Meeting Title: Brainforge x ABC Home and Commercial: Weekly Project Check Date: 2025-04-25 Meeting participants: Uttam Kumaran, Amber Lin, Janiecegarcia, Yvetteruiz, Scott_Harmon
WEBVTT
1 00:00:15.370 ⇒ 00:00:16.510 Uttam Kumaran: Hello!
2 00:00:16.990 ⇒ 00:00:19.530 JanieceGarcia: Hello, happy Friday.
3 00:00:21.620 ⇒ 00:00:22.600 Uttam Kumaran: Good morning!
4 00:00:23.030 ⇒ 00:00:23.960 JanieceGarcia: Morning, morning.
5 00:00:26.160 ⇒ 00:00:27.005 Uttam Kumaran: Matching.
6 00:00:28.152 ⇒ 00:00:31.269 JanieceGarcia: We sure are love! It.
7 00:00:32.970 ⇒ 00:00:33.860 Uttam Kumaran: Hey! Scott!
8 00:00:34.650 ⇒ 00:00:36.490 Scott_Harmon: Hey, Tom! Hi! Janice!
9 00:00:36.950 ⇒ 00:00:38.990 JanieceGarcia: Hello! Hello! Amber!
10 00:00:39.360 ⇒ 00:00:40.233 Amber Lin: Hi! There!
11 00:00:42.910 ⇒ 00:00:44.759 Scott_Harmon: Hi, amber! How are things in la.
12 00:00:45.724 ⇒ 00:00:48.200 Amber Lin: It’s actually, really, really cold.
13 00:00:48.200 ⇒ 00:00:49.460 Amber Lin: It is. Yeah.
14 00:00:49.460 ⇒ 00:00:50.060 JanieceGarcia: That’s.
15 00:00:50.530 ⇒ 00:00:59.019 Amber Lin: It is really, really cool it. It got better last week. And now this week I’m like wearing 2 layers of sweaters right now.
16 00:00:59.710 ⇒ 00:01:00.130 Scott_Harmon: Bye.
17 00:01:00.130 ⇒ 00:01:02.010 Amber Lin: When they wait.
18 00:01:02.450 ⇒ 00:01:03.420 Amber Lin: Yeah.
19 00:01:03.710 ⇒ 00:01:05.970 Scott_Harmon: Come and move to Texas. You’ll get over that quickly.
20 00:01:05.970 ⇒ 00:01:07.125 Amber Lin: Yeah, for sure.
21 00:01:09.620 ⇒ 00:01:10.650 Amber Lin: Well.
22 00:01:10.650 ⇒ 00:01:13.630 JanieceGarcia: Cold, like 2 days out of the week or out of the year.
23 00:01:14.286 ⇒ 00:01:15.139 Scott_Harmon: Right, yeah.
24 00:01:21.360 ⇒ 00:01:23.500 Amber Lin: Haven’t seen you last week, Scott.
25 00:01:24.400 ⇒ 00:01:25.359 Scott_Harmon: I’m sorry.
26 00:01:25.360 ⇒ 00:01:27.049 Amber Lin: Didn’t see you last week.
27 00:01:27.050 ⇒ 00:01:31.409 Scott_Harmon: Yeah, I had some extracurricular family stuff to take care of it.
28 00:01:31.836 ⇒ 00:01:32.689 Amber Lin: I see.
29 00:01:32.690 ⇒ 00:01:34.630 Scott_Harmon: Needed my attention. So.
30 00:01:35.680 ⇒ 00:01:36.190 JanieceGarcia: Fun, stuff.
31 00:01:36.490 ⇒ 00:01:41.320 Scott_Harmon: Back in the saddle this week. Yeah, no, not not no, not terribly. Not fun.
32 00:01:42.460 ⇒ 00:01:47.139 Scott_Harmon: No, we just had a we had a we had a
33 00:01:47.490 ⇒ 00:01:53.210 Scott_Harmon: death in the family. Was, it was expected it was. It was a 1 of our grandparents.
34 00:01:53.210 ⇒ 00:01:54.020 JanieceGarcia: I’m sorry.
35 00:01:54.020 ⇒ 00:01:56.070 Scott_Harmon: That’s okay. Thank you very much. But
36 00:01:58.060 ⇒ 00:02:01.670 Scott_Harmon: there was some drama with it. So we had to kind of wrestle with that.
37 00:02:01.670 ⇒ 00:02:02.890 Amber Lin: Oh, goodness!
38 00:02:03.410 ⇒ 00:02:04.360 Scott_Harmon: Hey?
39 00:02:05.340 ⇒ 00:02:06.770 Scott_Harmon: Part of life, I guess.
40 00:02:07.660 ⇒ 00:02:08.340 Amber Lin: Yeah.
41 00:02:09.060 ⇒ 00:02:14.390 JanieceGarcia: What is life without family drama? My mom is dealing with that right now with her and her sister.
42 00:02:15.950 ⇒ 00:02:20.189 JanieceGarcia: My grandparents are now gone, and they’re trying to sell their land and property. And yeah.
43 00:02:20.190 ⇒ 00:02:21.490 Amber Lin: Oh no!
44 00:02:22.050 ⇒ 00:02:24.280 Scott_Harmon: Yeah, yeah, yeah, yeah, it’s like, can we just.
45 00:02:24.710 ⇒ 00:02:25.460 JanieceGarcia: But
46 00:02:26.210 ⇒ 00:02:28.484 Scott_Harmon: Rama. Where would we be without it?
47 00:02:30.340 ⇒ 00:02:32.090 Scott_Harmon: Utah. I think you’re on mute or.
48 00:02:32.090 ⇒ 00:02:33.560 Uttam Kumaran: Oh, can you hear me now?
49 00:02:33.870 ⇒ 00:02:34.649 Scott_Harmon: There you go. Yeah.
50 00:02:34.650 ⇒ 00:02:51.852 Uttam Kumaran: Okay. No, my yeah, I was saying, it’s always a property. My dad had a very similar issue when his dad passed away. But he basically was like, you guys deal with it. I’m like he didn’t. He didn’t. He was like, I know, there’s gonna be drama. It’s yours. You guys can figure it out. I don’t want anything but.
51 00:02:52.300 ⇒ 00:02:58.269 JanieceGarcia: It’s it’s crazy. Well, my, the sad thing is, my mom does everything for it.
52 00:02:58.610 ⇒ 00:03:03.757 JanieceGarcia: But then my aunt wants to come in and make all the decisions. And my mom’s definitely
53 00:03:05.040 ⇒ 00:03:06.380 JanieceGarcia: over. So.
54 00:03:06.380 ⇒ 00:03:08.350 Scott_Harmon: You kind of like, you know my family.
55 00:03:09.535 ⇒ 00:03:11.230 Uttam Kumaran: Are we related.
56 00:03:11.230 ⇒ 00:03:11.590 JanieceGarcia: Bye.
57 00:03:12.705 ⇒ 00:03:14.935 Uttam Kumaran: Same story.
58 00:03:16.050 ⇒ 00:03:20.850 Scott_Harmon: Yeah, it turns out that lawyers and probate courts and wills and all that stuff can get a little bit.
59 00:03:21.380 ⇒ 00:03:22.190 JanieceGarcia: But fun.
60 00:03:22.720 ⇒ 00:03:26.080 Scott_Harmon: Not fun when you get family. That’s not all on the same page. So.
61 00:03:26.080 ⇒ 00:03:29.179 JanieceGarcia: 1st of all.
62 00:03:29.180 ⇒ 00:03:37.560 Scott_Harmon: Back to my go back to my lawyer and say, Now, are we sure this will is clear like? And then I called both my daughters, and said, Look, here’s the deal.
63 00:03:38.180 ⇒ 00:03:38.699 Amber Lin: To do that.
64 00:03:39.410 ⇒ 00:03:40.120 Scott_Harmon: Misunderstand.
65 00:03:42.850 ⇒ 00:03:45.300 JanieceGarcia: Yvette is in another meeting.
66 00:03:47.720 ⇒ 00:03:52.290 JanieceGarcia: So I don’t know when she’ll be joining, but she is in another meeting.
67 00:03:52.690 ⇒ 00:03:55.649 Amber Lin: Okay, is anyone else planning on.
68 00:03:55.650 ⇒ 00:03:56.880 Scott_Harmon: Even joining. Do you know.
69 00:03:56.880 ⇒ 00:03:58.990 JanieceGarcia: Steven’s on Pto. So is it just me.
70 00:03:59.930 ⇒ 00:04:02.950 Scott_Harmon: That’s good. Now we got the 18. We can get some stuff done. Finally.
71 00:04:03.259 ⇒ 00:04:09.140 Amber Lin: And Janice, there’s so much stuff this week. I’ll have to show you guys it’s it’s it’s incredible.
72 00:04:10.107 ⇒ 00:04:27.329 Amber Lin: Let me pull up the presentation. It’s pretty short today. So we’ll just blast through. And if you guys, if Utam, you want to talk to Scott about some of the phase 2 stuff. We can just use this time to talk about it as well. But let me send
73 00:04:29.650 ⇒ 00:04:31.080 Amber Lin: There we go.
74 00:04:31.780 ⇒ 00:04:36.959 JanieceGarcia: I don’t know how y’all work off of one screen. Sometimes I know Utami, you’re on multiple. I can tell by your eyes.
75 00:04:37.310 ⇒ 00:04:41.290 Uttam Kumaran: Yeah, yeah, I’m in like a control room. It’s a
76 00:04:42.127 ⇒ 00:04:46.289 Uttam Kumaran: but yeah, I just mentioned to Amber that I’m gonna get her monitor this week.
77 00:04:47.150 ⇒ 00:04:47.570 Uttam Kumaran: Okay.
78 00:04:48.580 ⇒ 00:04:52.228 Uttam Kumaran: I was like, I don’t know how I need to be like, I don’t know how you’re doing this.
79 00:04:52.420 ⇒ 00:04:57.320 Amber Lin: I mean this browser works. I have all my tabs.
80 00:04:58.230 ⇒ 00:05:06.980 Amber Lin: Yeah. And it lets me actually show meet in different places versus being stuck in my room, because if I have a monitor I’m gonna be stuck.
81 00:05:07.670 ⇒ 00:05:12.240 JanieceGarcia: Yeah, she might actually be coming.
82 00:05:21.820 ⇒ 00:05:23.349 Amber Lin: Do we wait a little bit.
83 00:05:23.350 ⇒ 00:05:24.779 JanieceGarcia: Yeah, she’s coming in.
84 00:05:25.030 ⇒ 00:05:26.000 Amber Lin: Okay. Awesome.
85 00:05:26.600 ⇒ 00:05:27.200 JanieceGarcia: Just
86 00:05:28.240 ⇒ 00:05:34.729 JanieceGarcia: no, you’re good. She’s getting getting back to her office and stuff. So give her some some time.
87 00:05:42.580 ⇒ 00:05:44.099 JanieceGarcia: Yeah, what time was your break.
88 00:05:45.870 ⇒ 00:05:47.609 JanieceGarcia: You want me to do it. 1015,
89 00:05:53.350 ⇒ 00:05:55.430 JanieceGarcia: because it’s in 15 min intervals.
90 00:05:55.750 ⇒ 00:05:58.610 JanieceGarcia: Intervals. Can’t speak today.
91 00:06:00.500 ⇒ 00:06:03.399 Scott_Harmon: Does that stuff show up that I’m doing this annotation stuff.
92 00:06:05.070 ⇒ 00:06:09.010 Scott_Harmon: Okay, I couldn’t tell. I’ve never done it before. And so I thought, Well, I’ll just see if it works.
93 00:06:09.640 ⇒ 00:06:11.019 Scott_Harmon: Huh! All right.
94 00:06:11.690 ⇒ 00:06:12.840 Amber Lin: Wow!
95 00:06:12.840 ⇒ 00:06:14.690 Uttam Kumaran: Oh, I use it all the time.
96 00:06:14.690 ⇒ 00:06:15.590 Scott_Harmon: Oh! Do you.
97 00:06:15.590 ⇒ 00:06:23.279 Uttam Kumaran: I cause sometimes you’re not. I don’t. It’s like top left. Oh, 8, like a little bit higher left, like little bit lower left. So I just draw.
98 00:06:23.470 ⇒ 00:06:25.639 Scott_Harmon: I’m sorry to muck up your slide. I guess it’s not on your slide.
99 00:06:26.490 ⇒ 00:06:29.140 Uttam Kumaran: You can hit the erase button, and you can erase them.
100 00:06:29.720 ⇒ 00:06:35.570 Scott_Harmon: Oh, got it? Yeah. Okay, well, that opens up all kinds of possibilities. I’m really.
101 00:06:35.780 ⇒ 00:06:38.339 Scott_Harmon: really irritating in these things. Amber. Okay.
102 00:06:38.955 ⇒ 00:06:42.030 Amber Lin: Helps drive the point across
103 00:06:47.490 ⇒ 00:06:49.059 Amber Lin: okay, waiting green event.
104 00:06:59.380 ⇒ 00:07:04.460 Amber Lin: I can always catch up and everybody has access to these slides.
105 00:07:10.360 ⇒ 00:07:15.970 Scott_Harmon: Why don’t we get started? And then I’m sure Yvette can if we send the slides. And she missed the 1st few we can.
106 00:07:16.590 ⇒ 00:07:17.030 Amber Lin: Yeah.
107 00:07:17.265 ⇒ 00:07:17.500 Scott_Harmon: Okay.
108 00:07:18.020 ⇒ 00:07:23.509 Uttam Kumaran: Yeah. And on the data stuff, amber. We reviewed a little bit yesterday. So but yeah, let’s get started.
109 00:07:23.510 ⇒ 00:07:24.440 Amber Lin: Okay, sounds good.
110 00:07:25.000 ⇒ 00:07:28.990 Amber Lin: So mo, mainly, 3 things.
111 00:07:29.190 ⇒ 00:07:30.310 Uttam Kumaran: Okay. Can you hear me now?
112 00:07:30.500 ⇒ 00:07:31.230 Scott_Harmon: Yeah.
113 00:07:31.230 ⇒ 00:07:32.620 Amber Lin: Yes, I can hear you.
114 00:07:35.680 ⇒ 00:07:39.909 Amber Lin: So 1st of all, on how we’re doing this week. So
115 00:07:40.030 ⇒ 00:07:43.270 Amber Lin: I think the main thing we should look at is
116 00:07:43.520 ⇒ 00:07:49.259 Amber Lin: the average use. So the total usage. And we’re at about
117 00:07:49.400 ⇒ 00:08:09.590 Amber Lin: 34 exchanges with Andy each day, which is pretty good, but, however, is pretty much consistent with last week, and so I would love for us to talk a little bit about. How do we boost that usage? So I’ll I have a slide for that, and we can talk about it in a bit.
118 00:08:09.890 ⇒ 00:08:11.340 Amber Lin: And so
119 00:08:11.470 ⇒ 00:08:24.709 Amber Lin: again on the leaderboard, as you can see this week, is all Janice. Janice is top one. We have still pretty much the same people that use it a lot.
120 00:08:25.359 ⇒ 00:08:29.759 Amber Lin: The other people I well, I’m glad that they used it.
121 00:08:30.060 ⇒ 00:08:37.280 Amber Lin: but I would love for them to use it more, and I know there’s people who are not using it yet.
122 00:08:37.419 ⇒ 00:08:41.059 Amber Lin: so we can go look at it and see who’s not using it.
123 00:08:41.973 ⇒ 00:08:47.169 Amber Lin: And we have good amount of feedback always would love more feedback.
124 00:08:47.280 ⇒ 00:08:49.670 Amber Lin: So that’s the state of these 2 things.
125 00:08:49.670 ⇒ 00:09:06.940 Uttam Kumaran: Yeah. And to give a sense of, we we talked to. Actually, you can go to the next slide, amber. We talked through each of these things in in a ton of detail yesterday. I still haven’t. I need to do probably a larger debrief with you, but I think we have a pretty good path on some action items. So
126 00:09:07.570 ⇒ 00:09:14.879 Uttam Kumaran: We also got a chance not only to meet as this crew, but we met with a few of the Csrs as well. So I have a
127 00:09:15.130 ⇒ 00:09:18.679 Uttam Kumaran: bunch of learnings that I didn’t get the chance to catch up with you on yesterday.
128 00:09:18.680 ⇒ 00:09:20.370 Scott_Harmon: What were couple?
129 00:09:20.490 ⇒ 00:09:23.679 Scott_Harmon: Could you just use the top top? 2.
130 00:09:24.596 ⇒ 00:09:40.053 Uttam Kumaran: Yeah. So a couple of things that people said so one I think there’s a mix there’s just 2 different. There’s like 3 different types of user bases. Right? There’s people that are new people who are basically onboarding. And there’s people who have, you know, seasoned, who have been here.
131 00:09:40.380 ⇒ 00:09:59.040 Uttam Kumaran: quite a while, and so I think part of the way we should consider how the bot communicates to them is their tenure. That’s 1 piece. The second piece is there’s still a couple of tweaks to be made across, handling for like abbreviations handling for like
132 00:09:59.080 ⇒ 00:10:25.099 Uttam Kumaran: just like certain lingo. That I think that didn’t. That may not be in the Central Doc, that we can add. The 3rd piece, I think, is just making it easier for Janice and team to update the central dock, which is the the training bot and stuff like that. So really just increasing the iteration cycles of us to give to get feedback, implement it and then get it deployed. I think overall, though, if I just look back at
133 00:10:26.510 ⇒ 00:10:38.299 Uttam Kumaran: some of my notes. We had some people that like joy, for example, is using it all the time. She’s new and she’s like super super excited to use it. We also have folks like Deborah.
134 00:10:38.300 ⇒ 00:11:00.080 Uttam Kumaran: who have been at the company for a long time. Honestly, their her knowledge of processes and systems are beyond currently like what Andy is doing in some degrees. And so for her, the pitch is more. How she could use that tool and actually helps us improve. Our documentation improves the experience for all the other Csrs.
135 00:11:00.580 ⇒ 00:11:03.760 Uttam Kumaran: yeah, there mainly were some stuff around acronyms.
136 00:11:04.286 ⇒ 00:11:11.350 Uttam Kumaran: And then everybody. I think we’ve promoted to use the feedback. Everybody is understanding how the thumbs up, thumbs down, works
137 00:11:12.020 ⇒ 00:11:13.520 Uttam Kumaran: things like that.
138 00:11:13.915 ⇒ 00:11:33.404 Uttam Kumaran: Everybody likes the oh, by the ways. And that was a huge hit. I think a lot. All 3 folks said it helped them remember to do that when they may not have remembered the script, or what the actual upsell was. Which is, which is huge. We didn’t get much feedback on the response times. I think it’s it’s fine right now.
139 00:11:35.060 ⇒ 00:11:36.230 Uttam Kumaran: yeah, that that was, I think.
140 00:11:36.230 ⇒ 00:11:42.131 YvetteRuiz: Emphasized on it. Udem! I think Joy emphasized on it, and I think April did a little bit. They said that it was. It was pretty good.
141 00:11:42.350 ⇒ 00:11:43.500 YvetteRuiz: Yeah, exactly.
142 00:11:43.500 ⇒ 00:11:44.550 YvetteRuiz: We’d like, yeah.
143 00:11:44.900 ⇒ 00:11:47.150 Uttam Kumaran: And then I asked a question about like
144 00:11:47.160 ⇒ 00:12:05.810 Uttam Kumaran: how the responses come to them like, do they like that? It’s sort of a narrative. Do they prefer like quick bullet points. Everybody said. Actually having it be more conversational, was helpful. They’re not necessarily reading it verbatim but it’s helpful if they’re sometimes lost for words on how to phrase
145 00:12:05.810 ⇒ 00:12:20.732 Uttam Kumaran: that they have that in front of them. So overall feedback was like very, very positive. I think, the biggest opportunity. We have some very specific things to improve with the Central Doc and those I think we’ll we can get through probably next week.
146 00:12:21.180 ⇒ 00:12:32.729 Uttam Kumaran: we also want to basically try to promote some of these heavy users. In order to share with more people how they’re using the bot. So I propose to the team that we should
147 00:12:32.990 ⇒ 00:12:39.300 Uttam Kumaran: potentially add everyone just to a slack channel all the Csrs, so that they can ask questions to us about
148 00:12:39.350 ⇒ 00:13:08.830 Uttam Kumaran: the bot updating things, questions on how to use it. It streamlines their ability to ask us directly, and then also people. Other people will then see the questions that are coming in amber. It’s really, really. I give the example of what we’re doing inside our company, right where we have some people that are very vocal. We have some people that are very shy with some people that are nervous about the technology. So we want to have different avenues. So I said both the slack group and potentially doing an office hours every other week, or something could be helpful.
149 00:13:09.500 ⇒ 00:13:33.030 Uttam Kumaran: And then the last thing, as I said, we’ll we’ll do those initiatives. If we still see that people are are sort of not using it. Then we’ll go meet one on one. But I want to do stuff in more in group settings. Where other people can see and they may just get the hang of it that way or understand who in the company they can go to for questions cause there’s only a few of us on our side, so this will quickly ramp beyond just the 20 people.
150 00:13:33.650 ⇒ 00:13:36.120 Scott_Harmon: Great super super summer in Utah. Thanks.
151 00:13:38.840 ⇒ 00:13:39.690 Amber Lin: Thank you.
152 00:13:39.880 ⇒ 00:13:59.549 Amber Lin: So to sum up where we’re at, the the only thing the main 2 or one thing we’re lacking is one about usage that we talked about and also improving, updating the knowledge base like we’re doing it really well right now. But what we want to is to make it even easier, so that would be a for the trainer. Bot!
153 00:13:59.840 ⇒ 00:14:04.849 Scott_Harmon: Are we going to get into the status of the trainer? Bot here, I assume, in subsequent slides.
154 00:14:08.065 ⇒ 00:14:12.222 Uttam Kumaran: Amber is amber frozen.
155 00:14:12.910 ⇒ 00:14:14.520 Scott_Harmon: I think I thunder with my question.
156 00:14:15.189 ⇒ 00:14:17.269 Uttam Kumaran: Yes, we will be.
157 00:14:18.680 ⇒ 00:14:22.419 YvetteRuiz: Yeah, that is one thing that we talked about with Udem yesterday, like, we really wanna
158 00:14:23.200 ⇒ 00:14:27.090 YvetteRuiz: get that better understanding of how that’s gonna we’re gonna start.
159 00:14:27.090 ⇒ 00:14:27.490 Uttam Kumaran: Yeah.
160 00:14:28.980 ⇒ 00:14:33.049 Scott_Harmon: It could be. We could be at the point where, and you just said it a minute ago.
161 00:14:33.860 ⇒ 00:14:46.100 Scott_Harmon: You know, we want to continue to get, especially the the less tenured Csrs, I think, can just we can get a ton of usage up now if we can get the knowledge loop
162 00:14:46.970 ⇒ 00:14:54.050 Scott_Harmon: shortened so that the more expert people can put knowledge in. Then I think you’re just going to see it compound.
163 00:14:54.280 ⇒ 00:15:00.689 Scott_Harmon: you know, you’ll see usage go up, because oh, there’s this kind of tricky bit of new knowledge that only I know.
164 00:15:01.040 ⇒ 00:15:01.430 Uttam Kumaran: Totally.
165 00:15:01.430 ⇒ 00:15:02.040 Scott_Harmon: Super easy, to.
166 00:15:02.040 ⇒ 00:15:02.520 YvetteRuiz: Yeah.
167 00:15:02.520 ⇒ 00:15:05.129 Scott_Harmon: Been there because people want to share knowledge right? Like.
168 00:15:05.130 ⇒ 00:15:05.690 JanieceGarcia: Right.
169 00:15:06.060 ⇒ 00:15:09.590 Scott_Harmon: Like one of the fun things about being an expert is
170 00:15:09.750 ⇒ 00:15:18.449 Scott_Harmon: sharing what you know like, that’s that’s what we want to. I think what we’ll really see adoption take off is when it becomes super easy for your more senior people
171 00:15:19.240 ⇒ 00:15:22.410 Scott_Harmon: to oh, I learned something that’s kind of yeah.
172 00:15:22.410 ⇒ 00:15:29.400 Scott_Harmon: you know it’s a tricky question. And then, like a minute, they can update the knowledge thing and and.
173 00:15:29.400 ⇒ 00:15:35.650 YvetteRuiz: You’re you’re spot on about that. And I think we had some conversation with him yesterday, too, is like the
174 00:15:35.790 ⇒ 00:15:47.500 YvetteRuiz: communication. You know. We we did an okay job with how we rolled it out. But I think we could have did a better job of where the the more not. You know senior people like what we’re talking about right now, right is.
175 00:15:47.880 ⇒ 00:15:59.929 YvetteRuiz: how is it going to benefit them? And they’re they’re the ones that are going to be teaching it right? So like it was interesting. Udem and I know when we met the past, Supervisor. Right? We would want her involvement in there. But she’s not been in there right.
176 00:15:59.930 ⇒ 00:16:00.310 Uttam Kumaran: Yeah.
177 00:16:00.310 ⇒ 00:16:28.359 YvetteRuiz: Where to, you know. Trying to explain it to her is like you want to be in there because your knowledge you want to make sure, you know, hey? I’m the person that they come to. They asked me the questions, so now I want to be able to redirect them to Andy. So me, and asking the questions, making sure that the responses I’m getting are accurate are super important. So getting the why to them is is going to be important. And that’s kind of the things that we’re gonna work on with the team.
178 00:16:28.730 ⇒ 00:16:31.750 Scott_Harmon: Just to piggyback on that event.
179 00:16:32.470 ⇒ 00:16:35.310 Scott_Harmon: One of the things that works really well
180 00:16:35.900 ⇒ 00:16:40.189 Scott_Harmon: for incentivizing more senior people to put in new knowledge is a
181 00:16:40.310 ⇒ 00:16:46.910 Scott_Harmon: not a contest, but like just like a dashboard that says, Oh, you know, event created 7 new.
182 00:16:47.280 ⇒ 00:16:53.100 Scott_Harmon: you know, pieces of knowledge. Last week. She won. You know she’s the top of the leaderboard. Those kind of things are really
183 00:16:55.300 ⇒ 00:16:56.789 YvetteRuiz: Love that idea. Scott.
184 00:16:56.790 ⇒ 00:17:01.690 Scott_Harmon: Don’t really like them. It’s it’s a lot like how open source software works.
185 00:17:01.690 ⇒ 00:17:03.670 Uttam Kumaran: Yes, the stars!
186 00:17:03.670 ⇒ 00:17:06.929 Scott_Harmon: A lot of kudos when they check in code. Lots of people use it like
187 00:17:07.881 ⇒ 00:17:13.350 Scott_Harmon: so there’s probably some really simple way to go, you know, just a weekly leaderboard with a.
188 00:17:13.359 ⇒ 00:17:37.789 Uttam Kumaran: Yeah. And you know what we’re thinking about. Scott is when we talk to Tim. We talked to him a little bit of how he’s provisioning. And he’s basically gonna do a Google group with all the people. And so we’re gonna use that as a basically a distro channel. So that one when new updates are coming out. Like, for example, we added a feature or some new knowledge, we can shout people out and and communicate that way. Additionally, I think, as part of some sort of weekly
189 00:17:38.069 ⇒ 00:17:40.799 Uttam Kumaran: thing around Andy, we should just publish
190 00:17:41.109 ⇒ 00:17:46.319 Uttam Kumaran: the who’s been using it. And just shout shout them out really publicly.
191 00:17:46.320 ⇒ 00:17:50.689 Scott_Harmon: Yeah, just produce a user digest, right? A Friday afternoon user digest with.
192 00:17:51.000 ⇒ 00:18:01.359 Scott_Harmon: you know, the people that are using it the most effectively. The people who created the coolest knowledge. And what’s cool about that is, people will see those new knowledge things
193 00:18:01.730 ⇒ 00:18:06.149 Scott_Harmon: and go. Oh, that’s an interesting topic. I might click on that right now, you know, like.
194 00:18:06.390 ⇒ 00:18:07.230 Scott_Harmon: yeah, for sure.
195 00:18:07.230 ⇒ 00:18:15.380 Scott_Harmon: You know, I’ve never had a phone call on it. But, huh! We’ve got a new skew for this, that, or the other thing. It’s a it’s a good way for people to learn when
196 00:18:15.910 ⇒ 00:18:18.450 Scott_Harmon: when they’re not on the phone
197 00:18:18.670 ⇒ 00:18:20.469 Scott_Harmon: with a gun to their head. So that’s.
198 00:18:20.470 ⇒ 00:18:21.100 YvetteRuiz: Yeah, yeah.
199 00:18:21.720 ⇒ 00:18:51.179 YvetteRuiz: Absolutely, I think. And and I and I did like the last week last Friday I did send an email. I copied you, Udem and amber to the entire team, letting them know. Hey, we successfully completed the world. Roll out with everyone. I shared the leaderboard. I shared what we updated. You know we got some positive responses, but I think you know to your point incentivizing them. You know. Because I thought it was interesting, and you may have caught that, you know, she kept on saying.
200 00:18:51.360 ⇒ 00:19:01.105 YvetteRuiz: y’all are spying on me. I was like, We’re we’re not spying. So again changing the to the why? Why is it important that we’re asking for usage of this.
201 00:19:01.420 ⇒ 00:19:02.970 Uttam Kumaran: Yes, yes, no. And
202 00:19:03.330 ⇒ 00:19:31.870 Uttam Kumaran: and it’s it’s really important that we consider that there are diff, that there are different types of users, and that the communication of the bot may be different, based on those people, but also the reason why they need to use it may be different. Some people are. They may have just lower usage because they’re pretty advanced, so most of their calls they don’t need help. Some of the junior folks like Joy. She’s using it. Every call, basically so we’re gonna have this mixed usage. So ultimately, maybe the barometer of usage.
203 00:19:31.910 ⇒ 00:20:01.098 Uttam Kumaran: I don’t know. We’ll have to think about what’s more impactful? Longer term. But for those, it’s really the narrative of hey giving the feedback back to the bot when it gets something wrong, helps everybody on the team, and also will eventually assist you when new new things are coming out, which at ABC. Is all the time your ability to learn those outside of the huddles and take advantage of those. The bot really streamlines versus them having to go recall. So I think we should just think of the the narrative piece a little bit
204 00:20:01.410 ⇒ 00:20:10.670 Uttam Kumaran: and think about the why for every person. But this is part of just adoption, and I think I’m glad we’re learning about it with this team, so that when we roll it out broader
205 00:20:10.750 ⇒ 00:20:12.550 Uttam Kumaran: it’s a lot smoother.
206 00:20:13.560 ⇒ 00:20:14.230 JanieceGarcia: Yeah.
207 00:20:14.230 ⇒ 00:20:36.900 YvetteRuiz: And I really like the cause of the other thing we talked about like when we are adding the other ones which we are, gonna add another group, which is our overflow people to the pest. Right? Because I think you’re gonna get a lot more usage and questions out of that. But the onboarding piece, when we send it like, what instructions are we sending, and what information are we sending, and what video are we sending them? So when we do that, I think that’s gonna be also impactful, that they get that.
208 00:20:37.600 ⇒ 00:20:38.390 Amber Lin: Oh.
209 00:20:39.260 ⇒ 00:21:07.909 Amber Lin: sounds good. Yeah. And and I know, put it down. I think it’ll be great to also mark different users by tenure, and then, we can work on creating different narratives for that. I think that’s something that isn’t blocked by any technic technical developments, but more of internal for this team to decide. Okay, what will motivate people the best. So we can definitely start doing that as well.
210 00:21:11.430 ⇒ 00:21:16.979 Amber Lin: Awesome. So here’s some good updates, examples that we did this week. So we were
211 00:21:17.540 ⇒ 00:21:22.759 Amber Lin: making sure that abbreviations are understood, especially for the more junior folks.
212 00:21:23.000 ⇒ 00:21:42.960 Amber Lin: we make sure that when it comes to keywords such as these, when it gets confusing that we do very specifically tell them what we do and what we don’t do, so that we make sure no business is lost, and we also explain we don’t do a certain certain business.
213 00:21:43.795 ⇒ 00:21:44.300 Amber Lin: and
214 00:21:44.300 ⇒ 00:22:02.230 Amber Lin: and this came up quite often. But this would be a great example of what an answer would look like. Right? So they have very specific instructions. It’s short, but it has the numbers, it has the prices, and it has a small oh, by the way, at the end.
215 00:22:03.720 ⇒ 00:22:15.729 Uttam Kumaran: Yeah. So people really appreciated that it had the Oh, by the way, cause one question I had was it, are we putting it like too often, and I think it’s actually helpful to have that. In fact, I think I think it was
216 00:22:16.110 ⇒ 00:22:17.849 Uttam Kumaran: maybe April. I forgot
217 00:22:17.960 ⇒ 00:22:22.810 Uttam Kumaran: who was just like, oh, actually I I totally should have said it, and it was right there. So then I was able to.
218 00:22:22.810 ⇒ 00:22:23.240 YvetteRuiz: I was able to.
219 00:22:23.240 ⇒ 00:22:25.256 Uttam Kumaran: Say, though, by the way,
220 00:22:25.660 ⇒ 00:22:26.140 Scott_Harmon: I’m gonna.
221 00:22:26.140 ⇒ 00:22:35.789 YvetteRuiz: And then Joy Joy said, That’s her bread and butter, I mean, that’s and and it’s great, because she’s a new person, and she’s building that habit from the beginning. She’s like, that’s yeah. Yeah.
222 00:22:35.790 ⇒ 00:22:39.849 Scott_Harmon: So this is absolutely awesome. I this
223 00:22:40.450 ⇒ 00:22:44.010 Scott_Harmon: so exciting. I’m now going to show my Ocd. Tendencies.
224 00:22:45.810 ⇒ 00:22:49.000 Scott_Harmon: Amber, would you please ask them to make the old.
225 00:22:49.000 ⇒ 00:22:50.640 Scott_Harmon: By the way, paragraph.
226 00:22:50.820 ⇒ 00:22:54.009 Amber Lin: That’s yes, that’s lousy.
227 00:22:54.010 ⇒ 00:23:00.900 Scott_Harmon: English. It’s lousy English. It’s a different topic. Longer paragraphs are more intimidating for people to read so.
228 00:23:00.900 ⇒ 00:23:01.510 Amber Lin: Oh!
229 00:23:01.660 ⇒ 00:23:05.110 Scott_Harmon: It’s so. It needs to be separated with a with a new line.
230 00:23:05.110 ⇒ 00:23:05.550 Uttam Kumaran: Yeah.
231 00:23:05.550 ⇒ 00:23:07.680 Scott_Harmon: That’s the biggest ever done.
232 00:23:07.990 ⇒ 00:23:08.610 Uttam Kumaran: Oh, yeah.
233 00:23:10.570 ⇒ 00:23:11.190 JanieceGarcia: Buddy.
234 00:23:11.480 ⇒ 00:23:12.759 Uttam Kumaran: No problem.
235 00:23:13.220 ⇒ 00:23:16.670 JanieceGarcia: And amber and utam. I did, actually.
236 00:23:16.670 ⇒ 00:23:33.600 JanieceGarcia: And the bee species event did give her stamp of approval. So I did send that to you guys, I didn’t want to just copy and paste it into the Central doc. So because of all the formatting and stuff that we talked about yesterday. So I did email it to the brain forge at go anteater. And then I Ccd, you.
237 00:23:33.600 ⇒ 00:23:55.939 Uttam Kumaran: Okay, yeah. And one thing I think on the improvements. And amber. I sent a note this morning is, I think we have. Janice is maintaining a document. We also have, I think, the document that Casey was working on. We also have another platform, so we don’t want to start to get into the struggle of having multiple documents as well. You know, that’s what we’re here to fix. So I think one thing is just consolidating as much to that core.
238 00:23:56.298 ⇒ 00:24:19.860 Uttam Kumaran: Google sheet as possible where we have not only documentation, on like, who’s using it? But we can also start to add the list of the issues that we’re tracking. And of course, on our side, we’re tracking it in linear, which is our project management tool. But just making sure that Janice will most likely not be the only person going forward providing suggestions. So I wanna make sure that all of those flow into our
239 00:24:20.020 ⇒ 00:24:42.349 Uttam Kumaran: project management flow, and then we can sort of take those off and then it’s just very, very easy for folks to be like. Here’s something that’s wrong, or here’s a piece of feedback like we want. We want to open the floodgates there and then. Our job is to triage and execute as fast as we can. So I just want to make it clear. If once we have folks in slack, potentially, they can add
240 00:24:42.580 ⇒ 00:24:53.849 Uttam Kumaran: stuff there and then for Janice, just making it clear where she can go to say, Hey, I was chatting with the bot. This thing is wrong. Here’s here’s why. Here’s why I think it’s a problem, and then our team can triage and take a look.
241 00:24:54.560 ⇒ 00:25:07.269 JanieceGarcia: Because what I was thinking after we had talked yesterday, too, with the bees, I actually thought about going ahead and putting it into the central doc, and then commenting and adding Amber and Casey, so that.
242 00:25:07.270 ⇒ 00:25:07.880 Uttam Kumaran: Yes.
243 00:25:07.880 ⇒ 00:25:17.879 JanieceGarcia: See right away. But I didn’t want to do. I didn’t want to do that until I knew it was okay for you guys, but I figured that would be a really good way to keep everything in one, Doc. And then also
244 00:25:18.060 ⇒ 00:25:20.760 JanieceGarcia: the back and forth. The triage.
245 00:25:20.920 ⇒ 00:25:49.489 Uttam Kumaran: I think that’s fine. I think the team now is gonna is figuring out sort of the best formatting and the trainer. But so I would rather we err on getting it into the document versus waiting another few days for us to like nail this process as long as it’s in the document. The AI will do a good job of picking it up. It’s just like, if it’s just formatted the right way, it may help it. Track it a little bit better and bring it up so, but if you can just add it there, don’t worry too much about the formatting now, and just comment us.
246 00:25:49.540 ⇒ 00:25:54.150 Uttam Kumaran: and then we’ll just make the cycle better. But yeah, thank you for that.
247 00:25:54.400 ⇒ 00:25:55.799 JanieceGarcia: Okay, perfect. Thank you.
248 00:25:58.200 ⇒ 00:26:08.790 Amber Lin: Sounds good. I’m writing all these down all these suggestions down in our backlog, so we can go look at it later. So
249 00:26:09.730 ⇒ 00:26:20.199 Amber Lin: for the formatting as we brought up. It’s a pretty big issue that we’re facing right now. This is part of our trainer bought development initiative, and I think for
250 00:26:20.540 ⇒ 00:26:46.360 Amber Lin: to get started with, I’ve just made this really simple document. I linked it in the bottom and I’ll send it in the email or individually, via chat. So just a simple guidelines of what’s good and what’s not so you guys can have the idea. And after that, what we want to do is we want to develop sort of an AI helper to help format these things, to make sure that
251 00:26:46.390 ⇒ 00:27:06.599 Amber Lin: we have a 1st pass on. Okay, is this good formatting before? Say, you have to reach the team because the team can’t always respond as fast as an AI chat or as Andy. That’s that was the whole point of us developing Andy. And after that we also want to have some auto suggested feedback
252 00:27:06.964 ⇒ 00:27:29.570 Amber Lin: updates. So remember the sheet that we had when the fee. When the Csrs give feedback, we want to eventually have it. Auto suggest, some updates based on that feedback that the trainers can then review and then add in. So in that case we’ll save a lot of time writing those feedbacks from scratch of writing those updates from scratch.
253 00:27:31.810 ⇒ 00:27:34.619 Amber Lin: This is our plan for the training part.
254 00:27:35.700 ⇒ 00:27:36.490 Scott_Harmon: So what is the.
255 00:27:36.490 ⇒ 00:27:36.990 YvetteRuiz: It’s really.
256 00:27:37.642 ⇒ 00:27:42.960 Scott_Harmon: Haven’t seen that trainer bots ui!
257 00:27:44.080 ⇒ 00:27:47.020 Scott_Harmon: What’s its? What’s its ui, Tom?
258 00:27:47.660 ⇒ 00:27:48.699 Scott_Harmon: Yeah, what’s it?
259 00:27:48.820 ⇒ 00:27:52.349 Scott_Harmon: Is it conversational? Is it? Interview that trainer like? What’s its?
260 00:27:52.640 ⇒ 00:27:54.169 Scott_Harmon: How does it interact?
261 00:27:54.300 ⇒ 00:28:00.229 Uttam Kumaran: Yeah. So my, our understanding right now is that we’re just gonna have it again in just in the Google
262 00:28:00.400 ⇒ 00:28:17.949 Uttam Kumaran: Workspace and then, ideally, you can use that to update the document or to reformat things, or basically it will. It will allow you, for example, when Janice wants to add something about bees go. The actual lift is not the fact that
263 00:28:18.180 ⇒ 00:28:35.410 Uttam Kumaran: she doesn’t know what to add. It’s actually the the difficulty is what structured at it. Where to add it? Am I covering all the bases, especially while that’s that one piece of thing is on her mind. She can close that out. So our goal for this is that it helps with the formatting. It helps for where to add it, and ideally it
264 00:28:35.540 ⇒ 00:28:38.999 Uttam Kumaran: it. You don’t really have to enter the central dock at all. You can
265 00:28:39.390 ⇒ 00:28:42.219 Uttam Kumaran: add that knowledge through the bot itself.
266 00:28:42.220 ⇒ 00:28:42.640 Scott_Harmon: So.
267 00:28:42.640 ⇒ 00:28:45.609 Uttam Kumaran: That would live as a work, as a chat bot in in Workspace.
268 00:28:45.610 ⇒ 00:28:52.370 Scott_Harmon: Okay. So maybe we could find some time to brainstorm on this a little bit like like
269 00:28:53.050 ⇒ 00:28:57.680 Scott_Harmon: I could see I could see 2 different approaches, and they’re they’re sort of like overlapped.
270 00:28:58.190 ⇒ 00:29:02.369 Scott_Harmon: One of them is where the the trainer thinks they’re saying.
271 00:29:02.790 ⇒ 00:29:08.770 Scott_Harmon: you know. Add this information, you know, to the central dock, and they’re they’re typing out the information
272 00:29:09.010 ⇒ 00:29:12.800 Scott_Harmon: in kind of a well formatted, you know, way.
273 00:29:14.990 ⇒ 00:29:19.430 Scott_Harmon: Another is where the bot is more interviewing them.
274 00:29:19.770 ⇒ 00:29:29.609 Scott_Harmon: and it’s just called an interview modality. So that kind of behavior would be like, I want to add some new knowledge, and the bot goes. Oh, what’s it about? It’s oh, it’s about a new.
275 00:29:29.780 ⇒ 00:29:40.519 Scott_Harmon: you know. It’s about a new pest service. Great. Okay? Why write a 1 sentence description first.st Okay, here’s the one sentence description. Great, right? Like it’s the bot knows what it needs
276 00:29:40.730 ⇒ 00:29:45.439 Scott_Harmon: and it it asks it? Asked the expert
277 00:29:45.600 ⇒ 00:29:51.280 Scott_Harmon: in little bursts, and then it goes off and updates the document. So
278 00:29:52.040 ⇒ 00:29:59.550 Scott_Harmon: and I guess we, you know, we might want to test it a little bit. One of them is where it’s basically just a smart text editor Tom. In other words.
279 00:29:59.550 ⇒ 00:30:00.230 Uttam Kumaran: Yeah.
280 00:30:00.230 ⇒ 00:30:12.589 Scott_Harmon: Here’s a blob of knowledge. I typed it out. Go make it the right format. So it looks right in the central dot. That’s 1 end. That’s just smart. A smart text editor, hey? This word didn’t make sense, or
281 00:30:13.200 ⇒ 00:30:15.560 Scott_Harmon: you know, the other one is.
282 00:30:15.690 ⇒ 00:30:21.519 Scott_Harmon: But the problem is that the the trainer needs to be able to write the whole thing out.
283 00:30:24.010 ⇒ 00:30:32.020 Scott_Harmon: The other one is that the bot is actually more actively interviewing the expert.
284 00:30:32.980 ⇒ 00:30:48.934 YvetteRuiz: It would be good to test them like you, said Scott. Because I could, you know, go both ways, I mean, because when that’s kind of what we’ve been doing with the B stuff, because we’re we now meet with the division managers on weekly basis and kind of say, Okay, what should it be? And we’re just kind of typing it all out.
285 00:30:49.390 ⇒ 00:31:01.169 Scott_Harmon: Yeah, I mean, it’s I don’t know which one’s right. I know that they but they’re just different right there in one. The I said, the Llm. Is more like I would just call it like a text editor, helper.
286 00:31:01.620 ⇒ 00:31:04.790 Scott_Harmon: To make sure a new knowledge is formatted properly.
287 00:31:04.920 ⇒ 00:31:07.930 Scott_Harmon: The other one is, it’s
288 00:31:08.960 ⇒ 00:31:18.449 Scott_Harmon: pulling exactly the right information out of the experts. Head, hey? To create this kind of note, you also need to tell me this. Yeah, you know. And so it understands.
289 00:31:20.380 ⇒ 00:31:25.390 Scott_Harmon: Everything that’s needed for that type of knowledge. Right? So
290 00:31:25.580 ⇒ 00:31:32.600 Scott_Harmon: I again, I don’t know the right answer. Maybe you could test them. But I do think you want to be intentional about that design choice with them.
291 00:31:33.184 ⇒ 00:31:37.330 Scott_Harmon: Cause. Those are a little bit different, you know. Paths.
292 00:31:37.530 ⇒ 00:31:41.819 Uttam Kumaran: Yeah, this is where I just my big takeaway from yesterday is that
293 00:31:42.320 ⇒ 00:32:08.821 Uttam Kumaran: it’s just to compress the feedback loops as much as possible. I think we we’re doing pretty well in that. I think it’s probably like a week end to end for us to even change something on the bot, or add new things. But I want to make it very, very easy for folks to identify what’s in that, doc, especially as it gets really big. And in like 48 h, be able to identify. There’s something missing. Find the source and then update it.
294 00:32:09.510 ⇒ 00:32:17.110 Uttam Kumaran: you know. And so so it ends up it. You basically like are updating the back end within just 48 h. I think this spot.
295 00:32:17.820 ⇒ 00:32:30.076 Uttam Kumaran: this bot, is really that, like what helps the second piece and ideally can assist folks like Janice or or trainers to to go update the knowledge. You know, way faster.
296 00:32:32.100 ⇒ 00:32:36.980 Scott_Harmon: So just to be clear what you and this could be right. I’m just trying to be explicit.
297 00:32:36.980 ⇒ 00:32:37.550 Uttam Kumaran: Yes.
298 00:32:37.550 ⇒ 00:32:46.339 Scott_Harmon: The design of the bot assumes that the trainer understands all the knowledge that’s needed.
299 00:32:46.670 ⇒ 00:32:49.450 Scott_Harmon: and can write that all out in one. Go.
300 00:32:50.270 ⇒ 00:32:52.330 Uttam Kumaran: So, yeah, so my feedback.
301 00:32:52.330 ⇒ 00:32:55.029 Scott_Harmon: It’s a little bit. It’s a little bit risky just to be clear.
302 00:32:55.030 ⇒ 00:33:00.719 Uttam Kumaran: So, my yeah, my feedback for the team is that this trainer bot should push back basically
303 00:33:01.410 ⇒ 00:33:08.298 Uttam Kumaran: have some set of requirements for like what it needs before it makes an update.
304 00:33:08.740 ⇒ 00:33:09.669 Scott_Harmon: The kind of stuff I’ve talked.
305 00:33:09.670 ⇒ 00:33:27.700 Uttam Kumaran: Sorry if I missed. Sorry if I missed this today, but it’s more dynamic in that way where you know. One of the questions yesterday was. And this is something maybe we were gonna test with with Denise on like, what? How even strict the Andy should be, because sometimes people just say, mosquito help. And you’re gonna get
306 00:33:27.870 ⇒ 00:33:45.880 Uttam Kumaran: just as poor of a response out if that’s the input, right? So at how strict should that be? On saying, well, I like, I need more context. But here’s a couple of things, and the but the trainer, but for sure is we want to have a very high standard for what ends up in
307 00:33:46.050 ⇒ 00:33:49.349 Uttam Kumaran: the back end. Otherwise it’s gonna degrade everything. So.
308 00:33:49.350 ⇒ 00:33:53.540 Scott_Harmon: Thumbs up for me. Right? You’ve so. However you do it.
309 00:33:53.740 ⇒ 00:34:02.729 Scott_Harmon: I think that the agent needs to push back or ask for more context or go wait a minute. I don’t quite understand yet, or whatever, so that the knowledge is
310 00:34:02.860 ⇒ 00:34:04.420 Scott_Harmon: really thorough.
311 00:34:04.830 ⇒ 00:34:05.450 Uttam Kumaran: Okay.
312 00:34:05.948 ⇒ 00:34:13.760 Scott_Harmon: And if it’s if it’s just taking and formatting text and not kind of
313 00:34:13.920 ⇒ 00:34:17.519 Scott_Harmon: checking for completeness, I think we’re gonna miss.
314 00:34:18.480 ⇒ 00:34:21.940 Scott_Harmon: You know you’re going to get a garbage in garbage out problem.
315 00:34:22.590 ⇒ 00:34:23.569 Scott_Harmon: Right, where?
316 00:34:23.780 ⇒ 00:34:24.370 Uttam Kumaran: Exactly.
317 00:34:24.370 ⇒ 00:34:39.059 Scott_Harmon: Denise. No offense, Denise. But oh, yeah, I forgot to mention. This only works on 3 storey houses, you know, whatever it is, right. Well, now, you’ve got knowledge in the knowledge document that doesn’t have all the necessary context. And
318 00:34:39.190 ⇒ 00:34:43.110 Scott_Harmon: you’ve created a a little bit of a knowledge rot problem.
319 00:34:43.110 ⇒ 00:34:46.599 Scott_Harmon: So. Yes, the trainer bot can say
320 00:34:46.749 ⇒ 00:34:56.529 Scott_Harmon: in a very nice way, hey, you need to tell me this. Does this work for 3 story houses? And Janice like, Oh, yeah, yes, of course, you know, like, so now the knowledge is is comprehensive.
321 00:34:56.869 ⇒ 00:35:04.039 Scott_Harmon: and you know, so I think we should have a fairly good publishing standard is what I’m saying.
322 00:35:04.429 ⇒ 00:35:07.889 Scott_Harmon: where the bot is making sure that we’ve got everything we need
323 00:35:08.269 ⇒ 00:35:09.909 Scott_Harmon: for a given bit of knowledge.
324 00:35:12.680 ⇒ 00:35:14.950 JanieceGarcia: I’d agree with what
325 00:35:15.270 ⇒ 00:35:19.799 JanieceGarcia: you’re saying, because we have to be careful. And Yvette and I have seen it already with.
326 00:35:19.910 ⇒ 00:35:22.010 JanieceGarcia: you know certain things.
327 00:35:22.220 ⇒ 00:35:24.409 JanieceGarcia: I’ll take the bees, for example.
328 00:35:24.800 ⇒ 00:35:31.450 JanieceGarcia: They are covered under pest control, but it’s got to be visible, and it’s got to be on the 1st story eaves.
329 00:35:31.650 ⇒ 00:35:54.475 JanieceGarcia: It’s gonna be additional if it’s on the second story. So those are things that we do have to think about. And we have to be very granular when it comes to our treatments, or you know, carpenter ants, they are covered inside the home because it’s under general pest inside the home. But anything on the exterior or on trees that’s going to be a 1 time. So those are things that we do have to think about when it comes to
330 00:35:55.110 ⇒ 00:35:55.800 JanieceGarcia: The trainer.
331 00:35:55.800 ⇒ 00:35:59.539 Scott_Harmon: If you could just pass that on to the designers who tell them, you know, and just.
332 00:35:59.540 ⇒ 00:36:00.160 Uttam Kumaran: Yeah.
333 00:36:00.160 ⇒ 00:36:12.259 Scott_Harmon: I think I think we’re all now agreeing that we want the trainer bot to be to really play an active role. I like that word dynamic when you’re creating new knowledge to make sure it’s complete, or to gather as much.
334 00:36:12.260 ⇒ 00:36:14.540 Uttam Kumaran: It should be pretty strict. Yeah.
335 00:36:14.540 ⇒ 00:36:20.010 Scott_Harmon: Yeah, I think that’s right. And so how you do that is kind of an implementation detail. I know you can figure that out. But
336 00:36:20.130 ⇒ 00:36:24.679 Scott_Harmon: I I think it needs to be better than just take whatever text is given.
337 00:36:25.070 ⇒ 00:36:26.669 Scott_Harmon: and drop it into the dock.
338 00:36:26.670 ⇒ 00:36:27.320 Uttam Kumaran: Yeah.
339 00:36:27.320 ⇒ 00:36:29.430 Scott_Harmon: In a in a nice format like that.
340 00:36:30.450 ⇒ 00:36:31.990 Scott_Harmon: That’s not.
341 00:36:31.990 ⇒ 00:36:38.130 Uttam Kumaran: I think, amber part of this formatting guide. We should just expand this into a broader guide about
342 00:36:38.990 ⇒ 00:36:47.043 Uttam Kumaran: about what we require, based on certain topics, and you could probably use AI to help you devise the instruction set. But
343 00:36:47.710 ⇒ 00:36:50.740 Uttam Kumaran: I think you kind of see the yeah. See the vision.
344 00:36:51.920 ⇒ 00:36:52.330 Amber Lin: Whoa!
345 00:36:52.330 ⇒ 00:36:53.280 Scott_Harmon: That makes sense amber.
346 00:36:53.280 ⇒ 00:36:54.080 Amber Lin: Okay.
347 00:36:55.080 ⇒ 00:37:03.380 Amber Lin: yeah, I’ve wrote that down since to me, I’m gonna communicate to the sea. Okay, which part can we do? First? st
348 00:37:03.984 ⇒ 00:37:13.280 Amber Lin: To have some have some benefits for the trainers immediately, and then what part takes more time to develop, develop, and how we’re.
349 00:37:13.280 ⇒ 00:37:13.640 Scott_Harmon: I don’t.
350 00:37:13.640 ⇒ 00:37:14.260 Amber Lin: Feedback.
351 00:37:14.260 ⇒ 00:37:15.630 Scott_Harmon: I’ve got a couple of examples.
352 00:37:15.630 ⇒ 00:37:16.400 Scott_Harmon: Formatting is good.
353 00:37:16.400 ⇒ 00:37:22.730 Scott_Harmon: I got a couple of examples from other domains of agents. That kind of behave like that that I’ll send if they’ll
354 00:37:23.310 ⇒ 00:37:26.789 Scott_Harmon: help, you know. Stimulate thinking for your designer.
355 00:37:26.930 ⇒ 00:37:27.920 Uttam Kumaran: Yes. Yeah.
356 00:37:38.810 ⇒ 00:37:39.570 Amber Lin: so
357 00:37:41.240 ⇒ 00:37:53.440 Amber Lin: I’ve noted all these down a lot of tickets. Now, i’ll groom through those, and i’ll I’ll talk to the team. I’ll get back to you guys with a more fleshed out plan.
358 00:37:59.310 ⇒ 00:38:00.230 Uttam Kumaran: Okay. Cool.
359 00:38:07.410 ⇒ 00:38:09.850 Amber Lin: Oh, my network is not the best.
360 00:38:10.980 ⇒ 00:38:12.454 Uttam Kumaran: We can still hear you, but
361 00:38:12.700 ⇒ 00:38:14.080 Scott_Harmon: You’re good, you’re good.
362 00:38:14.080 ⇒ 00:38:16.690 Amber Lin: Oh, okay, I see.
363 00:38:16.800 ⇒ 00:38:19.360 Amber Lin: Sounds good. Anything else.
364 00:38:21.770 ⇒ 00:38:25.989 Amber Lin: I know you were thinking about the phase 2 phase, 2 stuff.
365 00:38:25.990 ⇒ 00:38:49.891 Uttam Kumaran: Yeah, I think I’ll connect with Scott. I think we’re. I think we’re pretty set there. I think the only pieces on these meetings, amber moving forward. I think we want to, either on this meeting, or I want to do some other sort of business review meeting with Yvette in particular. Where we review. We just go very deep on metrics. I think it’d be great if Brian’s on there. It’d be great if Annie’s on there.
366 00:38:50.400 ⇒ 00:39:04.095 Uttam Kumaran: I feel like every 2 weeks seems like a good cadence, and maybe for the weeks off, we can do something that’s just more of an email. But I want to basically go through the go through the dashboard in detail.
367 00:39:04.570 ⇒ 00:39:13.110 Uttam Kumaran: not only the now that we have the overall call side, we can review that. And then, ideally, I think Brian could probably lead that. But we’re looking at
368 00:39:13.652 ⇒ 00:39:23.959 Uttam Kumaran: the calls where that bot was used. I want to look at. You know what was the average, you know, handling time of those who’s using it, what our eval scores were.
369 00:39:24.479 ⇒ 00:39:31.599 Uttam Kumaran: What like, what category of calls they were. I think we want to start going deep. I think probably another
370 00:39:31.860 ⇒ 00:39:36.629 Uttam Kumaran: meeting is is more required just to do that. And I think probably Annie on our side can
371 00:39:36.810 ⇒ 00:39:43.260 Uttam Kumaran: can lead that. But that’s that, was the only call out, and I’ll I have the notes from yesterday, so I’ll share a couple more things.
372 00:39:43.731 ⇒ 00:39:50.769 Uttam Kumaran: But would love to do that. I think Yvette, in particular, is very excited to use the dashboard more. We walked through it several times.
373 00:39:51.470 ⇒ 00:40:00.850 Uttam Kumaran: It’s just, you know, it’s a technical tool. So I want to make sure that they feel comfortable digging into it themselves. But even yesterday, for example, we looked at the thumb.
374 00:40:01.260 ⇒ 00:40:02.100 Uttam Kumaran: the
375 00:40:02.240 ⇒ 00:40:12.249 Uttam Kumaran: thumbs up, thumbs down by like department and like, there’s a lot of really great stuff in there that previously didn’t exist. So I just wanna make sure that that gets that gets adopted.
376 00:40:13.530 ⇒ 00:40:14.320 YvetteRuiz: Thank you.
377 00:40:14.320 ⇒ 00:40:14.870 Scott_Harmon: Sounds, like.
378 00:40:18.330 ⇒ 00:40:19.290 YvetteRuiz: Go ahead!
379 00:40:19.690 ⇒ 00:40:25.200 Scott_Harmon: I. I just have one question, what is the in the calendar? What’s the rollout date for the trainer? Bot.
380 00:40:29.420 ⇒ 00:40:31.520 Uttam Kumaran: Yeah, amber, maybe question.
381 00:40:32.053 ⇒ 00:40:51.659 Amber Lin: Depending on the features that we want to include. Right? So if we we can, I do you think we roll out features by features, so I want people to have been benefits from it as soon as possible, so the formatting will come pretty quickly, and then the one where it needs to
382 00:40:51.810 ⇒ 00:41:09.440 Amber Lin: specifically insert the updates into central will take a bit more time. But I do think that formatting incorrectly, having it ask questions to make sure that things are up to standard that will come a bit faster.
383 00:41:09.770 ⇒ 00:41:12.690 Amber Lin: So I would say it would take like a week or 2.
384 00:41:14.330 ⇒ 00:41:20.319 Uttam Kumaran: Yeah, I think I would. I would think it’s at least it’s at Max 2 weeks, but I think we should have something next week.
385 00:41:21.110 ⇒ 00:41:22.970 Scott_Harmon: Okay, I yeah, like, I say, I,
386 00:41:24.100 ⇒ 00:41:27.635 Scott_Harmon: maybe we could just do something internal, a little checkpoint on that.
387 00:41:28.530 ⇒ 00:41:31.709 Scott_Harmon: I just feel like a little bit of alignment now would really
388 00:41:32.160 ⇒ 00:41:35.389 Scott_Harmon: help save 2 or 3 weeks of rework on this thing.
389 00:41:35.730 ⇒ 00:41:39.066 Uttam Kumaran: Yeah, maybe, Amber, we can do a meeting on Wednesday.
390 00:41:40.470 ⇒ 00:41:44.590 Uttam Kumaran: let’s let’s just maybe just do an internal meeting on Wednesday, and you can add Scott just.
391 00:41:45.210 ⇒ 00:41:56.100 Uttam Kumaran: or if or if we want to do something even earlier in the week, that we can just look at what we’re planning on building. I agree. I just, I want to make some of these decisions upfront versus like seeing it come out later.
392 00:41:56.100 ⇒ 00:41:59.780 Scott_Harmon: And rework. Just a just a real 30 min is probably all we need.
393 00:41:59.780 ⇒ 00:42:00.200 Uttam Kumaran: Yeah.
394 00:42:00.460 ⇒ 00:42:01.500 Scott_Harmon: Just to get tight.
395 00:42:01.500 ⇒ 00:42:05.139 Amber Lin: Cool depends on where you’re free. Scott.
396 00:42:05.500 ⇒ 00:42:10.850 Scott_Harmon: Yeah, yeah, I’ve my schedule is very flexible except for Wednesday morning.
397 00:42:11.190 ⇒ 00:42:11.990 Amber Lin: Okay.
398 00:42:12.220 ⇒ 00:42:13.260 Scott_Harmon: I’m in a golf tournament.
399 00:42:13.514 ⇒ 00:42:14.529 Amber Lin: We can meet you.
400 00:42:14.530 ⇒ 00:42:17.811 Scott_Harmon: Tournament, and I lose money every week in this golf tournament.
401 00:42:18.800 ⇒ 00:42:22.960 Scott_Harmon: I’m gonna win money this week because I’ve started cheat.
402 00:42:24.020 ⇒ 00:42:25.460 YvetteRuiz: It’s very cheap.
403 00:42:25.894 ⇒ 00:42:34.575 YvetteRuiz: There’s a lot of golf tournaments going on next week. I know the exact teams have a tournament on Monday.
404 00:42:38.220 ⇒ 00:42:39.564 YvetteRuiz: Well, good luck, Scott.
405 00:42:39.900 ⇒ 00:42:40.660 Scott_Harmon: Thank you.
406 00:42:43.670 ⇒ 00:42:44.370 Amber Lin: Yeah.
407 00:42:45.070 ⇒ 00:43:00.709 Amber Lin: Okay. So I think there’s 2 calls. I would book. I will book a call probably would not before Wednesday, so that we can have the entire week to make sure that we’re developing on the right path, and then I’ll also book a call with you. If that for the dashboard.
408 00:43:01.910 ⇒ 00:43:02.470 Uttam Kumaran: Cool.
409 00:43:02.470 ⇒ 00:43:13.520 Uttam Kumaran: and then and then I’ll I’ll follow up amber. We can talk later today on, just like all the download from yesterday’s session. And then we can put together a little bit of a plan towards.
410 00:43:13.740 ⇒ 00:43:18.940 Uttam Kumaran: If we’re trying to do slack, or if we’re gonna do office hours like we’ll we’ll sort of start to think about that next week.
411 00:43:20.210 ⇒ 00:43:20.760 Amber Lin: Yeah.
412 00:43:20.760 ⇒ 00:43:22.200 YvetteRuiz: We’ll work on our part.
413 00:43:22.330 ⇒ 00:43:44.560 YvetteRuiz: We’ll work on our part to start, you know, getting the engagement, checking in with the agents and getting that going and stuff, and what incentives we’re going to be given for. You know the idea, Scott, throughout, and stuff like that. But we kind of have an idea already what? What? That’s gonna look like. So yeah. But overall, I mean, guys again. Thanks. I think this is great. I’m having you there yesterday was just awesome.
414 00:43:44.560 ⇒ 00:44:03.839 Uttam Kumaran: It was great. No, it’s amazing to see people talking about it, and like being able to walk the halls and and see folks using it. And again, I don’t I for me the the motive when I see people saying like, Oh, I don’t want to use it. That’s just more motivation right? That’s more work for us to do. So. I don’t get nervous about that. I’m just excited that
415 00:44:04.050 ⇒ 00:44:12.490 Uttam Kumaran: even with the limited training stuff we’ve done, some people are very excited about it, so it just goes to show if we take a couple more steps in that direction, I think we’ll be just fine, so.
416 00:44:12.920 ⇒ 00:44:13.550 YvetteRuiz: Yeah.
417 00:44:13.600 ⇒ 00:44:31.089 YvetteRuiz: again. Thanks, guys. And I did share with Udam, I mean, on Tuesday we made every we have our executive meeting. Bobby. You know he’s pretty excited about it as well, which is, which is good. And I think the Dms are we’re able to identify because one of the things that we we
418 00:44:31.090 ⇒ 00:44:45.606 YvetteRuiz: figured out in our Tuesday executive meeting was the customer referral program. And again, these topics we’re not having. If we’re not getting the feedback from them. And those are the big gaps that we know that we’ve been missing and stuff. So lots of good stuff. So thank you, guys overall.
419 00:44:46.350 ⇒ 00:44:47.999 Scott_Harmon: Great progress. Both teams.
420 00:44:48.850 ⇒ 00:44:51.090 Uttam Kumaran: Thanks. Everyone have a great weekend.
421 00:44:51.090 ⇒ 00:44:52.420 JanieceGarcia: Again have a good one.
422 00:44:52.600 ⇒ 00:44:54.600 Uttam Kumaran: Bye, bye.