒ 00:32:53.279 Katherine Bayless: currently include those A, you know, and then numeric value ones. This is, tracking their mappings year over year. Kyle might have a more complete file for this. I just have 23, 24, 25, 26 in here.
261 00:32:54.940 ⇒ 00:33:07.259 Katherine Bayless: program codes, these are the, like, the executive clubs, LIT, delegates, these are all of those different, code mappings. And then the job titles, this is the…
262 00:33:07.540 ⇒ 00:33:22.859 Katherine Bayless: So we have the question in red where we ask them to select their job title from the list, which is the same as what’s in here, basically, and then we have other places where we take their freeform title and map it out to those standardized ones, and so these are all of those mappings.
263 00:33:22.930 ⇒ 00:33:39.090 Katherine Bayless: So yeah, so I was just kind of dropping some of these data governance-y things into this directory, but we can ultimately kind of move them and put them wherever we think makes the most sense. I don’t think they need to persist as these CSVs specifically, it’s just this is the nature of the data at the moment.
264 00:33:39.110 ⇒ 00:34:03.559 Katherine Bayless: And so I can drop the other Fortune 500, and stuff like that in here, too, and then Kai was gonna drop in, sort of, like, one by one, the definitions for the different metrics, so that we’ve got these kind of, like, atomic definitions that we could bring in either to Snowflake or to the repo around, like, you know, what is a verified attendee, and what is, you know, the… what is senior level, and how do we
265 00:34:03.560 ⇒ 00:34:06.449 Katherine Bayless: Aggregate the different values into that.
266 00:34:06.450 ⇒ 00:34:07.509 Katherine Bayless: That kind of thing.
267 00:34:08.520 ⇒ 00:34:12.799 Awaish Kumar: Yeah, I think we also have a data platform documentation.
268 00:34:13.550 ⇒ 00:34:21.650 Awaish Kumar: like, Excel file. There also, you can have a list of metrics and the definitions, so we can actually maybe use that also for that.
269 00:34:22.130 ⇒ 00:34:29.519 Katherine Bayless: The only reason… because, like, Kai actually has it in Excel currently, too, the only thing that I was, like, kind of thinking ahead with it was, like.
270 00:34:30.860 ⇒ 00:34:32.080 Katherine Bayless: from a…
271 00:34:32.080 ⇒ 00:34:51.840 Katherine Bayless: changes standpoint, like, if it’s one sort of, you know, bigger Excel doc with everything in there, then we have to go into that every time we, like, learn a new definition. But if they’re kind of stored atomically as just individual metric definitions, then we can just replace the file in S3, or again, or the repo, wherever we decide to park them long-term.
272 00:34:51.840 ⇒ 00:35:05.150 Katherine Bayless: But we would have sort of this, like, concept of the definition that could persist, and, you know, we’d have versioning if we wanted to go back and see different previous versions, that kind of thing. Like, I just feel like there’s probably a little bit of knowledge loss if we keep it in Excel.
273 00:35:05.280 ⇒ 00:35:06.580 Katherine Bayless: Ultimately.
274 00:35:08.620 ⇒ 00:35:25.380 Awaish Kumar: Yes, and like… like, we are using dbt, there are ways to version control it and maybe write it as a macro, but it just depends how… how it looks, and if it is reusable. We can completely make it a macro and just use that anytime we want to.
275 00:35:26.390 ⇒ 00:35:29.610 Awaish Kumar: like, add a flag for his verify.
276 00:35:29.870 ⇒ 00:35:33.840 Awaish Kumar: And that way we can standardize the definitions.
277 00:35:34.200 ⇒ 00:35:37.990 Katherine Bayless: Yeah, yeah, that’s what I figured. Ultimately, they’ll make their way into code, yeah.
278 00:35:38.870 ⇒ 00:35:39.370 Awaish Kumar: Yep.
279 00:35:39.910 ⇒ 00:35:47.259 Awaish Kumar: And, for these mappings, like, do you want this to go into this model we are working for CS?
280 00:35:47.440 ⇒ 00:35:57.519 Katherine Bayless: Oh, I think you’ll have to, because, like, if we wanted to say, you know, how, like, what is the change in attendance around vehicle tech, at CES,
281 00:35:57.520 ⇒ 00:36:15.819 Katherine Bayless: we would need, like, the code mappings to know that it used to be vehicle tech on its own. At one point I think it was just car, and then it’s, you know, vehicle tech and advanced mobility, vehicle tech ampersand advanced mobility. So, like, these are going to be necessary to get correct year-over-year, aggregations.
282 00:36:15.830 ⇒ 00:36:23.250 Katherine Bayless: And then the program codes, sort of similarly, we do see different values for, like.
283 00:36:23.410 ⇒ 00:36:29.189 Katherine Bayless: some of these, not necessarily year over year, but between the different systems. Like, Merits calls them one thing.
284 00:36:29.350 ⇒ 00:36:46.880 Katherine Bayless: reg data, sometimes we get a different, slightly different value back out, because we get the text instead of the code. And then I also introduced another layer of complexity, because I created a form stack with, like, what Catherine thinks the code should be. So, like, all of those mappings are in here. And then the same, like, the job title stuff, yeah.
285 00:36:48.020 ⇒ 00:37:00.140 Awaish Kumar: Okay, yeah, and yeah, while looking at the product codes, I actually want to understand, like, we have… for each N&D, we have multiple product codes. What that exactly means? They are interested in those products, or…
286 00:37:00.320 ⇒ 00:37:15.590 Katherine Bayless: Yeah, so when they register for the event, we ask them, like, I think there’s, like, 48 different categories, and we say, like, which of these are you interested in? And we allow regular attendees to select up to 5. Media can select as many as they want.
287 00:37:15.930 ⇒ 00:37:34.130 Katherine Bayless: And then the four, excuse me, the five columns that are, like, rank 1, rank 2, rank 3, rank 4, Rank 5, those are where we ask the attendee to say, okay, these were the five categories you were interested in, now what’s your top most interested second, third, fourth, fifth? Yeah, exactly. And so then what we…
288 00:37:34.350 ⇒ 00:37:53.190 Katherine Bayless: Truthfully, we don’t use the data for much currently, but what we could use it for going forward would be to match attendees to exhibitors or to sessions, that kind of thing, to make recommendations. Right now, we use it mostly for analysis, like, you know, how many attendees came that were interested in AgDuck? Actually, here, let me… I’ll show you in Streamlit,
289 00:37:53.400 ⇒ 00:37:56.190 Katherine Bayless: How we tend to use this data, because that actually might be kind of helpful.
290 00:37:56.970 ⇒ 00:37:57.770 Katherine Bayless: He’s seeking.
291 00:38:00.960 ⇒ 00:38:03.260 Katherine Bayless: Hmm… This one.
292 00:38:07.350 ⇒ 00:38:16.420 Katherine Bayless: So, this is a report that I had built for, for Dave Fennessy, who’s, on our marketing team that helps a lot with, like, sales and lead gen.
293 00:38:16.560 ⇒ 00:38:23.200 Katherine Bayless: And so, basically, these product categories, he can go through and select whichever one he’s interested in.
294 00:38:23.320 ⇒ 00:38:28.969 Katherine Bayless: This will filter it to be just the people that had said that it was their top-ranked choice.
295 00:38:29.200 ⇒ 00:38:41.079 Katherine Bayless: And then these will kind of go through all of the other attendee demographics for people interested in this product category. And so he’ll use this for developing, like, marketing collateral and stuff like that.
296 00:38:42.570 ⇒ 00:38:45.619 Awaish Kumar: Okay, is it built on top of raw data, or…
297 00:38:45.710 ⇒ 00:39:00.059 Katherine Bayless: Yeah, right now, this is just built on top of the raw CES 2026 data, but once you guys have the models completed, we’ll remap it to that, because he would like to be able to see, like, the year-over-year comparisons, which aren’t possible in this right now.
298 00:39:00.630 ⇒ 00:39:01.840 Awaish Kumar: Okay.
299 00:39:02.460 ⇒ 00:39:03.170 Katherine Bayless: Yeah.
300 00:39:05.560 ⇒ 00:39:08.030 Awaish Kumar: Yeah, I think then…
301 00:39:09.790 ⇒ 00:39:15.499 Awaish Kumar: I have a lot of things that, if you’ve answered, I can go back and work on those models.
302 00:39:15.640 ⇒ 00:39:22.670 Awaish Kumar: And then, yeah, in parallel, actually, he might be working on some few more models for CES, and then…
303 00:39:24.500 ⇒ 00:39:31.940 Awaish Kumar: Yeah, and I’m… yeah, we are also going to create tickets for whatever we have discussed, and, like, for the open questions.
304 00:39:32.050 ⇒ 00:39:35.070 Awaish Kumar: And, yeah, that’s… I think that’s the plan.
305 00:39:35.900 ⇒ 00:39:41.269 Katherine Bayless: Okay, that sounds good. Do we want to talk about the identity stitching piece at all, or…
306 00:39:42.430 ⇒ 00:39:52.200 Awaish Kumar: Yeah, we can talk about it, like, that’s… yeah, what… what we started, and I think we… we got a go-ahead from Kyle to actually work on that.
307 00:39:52.990 ⇒ 00:39:55.949 Awaish Kumar: But, yeah, we can discuss that.
308 00:39:57.240 ⇒ 00:40:03.310 Katherine Bayless: Yeah, I mean, if you guys don’t have any questions, then that’s fine too. I wasn’t sure if there were still action items around it, since I had.
309 00:40:03.310 ⇒ 00:40:07.199 Awaish Kumar: Yeah, it’s like, we just, like, whatever,
310 00:40:07.360 ⇒ 00:40:14.120 Awaish Kumar: we thought could be the best way to approach this. We kind of proposed the solution.
311 00:40:14.230 ⇒ 00:40:20.010 Awaish Kumar: And we just need your feedback on, like, does that sound good, or… Are you okay with that?
312 00:40:20.700 ⇒ 00:40:21.280 Katherine Bayless: Okay.
313 00:40:21.790 ⇒ 00:40:23.980 Katherine Bayless: Then I can take a deep dive through the document.
314 00:40:24.260 ⇒ 00:40:40.749 Kyle Wandel: Yeah, when we talked a little bit about it yesterday, they said around the high level, they got 99, like, around 90% matches, which I feel like is pretty darn good. And then maybe we can take that list of that 10%, and then kind of send it over to membership and give their viewpoint. I mean, a lot of it…
315 00:40:40.960 ⇒ 00:40:55.049 Kyle Wandel: I’m actually doing, like, a little bit of cleanup now, kind of, like, just looking at 2026 stuff for CES, like, comparing names to, do that, whatever, Kinsey request that we were talking about earlier, and, like.
316 00:40:55.290 ⇒ 00:41:01.329 Kyle Wandel: 90% is pretty darn good, so, because it’s… this is really bad in a lot of areas.
317 00:41:02.160 ⇒ 00:41:03.609 Kyle Wandel: But we could get that.
318 00:41:03.760 ⇒ 00:41:05.309 Kyle Wandel: Last hump over again.
319 00:41:07.300 ⇒ 00:41:25.629 Kyle Wandel: I think I did mention that we want to match, and I think it was last week or two weeks, that we did… I did kind of mention that the registration is… should be the overarching universe, or maybe that was earlier this week, but the registration should be the start, and then everything should feed into that, like, should join into that, or match into that, basically.
320 00:41:28.700 ⇒ 00:41:35.740 Awaish Kumar: Yeah, I think that this is, like… I took this source as an example, and for that source, we have, like.
321 00:41:35.940 ⇒ 00:41:42.930 Awaish Kumar: Around 14,893 rows, where Out of those,
322 00:41:44.010 ⇒ 00:41:52.590 Awaish Kumar: like, for the 9,500, we actually have an Impexum ID through which we can join the… and get the organization name.
323 00:41:52.870 ⇒ 00:41:57.500 Awaish Kumar: For the ones where it’s null, we try to find it through email, company.
324 00:41:57.750 ⇒ 00:42:00.639 Awaish Kumar: Phone and website, and we can actually get,
325 00:42:01.000 ⇒ 00:42:16.909 Awaish Kumar: A few matches from each of these, and we are left with only 1,637, but then this is much more, like, a strategy to get even… to get even more matches, but that’s, like.
326 00:42:17.190 ⇒ 00:42:25.430 Awaish Kumar: Really depends on how you think, like, website domain, or try to get from the link, get the domain, and try to match it, or…
327 00:42:26.070 ⇒ 00:42:30.770 Awaish Kumar: Or the booths will have some new information to figure out…
328 00:42:31.500 ⇒ 00:42:37.039 Katherine Bayless: Yeah, so I think, truthfully, I mean, seeing that rate of matching,
329 00:42:37.890 ⇒ 00:42:40.279 Katherine Bayless: Actually, okay, let me back up. So…
330 00:42:40.300 ⇒ 00:42:50.840 Katherine Bayless: the nature of the data that you’re looking at here, right? So, the Impexium slash remembers, that database is used for the membership team, and so…
331 00:42:50.840 ⇒ 00:43:01.590 Katherine Bayless: all of the companies in there are either going to be members, or prospective members, or some other company of interest, former members, that kind of thing, right? But they’re only going to be the companies
332 00:43:01.590 ⇒ 00:43:05.010 Katherine Bayless: That have had some sort of interaction with our membership team.
333 00:43:05.280 ⇒ 00:43:28.160 Katherine Bayless: the exhibit space data is coming from the Salesforce CRM, and so that team focuses strictly on exhibit space sales at CES, and so their data is going to contain only those companies that have had either some… in this case, this is only the companies that have ended up exhibiting. That database does have more data on, like, prospects and former and that kind of thing.
334 00:43:28.160 ⇒ 00:43:30.040 Katherine Bayless: That we’re not seeing here, but…
335 00:43:30.040 ⇒ 00:43:33.790 Katherine Bayless: The gap between the two systems are probably
336 00:43:33.860 ⇒ 00:43:46.299 Katherine Bayless: Probably… some of it’s gonna be dirty data, which, yeah, like, we can try and match, but there’s also probably a number of, like, exhibitors who aren’t qualified for membership based on their company’s location, size, or type.
337 00:43:46.300 ⇒ 00:43:55.069 Katherine Bayless: And so, there are probably situations where they haven’t created a record in remembers at all for that company because they didn’t need to.
338 00:43:55.070 ⇒ 00:44:12.930 Katherine Bayless: going forward, we’ll create those records if… because we’re using remembers as the backbone for the identity stitching, so we’ll need to create any missing companies in remembers, so that we can use them, or use that database as the source for the matching
339 00:44:12.930 ⇒ 00:44:15.400 Katherine Bayless: Across systems. But there are probably
340 00:44:15.460 ⇒ 00:44:22.689 Katherine Bayless: legitimately quite a few companies that are not in remembers, because they weren’t necessary to be in there before.
341 00:44:24.130 ⇒ 00:44:26.489 Awaish Kumar: Okay, and also, like,
342 00:44:26.740 ⇒ 00:44:46.740 Awaish Kumar: So, like, the thing I mentioned about being 90% there, or something like that, is just for this source. This is, like, we are taking one source as an example and trying to lay out an approach, and we are going to repeat that for each individual table. I don’t know how we are going to do,
343 00:44:46.920 ⇒ 00:44:54.480 Awaish Kumar: In terms of, accuracy with the other tables, but that’s… that’s… if approach is good, we can try that out and see if it…
344 00:44:55.270 ⇒ 00:45:03.659 Katherine Bayless: Yeah, exactly, and I think if this approach is what we stick with, then it’s, like, the remainder that just don’t have a match, like Kyle said, we can kind of
345 00:45:03.770 ⇒ 00:45:10.520 Katherine Bayless: Curate them into a dataset, ask the membership team to just, like, verify that we’re not missing something obvious, like.
346 00:45:10.680 ⇒ 00:45:27.830 Katherine Bayless: maybe they know exactly where that record is, and we just didn’t match it in any of our logic, or it is new to that system and needs to be created. And so, yeah, then as you repeat the exercise on the other datasets, we would find more companies that we either need to add or match.
347 00:45:28.340 ⇒ 00:45:28.900 Awaish Kumar: Yep.
348 00:45:30.950 ⇒ 00:45:31.610 Katherine Bayless: Yeah.
349 00:45:32.350 ⇒ 00:45:33.080 Awaish Kumar: Okay.
350 00:45:33.560 ⇒ 00:45:36.860 Awaish Kumar: I think, like, that’s what we…
351 00:45:37.240 ⇒ 00:45:49.040 Awaish Kumar: we have to implement some of it, it’s already, like, 70% there in terms of implementation, which I did during building this document, but there are missing pieces we need to…
352 00:45:49.600 ⇒ 00:45:51.440 Awaish Kumar: do that, and I think we sh…
353 00:45:51.620 ⇒ 00:45:56.830 Awaish Kumar: Yeah, today’s Friday, so, yeah, in the next… early next week, we might be able to finish that.
354 00:45:57.300 ⇒ 00:45:59.030 Katherine Bayless: Okay, cool. Yeah.
355 00:45:59.810 ⇒ 00:46:10.910 Katherine Bayless: The, the other thing that I’ll look into, too, is, like, we do have a Power Automate key, that we got from remembers as part of doing the data share thing, and so…
356 00:46:11.180 ⇒ 00:46:21.190 Katherine Bayless: for the purpose of sending some of these, like, unmatched companies, you know, back to create records. And also, like, aliases and domains that we hadn’t…
357 00:46:21.430 ⇒ 00:46:33.530 Katherine Bayless: that we found this way, but didn’t have, and remembers to start with, like, those records we’d want to write back to that database, and so maybe that Power Automate thing will come in handy there. If not, there is a way to, like, upload a CSV on the backend.
358 00:46:33.530 ⇒ 00:46:34.240 Awaish Kumar: But yeah.
359 00:46:34.730 ⇒ 00:46:35.860 Kyle Wandel: I agree, yeah. Okay.
360 00:46:35.970 ⇒ 00:46:40.370 Awaish Kumar: But what is Power Automate thing? Like, is that an API?
361 00:46:40.860 ⇒ 00:47:02.180 Katherine Bayless: I, I love that you don’t know, actually, because I am kind of upset that we might have to use it. So it’s, it’s Microsoft’s, like, version of Zapier, basically. And so, like, Power Automate is their, like, no-code integration tool, and basically remembers, as a vendor, has designed a set of Power Automate, like, actions.
362 00:47:02.180 ⇒ 00:47:12.870 Katherine Bayless: That you can take using this key. It is essentially an API key, but it’s a specific set of vendor-configured functionality that it can access. Yeah, it’s clunky.
363 00:47:13.270 ⇒ 00:47:15.640 Awaish Kumar: Yeah, we have Zapier a lot.
364 00:47:15.640 ⇒ 00:47:17.350 Katherine Bayless: Yeah, it’s like that, yeah.
365 00:47:18.460 ⇒ 00:47:23.300 Awaish Kumar: Now there’s a lot of automation tools that are similar to Zapier, it’s hard to remember everything.
366 00:47:23.910 ⇒ 00:47:24.540 Katherine Bayless: Yeah.
367 00:47:24.780 ⇒ 00:47:25.400 Katherine Bayless: Yeah.
368 00:47:25.400 ⇒ 00:47:32.770 Awaish Kumar: Yeah, we can obviously use that, and there’s no problem if we have API key already, and once we have the list, we can maybe…
369 00:47:32.880 ⇒ 00:47:33.980 Awaish Kumar: Try to lose that.
370 00:47:34.200 ⇒ 00:47:34.900 Awaish Kumar: Okay.
371 00:47:35.510 ⇒ 00:47:41.999 Katherine Bayless: Yeah, and it doesn’t have to be something that’s part of this push right now, it’s just a thing we can leverage down the road, but yeah.
372 00:47:42.270 ⇒ 00:47:50.069 Katherine Bayless: Speaking of the remembers data share, the other question for me to bring to the team today was, so…
373 00:47:50.150 ⇒ 00:48:04.039 Katherine Bayless: the, the engineering team on their side, I guess they worked with Snowflake to have it so that they could do the cross-region replication thing without needing to stand up, like, a separate set of infrastructure on their side, so…
374 00:48:04.040 ⇒ 00:48:17.819 Katherine Bayless: they have connected us to a new version of the data share, and I think, as far as I understand, all I need to do is go in and rename the one we’ve been using to, like, you know, underscore old.
375 00:48:17.820 ⇒ 00:48:36.349 Katherine Bayless: and then rename this new one to match the name of the one we’d been using, and then everything just kind of magically keeps working? And then eventually they’ll shut off the old data share. But is that as simple as I think it is, or should I be more cautious around swapping out the data share?
376 00:48:37.130 ⇒ 00:48:38.240 Ashwini Sharma: It, it should work.
377 00:48:38.680 ⇒ 00:48:43.840 Ashwini Sharma: Yeah. It should work. We just have to look at the grants and access on the new database.
378 00:48:44.170 ⇒ 00:48:52.119 Ashwini Sharma: Just a quick question on that, that is in our infrastructure, like, in CTA’s infrastructure, or is it still there?
379 00:48:52.870 ⇒ 00:49:01.729 Katherine Bayless: Yeah, so here I can actually… I’ll show you. So it’s… it’s still in theirs, like, it’s still an external data share coming to us via Snowflake, but I guess they were able to do something…
380 00:49:02.090 ⇒ 00:49:05.569 Katherine Bayless: That made it so that they were… oops, what did I do?
381 00:49:10.130 ⇒ 00:49:27.180 Katherine Bayless: like, previously, they were… they had to have their own AWS account in our region in order to share the data with us, and so they had set that up, and I think now they can share it across region without that, and so they were going to kind of consolidate their AWS infrastructure back down.
382 00:49:27.270 ⇒ 00:49:32.030 Katherine Bayless: So it’s this one, the AMS Data Share V2.
383 00:49:32.310 ⇒ 00:49:43.540 Katherine Bayless: is the new one, and so I went ahead and we connected to it, made sure it worked, all the things, and so it’s the exact same as the previous data share, it’s just coming.
384 00:49:43.540 ⇒ 00:49:49.430 Awaish Kumar: If the names and the schema is the same, then it will just work.
385 00:49:49.670 ⇒ 00:49:52.219 Awaish Kumar: Unless there is schema changes.
386 00:49:52.660 ⇒ 00:49:55.579 Katherine Bayless: No, there shouldn’t be any schema changes, yeah.
387 00:49:56.650 ⇒ 00:49:58.329 Katherine Bayless: To proceed the exact same thing.
388 00:49:59.210 ⇒ 00:49:59.820 Awaish Kumar: Okay.
389 00:50:00.450 ⇒ 00:50:18.359 Katherine Bayless: And then also, as part of that, this is the custom data that we have in the system. Truthfully, I don’t think we have a lot of stuff going on in here, and even if there is data in these tables, it might be older data that’s not necessarily in current use, so…
390 00:50:18.480 ⇒ 00:50:32.320 Katherine Bayless: I don’t think we need to worry about prioritizing exploring this, but if a question comes up and we can’t seem to find the data anywhere in the rest of the system, it might be hiding in here. But otherwise, I think this is just… we can let it sit until we have a use case for it.
391 00:50:32.750 ⇒ 00:50:38.320 Awaish Kumar: Okay, and for the other piece, like, do you want to rename it today? Like, maybe, I don’t know if it
392 00:50:38.520 ⇒ 00:50:42.659 Awaish Kumar: Because today’s Friday, I don’t know if that breaks anything.
393 00:50:43.800 ⇒ 00:50:50.690 Awaish Kumar: Yeah, maybe. If something like that, if not really urgent, we can maybe do it on one day.
394 00:50:51.540 ⇒ 00:51:01.360 Katherine Bayless: Yeah, we can do it Monday if you want. I told the, the engineer, I told him I’d get him an answer, by next week if we were up and running, so yeah, I think Monday’s fine, if that’s the preference.
395 00:51:01.540 ⇒ 00:51:02.270 Katherine Bayless: But…
396 00:51:02.630 ⇒ 00:51:08.050 Awaish Kumar: Yeah, like, I… I don’t know if… I’m just, considering the…
397 00:51:08.300 ⇒ 00:51:15.100 Awaish Kumar: that if something fails, then if there are bigger failures, then I don’t know how long that’s going to take us to.
398 00:51:15.430 ⇒ 00:51:17.400 Katherine Bayless: Yeah, yeah, yeah, totally fair.
399 00:51:17.750 ⇒ 00:51:24.930 Awaish Kumar: If, if, yeah, in the ideal case, it can… might work, just as it is, so… but, yeah.
400 00:51:26.670 ⇒ 00:51:33.840 Katherine Bayless: Yeah, okay, well, we can leave it for Monday, that’s totally fine. But yeah, so yeah, we’re just switching from this one over to this one.
401 00:51:34.290 ⇒ 00:51:34.960 Awaish Kumar: Okay.
402 00:51:37.350 ⇒ 00:51:43.410 Awaish Kumar: Yep, I think that’s… I think that we can do also, Ashwini, on our side, in dbt.
403 00:51:43.410 ⇒ 00:51:44.899 Katherine Bayless: Oh yeah, totally.
404 00:51:47.720 ⇒ 00:51:48.589 Awaish Kumar: Yeah, I think we can.
405 00:51:48.590 ⇒ 00:51:50.490 Ashwini Sharma: PB renaming has to be done.
406 00:51:50.600 ⇒ 00:51:52.090 Ashwini Sharma: On Snowflake only.
407 00:51:52.280 ⇒ 00:51:58.880 Ashwini Sharma: Renaming of a DB has to be done on Snowflake. Everything else should work fine after that. We don’t have to do anything else.
408 00:51:59.920 ⇒ 00:52:04.010 Katherine Bayless: I think OISH means you could just repoint the dbt code to the new name.
409 00:52:04.250 ⇒ 00:52:05.890 Katherine Bayless: Like, 6 in one, half dozen.
410 00:52:05.890 ⇒ 00:52:09.949 Ashwini Sharma: Yeah, yeah, that also works, yeah. You don’t have to do anything, right, yeah?
411 00:52:09.950 ⇒ 00:52:12.699 Awaish Kumar: Yeah, so just create a ticket, actually, for us, and then you.
412 00:52:12.700 ⇒ 00:52:13.270 Ashwini Sharma: Okay.
413 00:52:13.270 ⇒ 00:52:14.699 Awaish Kumar: Then just do that on Monday.
414 00:52:14.980 ⇒ 00:52:16.010 Ashwini Sharma: Alright, yep.
415 00:52:16.360 ⇒ 00:52:16.760 Katherine Bayless: Okay.
416 00:52:20.440 ⇒ 00:52:27.160 Katherine Bayless: Okay. Team, anything else on our collective brains? I think those are all the things I had.
417 00:52:33.280 ⇒ 00:52:34.040 Katherine Bayless: Okay?
418 00:52:34.250 ⇒ 00:52:35.760 Katherine Bayless: Cool?
419 00:52:35.760 ⇒ 00:52:39.670 Awaish Kumar: Okay, thank you then. I think that’s wonderful.
420 00:52:40.060 ⇒ 00:52:44.550 Katherine Bayless: Alright, well, thank you, and if you guys come in with any questions, just let us know. We’re here to help.
421 00:52:45.060 ⇒ 00:52:48.720 Awaish Kumar: Okay, yeah, sure. Thank you, thank you so much.
422 00:52:48.720 ⇒ 00:52:49.609 Katherine Bayless: Have a great weekend.
423 00:52:50.230 ⇒ 00:52:50.900 Awaish Kumar: Endo.