Meeting Title: Henry - Hannah - Amplitude Case Study Date: 2025-10-08 Meeting participants: Hannah Wang, Henry Zhao
WEBVTT
1 00:01:55.740 ⇒ 00:01:58.910 Hannah Wang: Hey, I think you’re muted.
2 00:01:58.910 ⇒ 00:02:00.100 Henry Zhao: Hey, Hannah, how’s it going?
3 00:02:00.430 ⇒ 00:02:01.210 Hannah Wang: Good.
4 00:02:01.910 ⇒ 00:02:06.610 Hannah Wang: Give me one second.
5 00:02:09.850 ⇒ 00:02:11.319 Henry Zhao: Where are you based out of, by the way?
6 00:02:11.740 ⇒ 00:02:12.540 Hannah Wang: Huh?
7 00:02:12.910 ⇒ 00:02:14.059 Henry Zhao: Where are you based out of?
8 00:02:14.060 ⇒ 00:02:14.920 Hannah Wang: LA.
9 00:02:15.370 ⇒ 00:02:16.240 Henry Zhao: Oh, cool.
10 00:02:16.240 ⇒ 00:02:17.839 Hannah Wang: Yeah, where are you right now?
11 00:02:18.170 ⇒ 00:02:19.060 Henry Zhao: Arizona.
12 00:02:19.310 ⇒ 00:02:20.720 Hannah Wang: I feel like you’re kind of.
13 00:02:20.720 ⇒ 00:02:21.460 Henry Zhao: Very close.
14 00:02:21.460 ⇒ 00:02:25.650 Hannah Wang: all over the… or, like, weren’t you in New York for a little bit, and then, like, in…
15 00:02:25.840 ⇒ 00:02:28.699 Hannah Wang: Some other country for a little bit at this…
16 00:02:28.700 ⇒ 00:02:34.329 Henry Zhao: Yes, so my family is in New York, so whenever I visit them, I… meet up with Robert.
17 00:02:35.780 ⇒ 00:02:39.230 Hannah Wang: But your home, or, like, you live… reside in Arizona?
18 00:02:39.230 ⇒ 00:02:39.930 Henry Zhao: Yeah,
19 00:02:39.930 ⇒ 00:02:44.520 Hannah Wang: Okay. Oh, nice. I thought you were on the East Coast, so it’s nice having someone…
20 00:02:45.150 ⇒ 00:02:48.430 Hannah Wang: Kind of closer to my time zone.
21 00:02:48.430 ⇒ 00:02:51.490 Henry Zhao: Yeah, but I’m kind of just, like, 24-7, I’m kind of just on call all the time.
22 00:02:51.890 ⇒ 00:02:54.690 Hannah Wang: That’s… that’s no way to live.
23 00:02:55.090 ⇒ 00:02:55.560 Henry Zhao: I like it.
24 00:02:55.560 ⇒ 00:02:56.629 Hannah Wang: Cool. Oh, okay.
25 00:02:56.630 ⇒ 00:02:58.650 Henry Zhao: No, I think it’s less stressful, actually.
26 00:02:59.020 ⇒ 00:03:02.899 Henry Zhao: Because when I’m, like, set hours, I have, like, procrastinate so much.
27 00:03:04.040 ⇒ 00:03:05.010 Henry Zhao: Yeah, so…
28 00:03:05.010 ⇒ 00:03:07.780 Hannah Wang: I see. Different working personalities.
29 00:03:07.780 ⇒ 00:03:08.380 Henry Zhao: Yeah, exactly.
30 00:03:08.380 ⇒ 00:03:19.540 Hannah Wang: Okay, so with the case studies, this is the first time that we’ve done this, so I might ask for a couple moving forward. So just give you the lay of the land.
31 00:03:20.460 ⇒ 00:03:44.069 Hannah Wang: basically ask you a bunch of questions, and I take the transcript from this meeting, create it to AI, and then I build a case study out of it, and then afterwards, I ask whoever I interviewed, and also, like, data or AI team to also look at it. So even… so for the questions, even if the answer’s super obvious, you can just
32 00:03:44.600 ⇒ 00:03:51.009 Hannah Wang: say things out loud so it’s captured in the transcript, and then that’ll be good. So…
33 00:03:51.300 ⇒ 00:03:57.890 Hannah Wang: Okay. I know we are doing Amplitude today, I don’t know…
34 00:03:58.270 ⇒ 00:04:07.060 Hannah Wang: Yeah, Utena was just saying, oh, a general amplitude case study. I don’t know what client… I don’t… I don’t think it’s client-specific,
35 00:04:07.240 ⇒ 00:04:13.380 Hannah Wang: So I don’t really know how to build out this case study, because all our previous ones were for
36 00:04:13.960 ⇒ 00:04:17.570 Hannah Wang: particular clients, even if it was anonymized.
37 00:04:17.570 ⇒ 00:04:18.170 Henry Zhao: Okay.
38 00:04:18.170 ⇒ 00:04:21.119 Hannah Wang: So I don’t know if you had anything in mind for…
39 00:04:21.380 ⇒ 00:04:23.159 Henry Zhao: Just walking me through.
40 00:04:23.410 ⇒ 00:04:28.660 Hannah Wang: like… Something with amplitude for a case study.
41 00:04:28.660 ⇒ 00:04:29.680 Henry Zhao: Yeah, sure.
42 00:04:29.680 ⇒ 00:04:34.499 Hannah Wang: Okay, so I’m gonna start… I’m gonna pull up my list of questions.
43 00:04:35.380 ⇒ 00:04:38.020 Henry Zhao: Should I keep this more broad or more general?
44 00:04:39.150 ⇒ 00:04:41.209 Henry Zhao: Sorry, more broader, more specific.
45 00:04:41.210 ⇒ 00:04:43.800 Hannah Wang: I got you.
46 00:04:46.270 ⇒ 00:04:59.189 Henry Zhao: Because the thing is, Amplitude is, like, one of the tools that does product analytics, but there’s also Mixpanel, and I know some companies have MixPanel, so I don’t want… I don’t know if I should tailor this to either-or, or people that are just specifically using Amplitude.
47 00:04:59.190 ⇒ 00:05:01.460 Hannah Wang: You should do amplitude, because next cycle.
48 00:05:01.820 ⇒ 00:05:05.680 Hannah Wang: panel slotted, so it might be you, as well, so…
49 00:05:05.680 ⇒ 00:05:09.820 Henry Zhao: It’s gonna be the same thing, except for… yeah, it’s gonna be the same thing, except for Mixpanel instead of Amplitude.
50 00:05:09.820 ⇒ 00:05:11.210 Hannah Wang: Yeah, totally.
51 00:05:11.210 ⇒ 00:05:12.040 Henry Zhao: Yeah, same thing.
52 00:05:12.280 ⇒ 00:05:29.239 Hannah Wang: Okay, so I guess the first question… I don’t know if… I think you’ve taken a look at our case studies, but we have, like, an at-a-glance section, so I guess my first question is, what is the project type? I don’t know if… yeah, sure. Like, what is the project type? Yeah.
53 00:05:29.240 ⇒ 00:05:38.530 Henry Zhao: Yeah, so it’s product analytics. It’s basically analyzing how users of a specific product… usually it’s digital, right, because these things are implemented on the web.
54 00:05:38.740 ⇒ 00:05:46.120 Henry Zhao: How they interact with the product, whether it’s specific tabs, specific buttons, or specific, features.
55 00:05:46.360 ⇒ 00:05:52.410 Henry Zhao: Amplitude logs these events specifically, and then looks at where do people drop off, where they shouldn’t be dropping off.
56 00:05:52.680 ⇒ 00:05:56.020 Henry Zhao: Where do they maybe have some confusion?
57 00:05:56.820 ⇒ 00:06:00.280 Henry Zhao: And then…
58 00:06:01.560 ⇒ 00:06:17.060 Henry Zhao: there’s also a lot of, like, converting from free to paid plans, right? Like, so, what do people that convert from free to paid plans do? Like, what are the features they find useful? What are the ways that they’re using the product? So a lot of it is just very product analytics focused. Okay, cool.
59 00:06:17.060 ⇒ 00:06:25.030 Hannah Wang: My next question is, what’s the project duration? But if there’s no client, then I guess you can’t…
60 00:06:25.510 ⇒ 00:06:26.299 Hannah Wang: Or if there’s no…
61 00:06:26.300 ⇒ 00:06:43.949 Henry Zhao: Yeah, if it’s, like, an app or a website that’s, like, one page, it might be, like, three weeks, very, very simple. But if it’s, like, a suite of tools where there’s a bunch of different apps, different pages, different types of products, different ways you can interact with the product, it might be three to five months, I would say.
62 00:06:43.950 ⇒ 00:06:44.720 Hannah Wang: Okay.
63 00:06:44.970 ⇒ 00:06:45.520 Henry Zhao: Yeah.
64 00:06:45.520 ⇒ 00:06:53.879 Hannah Wang: And then, usually, who are the team members involved? I know there’s, like, a PM, obviously, and then a data analyst,
65 00:06:54.110 ⇒ 00:06:54.890 Hannah Wang: And then…
66 00:06:54.890 ⇒ 00:07:06.310 Henry Zhao: I would say the… I would say the big piece is the engineer that can implement Amplitude. So basically, they need to add a piece of JavaScript to every event that they want fired, and usually that’s the bottleneck, so…
67 00:07:07.370 ⇒ 00:07:19.040 Henry Zhao: This is something I talked to Robert and Utam about, is like, do we want to provide this as a service? Because that’s usually the bottleneck, where companies don’t have enough resources to do that. So do we want to provide that and add that to our suite of products?
68 00:07:19.170 ⇒ 00:07:23.019 Henry Zhao: So basically, right, if you have a website or a product.
69 00:07:23.290 ⇒ 00:07:36.450 Henry Zhao: People are using it. Whenever an action is done, you need to fire, like, something too amplitude to log that and say, the person did this, this is the person, this is their traits, whatever country they’re from, what device they’re using, etc.
70 00:07:36.620 ⇒ 00:07:52.030 Henry Zhao: That all needs to be programmed in, right? So that’s usually the bottleneck. Once the data actually flows into Amplitude, the analysis is pretty easy. We just need to make sure that it is standardized, that the data is clean, that it’s flowing in as we expected, like, there’s no duplication.
71 00:07:52.350 ⇒ 00:07:54.850 Henry Zhao: And then we can just build charts, and it should be pretty easy.
72 00:07:55.470 ⇒ 00:08:03.690 Hannah Wang: So, is Ampli… like, I… I kind of know, like, so is Post Hoc a… is it a similar tool to Amplitude?
73 00:08:03.690 ⇒ 00:08:09.009 Henry Zhao: I think so. I’ve actually heard about PostHog, I’ve never looked into it.
74 00:08:10.470 ⇒ 00:08:13.780 Hannah Wang: Cause it sounds like… Famil… it just sounds like, oh yeah.
75 00:08:13.780 ⇒ 00:08:14.549 Henry Zhao: I think so.
76 00:08:14.710 ⇒ 00:08:15.460 Hannah Wang: No? Okay.
77 00:08:15.460 ⇒ 00:08:16.540 Henry Zhao: Yeah.
78 00:08:16.900 ⇒ 00:08:21.970 Henry Zhao: Because I have heard of post hoc, but yeah, I don’t think it’s… Something.
79 00:08:22.180 ⇒ 00:08:31.409 Hannah Wang: Okay. Cool. I think you are cutting in and out, so I’m gonna wait.
80 00:08:31.410 ⇒ 00:08:32.039 Henry Zhao: Hmm?
81 00:08:32.730 ⇒ 00:08:33.429 Henry Zhao: Nope.
82 00:08:33.860 ⇒ 00:08:39.569 Hannah Wang: Yeah, your video’s frozen, but I can kind of hear your audio, so I…
83 00:08:40.250 ⇒ 00:08:45.970 Hannah Wang: Don’t know if it’s my internet or your internet.
84 00:08:45.970 ⇒ 00:08:47.720 Henry Zhao: Can you see me?
85 00:08:48.550 ⇒ 00:08:52.940 Hannah Wang: I can’t see you, but I can hear you, so I… That’s mine. Yeah, okay.
86 00:08:54.750 ⇒ 00:09:00.210 Hannah Wang: Okay, no worries. Cool. So, the next thing is…
87 00:09:00.710 ⇒ 00:09:17.100 Hannah Wang: I ask about the context, so usually what’s, like… yeah, sure, camera off is good. What’s, like, the working environment of these clients before we start trying to use Amplitude to help them fix their stuff?
88 00:09:17.510 ⇒ 00:09:26.779 Henry Zhao: Yeah, so it’s basically, usually it’s like, they rolled out a bunch of new features, and they’re like, we don’t know if people are using this, we don’t know if they’re using it in the right way,
89 00:09:27.180 ⇒ 00:09:36.970 Henry Zhao: We don’t know, like, what are the most useful features, what are the things that people find most valuable, and then I think product analytics can really help us define that and find some data to back that up.
90 00:09:39.140 ⇒ 00:09:41.040 Henry Zhao: Got it.
91 00:09:42.400 ⇒ 00:09:50.779 Hannah Wang: What are, like, specific limitations or constraints that clients usually have that make it hard for them to do? Is it just lack of people?
92 00:09:50.780 ⇒ 00:09:51.609 Henry Zhao: and young.
93 00:09:51.610 ⇒ 00:09:54.049 Hannah Wang: Okay, lack of engineering.
94 00:09:54.050 ⇒ 00:09:56.470 Henry Zhao: Yeah, I would say that’s pretty much it.
95 00:09:57.170 ⇒ 00:09:59.150 Henry Zhao: Okay. Yeah.
96 00:09:59.150 ⇒ 00:10:01.720 Hannah Wang: And then, usually, like.
97 00:10:01.720 ⇒ 00:10:02.740 Henry Zhao: Basically, you know.
98 00:10:02.740 ⇒ 00:10:11.350 Hannah Wang: Do our clients try to do it themselves before asking for help? Like, are there any previous efforts to solve.
99 00:10:11.350 ⇒ 00:10:12.620 Henry Zhao: light on that yet?
100 00:10:12.620 ⇒ 00:10:13.690 Hannah Wang: Of, like, not knowing.
101 00:10:13.690 ⇒ 00:10:28.310 Henry Zhao: Yes, okay, so I would say one thing I think a lot of companies do is they try to do it in Google Analytics, but that’s really different, right? Google Analytics, I think, is meant towards more, like, optimizing your Google campaigns, so it’s, like, very Google-focused, it’s very…
102 00:10:28.410 ⇒ 00:10:37.069 Henry Zhao: I would say biased, maybe, is a good word for it, but if you want unbiased, like, actual, like, own your own data, you should probably use Amplitude.
103 00:10:37.500 ⇒ 00:10:38.880 Hannah Wang: Okay. I actually.
104 00:10:38.880 ⇒ 00:10:40.220 Henry Zhao: Yeah.
105 00:10:40.220 ⇒ 00:10:40.750 Hannah Wang: Go ahead.
106 00:10:41.200 ⇒ 00:10:48.560 Henry Zhao: Yeah, you’ll see that people that implement both, they’re like, the data doesn’t match, but that’s, like, that’s the reason, right? Because Google wants to do what’s best for them, and then…
107 00:10:48.710 ⇒ 00:10:51.360 Henry Zhao: Amplitude is just, like, sending you the data purely.
108 00:10:51.360 ⇒ 00:10:54.229 Hannah Wang: I see, so it’s more agnostic. Okay.
109 00:10:54.740 ⇒ 00:11:12.429 Hannah Wang: Yeah, Google Analytics, I… so I’m trying to get… I don’t know if you know this, but I’m also trying to do wedding photography on the side of this as well, and I, like, hooked up my website to Google Analytics so I can, like, look at where users are, like, clicking into stuff, so…
110 00:11:13.120 ⇒ 00:11:20.450 Hannah Wang: But that’s, like, the extent of what I know for, like, data. So I just hooked up my website, and yeah, so…
111 00:11:20.450 ⇒ 00:11:21.850 Henry Zhao: It’s just interesting.
112 00:11:21.850 ⇒ 00:11:27.660 Hannah Wang: Okay, so the next section is challenges, so…
113 00:11:28.440 ⇒ 00:11:45.979 Hannah Wang: I feel like these are similar questions to context, but, this section is more to, like, uncover pain points, which I feel like you’ve already mentioned, but just say it again, so I capture it in the transcript. What specific problems or frustrations do people face with
114 00:11:46.000 ⇒ 00:11:51.230 Hannah Wang: Before, like, needing to implement Amplitude and track all these analytics.
115 00:11:51.940 ⇒ 00:11:55.349 Henry Zhao: I wouldn’t say it’s, like, necessarily a frustration.
116 00:11:55.670 ⇒ 00:12:00.330 Henry Zhao: But I think there’s just, like, no insight into how the product’s being used.
117 00:12:00.600 ⇒ 00:12:06.610 Henry Zhao: And then whenever new features are rolled out, they don’t know if it’s being implemented, or…
118 00:12:07.520 ⇒ 00:12:23.439 Henry Zhao: they don’t know, like, what the loyal users are typically like, right? So, whenever you have a product, you want loyal users to keep coming back and generate more loyal users, and also in marketing, you want to do lookalike campaigns, you want to bring in more people that are like your loyal users, right?
119 00:12:23.550 ⇒ 00:12:28.519 Henry Zhao: So it’s getting a better understanding, I think, of your user base and their user journey.
120 00:12:29.030 ⇒ 00:12:31.750 Hannah Wang: Okay, and this might be obvious.
121 00:12:32.030 ⇒ 00:12:37.570 Hannah Wang: The answer might be obvious, but what’s the consequence of, like, not knowing all those insights?
122 00:12:37.570 ⇒ 00:12:38.270 Henry Zhao: sum up.
123 00:12:39.150 ⇒ 00:12:45.930 Henry Zhao: For the business? I would say it’s… a lot of people say it’s like flying blind, right? So…
124 00:12:46.820 ⇒ 00:12:58.590 Henry Zhao: Yeah, it’s like doing product development, doing marketing, and doing… meetings, basically flying blind. Like, we don’t know our customer, we think this is how people use it, but the reality often is not the same, right?
125 00:12:59.240 ⇒ 00:13:03.760 Hannah Wang: Got it. And… I guess, what would happen if…
126 00:13:03.930 ⇒ 00:13:13.239 Hannah Wang: we didn’t… like, if you didn’t… if clients didn’t track any of these metrics and analytics, like, what would happen? I guess the business would…
127 00:13:13.960 ⇒ 00:13:21.749 Hannah Wang: be steered into another direction, and yeah, they would continue flying blind and, like, lose revenue or retention, I don’t know, like, I guess…
128 00:13:22.220 ⇒ 00:13:25.279 Henry Zhao: Yeah, I’m kind of a little bit torn on this, yeah.
129 00:13:25.900 ⇒ 00:13:28.680 Henry Zhao: like, I think you can always get lucky, right? But…
130 00:13:28.880 ⇒ 00:13:35.390 Henry Zhao: I think it’s better, and I think for everyone’s jobs, right, it’s better to show that you are making decisions based off of data.
131 00:13:35.520 ⇒ 00:13:38.879 Henry Zhao: You know what I mean? Cuz…
132 00:13:38.880 ⇒ 00:13:39.380 Hannah Wang: Yeah.
133 00:13:39.380 ⇒ 00:13:40.670 Henry Zhao: How do I word this?
134 00:13:42.270 ⇒ 00:13:44.540 Henry Zhao: So, if you put out a product that you think people want.
135 00:13:44.710 ⇒ 00:13:46.640 Henry Zhao: They either want it or they don’t want it.
136 00:13:46.890 ⇒ 00:13:52.070 Henry Zhao: If they do want it, you got lucky. If they don’t want it, you can correct course, right? So…
137 00:13:52.290 ⇒ 00:13:57.099 Henry Zhao: I think it’s maybe working… maybe pivoting faster, because you have data to go off of.
138 00:13:57.820 ⇒ 00:13:58.760 Hannah Wang: Totally. Okay.
139 00:14:00.100 ⇒ 00:14:24.929 Hannah Wang: So next is kind of going into the nitty-gritty, and feel free to be super technical, even if you think I won’t understand, which I might not, but… So I guess, yeah, so the next part is the solution. So usually, what does the solution involve? I know you mentioned, like, hooking a JavaScript to an event or something, and, like, sending that to Amplitude. Like, I guess you can also share your screen with me if we have Amplitude, and you can just walk me through
140 00:14:24.990 ⇒ 00:14:27.710 Hannah Wang: The entire flow of setting everything up.
141 00:14:28.500 ⇒ 00:14:37.539 Henry Zhao: Yeah, I don’t think I need to share my screen, but basically, I think the first step is listing out the business problems that you want to solve, right? So…
142 00:14:37.910 ⇒ 00:14:42.209 Henry Zhao: Based on what your company is, or what you’re trying to achieve, like, what are the…
143 00:14:42.680 ⇒ 00:14:56.340 Henry Zhao: objectives. What do we want to achieve? Do we want to increase loyal users? Do we want to see how people are using the product? Do we want to, do A-B testing? Like, those are the kind of things you want to answer first.
144 00:14:56.520 ⇒ 00:15:02.049 Henry Zhao: Then, the next step is figuring out what events do I need to track in order to accomplish that, right?
145 00:15:02.310 ⇒ 00:15:10.849 Henry Zhao: So… if I want to see, like, are people implementing new features, are using our new features, then we need to obviously have event tracking for new features.
146 00:15:11.050 ⇒ 00:15:19.549 Henry Zhao: If we want to see… what are people… so, like, if we’re a search tool, right, and we want to see what are people searching, then we need to log what are people searching?
147 00:15:19.550 ⇒ 00:15:20.150 Hannah Wang: Yeah.
148 00:15:20.150 ⇒ 00:15:33.839 Henry Zhao: Obviously, right? So, like, the search event needs to be a logged event, and then, like, you have a bunch of parameters that you can add, right? So, like, what country are they from? What device are they using? But I would want to see, like, what are they searching for? And then maybe do some sort of categorization.
149 00:15:34.100 ⇒ 00:15:36.380 Henry Zhao: So are they searching for…
150 00:15:37.100 ⇒ 00:15:46.899 Henry Zhao: politics, government, are they searching for marketing, economics, things like that, right? So that I can solve those business issues. That’s why business issues is first.
151 00:15:47.240 ⇒ 00:15:52.040 Henry Zhao: Once I have that, I would put a list, I’d have a list of events that we want to track.
152 00:15:52.310 ⇒ 00:15:54.469 Henry Zhao: I would then say, like, are these possible?
153 00:15:54.700 ⇒ 00:16:00.240 Henry Zhao: And then, how do we want to name these events, right? So…
154 00:16:01.290 ⇒ 00:16:06.290 Henry Zhao: Usually, I want to, like, group them by, like, function, or by, like, product type.
155 00:16:06.700 ⇒ 00:16:10.729 Henry Zhao: So, let’s say that… How do I put this?
156 00:16:10.960 ⇒ 00:16:22.329 Henry Zhao: Like, if events are related, I want them to have similar names, right? So that I, when I pull the data, I can, like, just say, like, give me the first three words are user search function.
157 00:16:22.720 ⇒ 00:16:27.089 Henry Zhao: And then it could be, like, user search function politics, user search function government, user search function.
158 00:16:27.250 ⇒ 00:16:37.269 Henry Zhao: economics, and then I can, like, group those into the same bucket. Then I would go to the engineer and say, okay, we need these events implemented, here’s where you would fire them, and then they would have to implement that.
159 00:16:37.510 ⇒ 00:16:47.649 Henry Zhao: And then I would create dashboards, right? So then once the events are coming in and I test it, they’re okay, I would start implementing the dashboards, the analysis, and then deliver that to the client.
160 00:16:48.060 ⇒ 00:16:53.789 Hannah Wang: Okay, so is Amplitude the dashboard part of everything?
161 00:16:54.200 ⇒ 00:17:04.470 Henry Zhao: They receive the events, and then you can look at the events, you can make dashboards off of them, so you can say, like, tell me all the people that did this event in the past 7 days, past 30 days.
162 00:17:04.560 ⇒ 00:17:13.320 Henry Zhao: And make charts off of that. You can also make, like, user journeys, so you can say, like, people that did A, what did they do next? What did they do after that? So you can see, like.
163 00:17:13.319 ⇒ 00:17:25.440 Henry Zhao: people that logged on, then they went to the homepage. On the homepage, they, I don’t know, added a new user, and then out of those people that added a new user, like, 80% churned, so, like, we don’t want…
164 00:17:25.460 ⇒ 00:17:27.979 Henry Zhao: Maybe something’s broken, or we wanna, like, fix that.
165 00:17:28.210 ⇒ 00:17:28.800 Hannah Wang: Okay.
166 00:17:29.100 ⇒ 00:17:35.859 Hannah Wang: So, Amplitude is not a data visualization tool, it’s… It is.
167 00:17:35.860 ⇒ 00:17:36.240 Henry Zhao: Yes.
168 00:17:36.240 ⇒ 00:17:37.159 Hannah Wang: It is? Okay.
169 00:17:37.160 ⇒ 00:17:38.770 Henry Zhao: It has, yeah, it has, yeah.
170 00:17:38.770 ⇒ 00:17:40.550 Hannah Wang: It has that, but what’s, like, the main…
171 00:17:40.660 ⇒ 00:17:44.940 Hannah Wang: Functionality, just product analytics tool? Like, what is…
172 00:17:44.940 ⇒ 00:17:50.559 Henry Zhao: It really just captures the events, and then allows you to use those events to make visualizations.
173 00:17:50.560 ⇒ 00:17:51.330 Hannah Wang: Okay.
174 00:17:52.800 ⇒ 00:17:54.479 Hannah Wang: Got it.
175 00:17:54.480 ⇒ 00:17:56.969 Henry Zhao: So you can think of, like, if you were running a circus, right?
176 00:17:56.970 ⇒ 00:17:57.380 Hannah Wang: Yes.
177 00:17:57.380 ⇒ 00:18:01.829 Henry Zhao: Ampitude would be that person at the entrance of the circus saying, like, okay.
178 00:18:02.390 ⇒ 00:18:19.279 Henry Zhao: Clowns came in, we opened the circus, customers came in, they paid X amount of dollars, and then you can have all this data and then say, like, how many people came in, how much money did we make today, how many clowns came in, how long did we operate, like, those type of things, based on logging events.
179 00:18:19.560 ⇒ 00:18:20.460 Hannah Wang: I see.
180 00:18:20.460 ⇒ 00:18:21.800 Henry Zhao: Everything is event-based, right?
181 00:18:21.800 ⇒ 00:18:24.060 Hannah Wang: I love that you used circus as the.
182 00:18:24.060 ⇒ 00:18:42.609 Henry Zhao: Yeah, I don’t know, that’s, like, the first thing that came to mind. But, like, to see, like, how long your circus is open, for example, you would have to say, alright, circus open at 8am, and the circus closed at 8pm, and you do, like, a calculated field where you say, alright, give me circus close time minus circus open time, to say, like, the circus is open 12 hours today.
183 00:18:42.610 ⇒ 00:18:43.060 Hannah Wang: Okay.
184 00:18:43.060 ⇒ 00:18:45.749 Henry Zhao: But everything has to be event-based, like, you can’t just do, like.
185 00:18:46.410 ⇒ 00:18:47.639 Henry Zhao: You know what I mean? Like…
186 00:18:48.350 ⇒ 00:18:53.969 Henry Zhao: It’s event-based, and then there’s user properties-based. So, events happen, and then there’s also users…
187 00:18:54.080 ⇒ 00:18:58.480 Henry Zhao: Property, where it’s, like, every person that comes into the circus gets, like, a profile.
188 00:18:58.480 ⇒ 00:19:01.589 Hannah Wang: It’s like, this is Hannah, this is Henry, this is whatever.
189 00:19:01.600 ⇒ 00:19:17.100 Henry Zhao: this is their device, this is their IP address, this is their customer ID, which is usually very helpful. And then you kind of join those together to say, who are our customers? What kind of properties do they have? Do we have more Android users? Do we have more Apple users? Those kind of things.
190 00:19:17.100 ⇒ 00:19:18.669 Hannah Wang: Got it. Okay.
191 00:19:19.010 ⇒ 00:19:25.960 Hannah Wang: And then, in addition to Amplitude, like, what other tools are usually used during that whole workflow?
192 00:19:25.960 ⇒ 00:19:29.390 Henry Zhao: I would say emphasis is pretty standalone.
193 00:19:29.560 ⇒ 00:19:36.419 Hannah Wang: Okay. Okay. Yeah, because usually in the case study, we have, like, a tool section, but if this… yeah, if it’s standalone, then that makes sense.
194 00:19:36.420 ⇒ 00:19:40.909 Henry Zhao: But you can always export to CSV and, like, do additional analysis, I just… it’s not that common.
195 00:19:40.910 ⇒ 00:19:44.959 Hannah Wang: Okay. Gotcha. Okay, so the next section is, I guess.
196 00:19:44.960 ⇒ 00:19:45.400 Henry Zhao: Pretty much.
197 00:19:45.400 ⇒ 00:19:52.509 Hannah Wang: results. So, yeah, measuring, like, quantifying the improvements that Amplitude does for clients. So,
198 00:19:53.880 ⇒ 00:19:56.280 Hannah Wang: And in this section, I usually like to ask.
199 00:19:56.280 ⇒ 00:19:56.960 Henry Zhao: Like.
200 00:19:56.960 ⇒ 00:20:03.600 Hannah Wang: If there’s any specific metrics that… or numbers that you can, like, call out to me that, like.
201 00:20:03.600 ⇒ 00:20:04.610 Henry Zhao: Yeah, I would say, like…
202 00:20:04.610 ⇒ 00:20:05.960 Hannah Wang: Blah, blah, blah.
203 00:20:06.360 ⇒ 00:20:14.369 Henry Zhao: I would say, like, increased retention, because you kind of get to analyze where the churn is coming from, and kind of fix those gaps. Yeah.
204 00:20:14.570 ⇒ 00:20:21.560 Henry Zhao: Cutting new features that are not as successful as you would hope they would be, identifying bugs,
205 00:20:25.300 ⇒ 00:20:26.780 Henry Zhao: I think that’s about it.
206 00:20:26.780 ⇒ 00:20:33.730 Hannah Wang: Okay. And then also just being able to provide metrics to leadership or stakeholders, whoever it may be, that just wants to know, like.
207 00:20:33.730 ⇒ 00:20:37.739 Henry Zhao: How many people are using it? Like, should we put in more investment funding? Those kind of things.
208 00:20:37.740 ⇒ 00:20:42.490 Hannah Wang: Usually, it saves Money, I’m assuming, as well.
209 00:20:42.490 ⇒ 00:20:46.290 Henry Zhao: Yeah. Okay. I’m posters are kind of expensive, so…
210 00:20:46.290 ⇒ 00:20:50.629 Hannah Wang: Okay. Well, hopefully in the long run it… yeah.
211 00:20:51.010 ⇒ 00:20:53.800 Hannah Wang: overall saves money for the business, I guess.
212 00:20:53.800 ⇒ 00:20:57.309 Henry Zhao: But if they’re expensive and people are still buying it, it must be providing value, right?
213 00:20:57.310 ⇒ 00:20:59.019 Hannah Wang: Yeah, totally. Okay.
214 00:20:59.020 ⇒ 00:20:59.790 Henry Zhao: Yeah.
215 00:20:59.790 ⇒ 00:21:05.969 Hannah Wang: Let’s see… Okay, I think that’s…
216 00:21:06.070 ⇒ 00:21:07.760 Henry Zhao: Pretty…
217 00:21:07.760 ⇒ 00:21:10.849 Hannah Wang: Good, and just for, like, my knowledge, like, what…
218 00:21:10.850 ⇒ 00:21:11.389 Henry Zhao: Oh, God.
219 00:21:11.390 ⇒ 00:21:12.170 Hannah Wang: client.
220 00:21:13.470 ⇒ 00:21:16.850 Hannah Wang: You know, like, I guess, wait, how do I say this?
221 00:21:16.850 ⇒ 00:21:17.360 Henry Zhao: off.
222 00:21:17.360 ⇒ 00:21:18.010 Hannah Wang: Like, which…
223 00:21:18.010 ⇒ 00:21:19.140 Henry Zhao: No, he’s in fact.
224 00:21:19.140 ⇒ 00:21:20.240 Hannah Wang: clients use.
225 00:21:20.240 ⇒ 00:21:22.999 Henry Zhao: Did we use… implement amplitude?
226 00:21:23.000 ⇒ 00:21:24.220 Hannah Wang: analytics for.
227 00:21:24.220 ⇒ 00:21:27.279 Henry Zhao: Default and README are the ones I know of.
228 00:21:27.280 ⇒ 00:21:31.540 Hannah Wang: Okay, I’m assuming there’s probably other ones that… okay, I just wanted to.
229 00:21:31.540 ⇒ 00:21:33.080 Henry Zhao: Uths will probably know better, yeah.
230 00:21:33.080 ⇒ 00:21:37.180 Hannah Wang: Okay, because usually I try to, like, I’m also on the sales side, so I try to, like.
231 00:21:38.150 ⇒ 00:21:40.440 Hannah Wang: these into buckets so I can, like.
232 00:21:41.030 ⇒ 00:21:44.109 Hannah Wang: yeah, if there’s any lookalike company or something, I can know.
233 00:21:46.060 ⇒ 00:21:50.559 Henry Zhao: But I think… I think pretty much every company that’s not, like, brick and mortar could benefit…
234 00:21:55.000 ⇒ 00:21:57.749 Henry Zhao: Fit from Amplitude or MixedPanel. Thank you.
235 00:21:57.920 ⇒ 00:21:58.920 Henry Zhao: Absolutely.
236 00:21:58.920 ⇒ 00:22:00.430 Hannah Wang: Okay, you got caught up, but you’re back.
237 00:22:00.430 ⇒ 00:22:01.570 Henry Zhao: I think I froze again, right?
238 00:22:01.920 ⇒ 00:22:02.969 Hannah Wang: Yeah, that’s okay.
239 00:22:02.970 ⇒ 00:22:03.300 Henry Zhao: Yeah, yeah.
240 00:22:03.300 ⇒ 00:22:07.509 Hannah Wang: I think I got everything that’s not brick and mortar could benefit from Amplitude or Mixpanel, is what you said.
241 00:22:07.510 ⇒ 00:22:08.969 Henry Zhao: Yeah, actually, I think it was cause I…
242 00:22:09.200 ⇒ 00:22:12.030 Henry Zhao: I said seriously or something, and Siri got activated.
243 00:22:12.030 ⇒ 00:22:24.000 Hannah Wang: Oh, okay. Okay, cool. And then, I guess if we have a mixed panel case study, like, wouldn’t it literally be the same?
244 00:22:24.000 ⇒ 00:22:24.859 Henry Zhao: Same thing.
245 00:22:25.580 ⇒ 00:22:28.210 Henry Zhao: You can literally Ctrl-F amplitude to Mixpanel, yeah.
246 00:22:28.340 ⇒ 00:22:33.719 Hannah Wang: I don’t know how I would make it different, except Logos and the names.
247 00:22:33.920 ⇒ 00:22:53.100 Henry Zhao: Okay, I will say one difference. So, one difference is their pricing structures. So, I think… I don’t remember which one is which right now, but one of them charges based on users tracking, the other one is based on event tracking. So, I worked at one company where they were trying to get a lot of different new users, but they don’t do much, right? So think about it, it’s like, a million people come in, and they do one thing.
248 00:22:53.820 ⇒ 00:23:04.260 Henry Zhao: So that’s 1 million users, 1 million events. And then you have customers that are, like, you have few customers, but they do a lot of things. So maybe it’s, like, 5 customers, each person does a million things.
249 00:23:04.260 ⇒ 00:23:04.740 Hannah Wang: I see.
250 00:23:04.740 ⇒ 00:23:10.820 Henry Zhao: So that’s gonna be 5 customers, 5 million events. So you wanna pay whatever’s cheaper based on your business model.
251 00:23:11.490 ⇒ 00:23:11.850 Hannah Wang: Got it.
252 00:23:11.850 ⇒ 00:23:16.069 Henry Zhao: So, like, the one that’s 5 customers, 5 million, I’m gonna wanna pay on customers, because that’s 5 people.
253 00:23:16.330 ⇒ 00:23:16.970 Hannah Wang: Yeah.
254 00:23:16.970 ⇒ 00:23:22.380 Henry Zhao: And 5 million events, whereas the one that’s, like, charging on events, I want to do the 1 million events, 1 million people one.
255 00:23:22.560 ⇒ 00:23:26.720 Henry Zhao: Because 1 million people is gonna cost, like, $35,000 a year. It’s gonna be crazy.
256 00:23:27.920 ⇒ 00:23:36.980 Hannah Wang: I wonder if those companies, like, know… that’s, like, kind of hacking the system, like, I wonder why they chose that pricing model.
257 00:23:36.980 ⇒ 00:23:40.920 Henry Zhao: It’s not hacknessism, I think that’s, like, how Amplitude and Mixed Panel differentiate themselves in the market.
258 00:23:40.920 ⇒ 00:23:42.019 Hannah Wang: I see. Okay.
259 00:23:43.360 ⇒ 00:23:44.360 Hannah Wang: Okay, yeah.
260 00:23:44.360 ⇒ 00:23:51.930 Henry Zhao: because we’re obviously paying them to store the data, right? Like, if I have 50 million customers, that’s a lot of data, but the way that they store it, whether it’s Parquette or…
261 00:23:52.250 ⇒ 00:23:54.289 Henry Zhao: Like, by rows, like…
262 00:23:54.560 ⇒ 00:23:59.840 Henry Zhao: it depends on whether you want to sort by user or by event, so I think… Okay. …that’s how they differentiate themselves, yeah.
263 00:23:59.840 ⇒ 00:24:00.480 Hannah Wang: I see.
264 00:24:00.690 ⇒ 00:24:14.720 Hannah Wang: Okay, then I think we knocked out two… I guess two case studies in one. I’ll just CTRL-F, and then copy-paste to just change logos and add that caveat, and then I think we should be good.
265 00:24:14.720 ⇒ 00:24:18.430 Henry Zhao: And I think, just also, in my opinion, Mixpanel’s a little bit easier to use, I would say.
266 00:24:18.590 ⇒ 00:24:20.419 Hannah Wang: like, UX-wise, just like…
267 00:24:20.420 ⇒ 00:24:23.670 Henry Zhao: Yeah, but that’s just my opinion. I don’t know how other people feel, yeah.
268 00:24:23.670 ⇒ 00:24:24.980 Hannah Wang: Okay, cool.
269 00:24:27.870 ⇒ 00:24:36.679 Hannah Wang: Okay, yeah, I think this is good. I will crank out the case study for both, and then probably have you… probably have you take a look at it sometime.
270 00:24:37.770 ⇒ 00:24:45.569 Hannah Wang: And, feel free to, like, leave comments or correct anything that’s wrong, and I think we should be good from there.
271 00:24:45.570 ⇒ 00:24:47.000 Henry Zhao: Okay, sounds good.
272 00:24:47.200 ⇒ 00:24:54.969 Hannah Wang: Thanks, yeah, I appreciate your time, and I feel like you explained things really clearly, so I… Oh, thank you. Just from, like, a knowledge standpoint of me…
273 00:24:54.970 ⇒ 00:24:56.020 Henry Zhao: I feel like I don’t.
274 00:24:56.020 ⇒ 00:25:00.950 Hannah Wang: Really? No, your analogies also help, so, like… Oh my god, I got…
275 00:25:01.230 ⇒ 00:25:04.480 Henry Zhao: People keep telling me I have the worst analogy, so I’m kind of, like, self-conscious about it, but it’s…
276 00:25:04.480 ⇒ 00:25:09.209 Hannah Wang: No, I love it. It’s help. Like, I’d rather have analogies than have, like.
277 00:25:09.690 ⇒ 00:25:10.489 Henry Zhao: Is that what we have two?
278 00:25:10.490 ⇒ 00:25:16.390 Hannah Wang: technical stuff that I’m like, I don’t… just goes over my head, so keep up the analogies, it’s good.
279 00:25:16.390 ⇒ 00:25:17.949 Henry Zhao: Okay, thank you. Appreciate that.
280 00:25:18.550 ⇒ 00:25:21.649 Hannah Wang: Yeah, well, I’ll talk to you on talk, and have a good one.
281 00:25:22.060 ⇒ 00:25:23.369 Henry Zhao: Sounds good, take care, Hannah.
282 00:25:23.540 ⇒ 00:25:24.240 Hannah Wang: Right?