Meeting Title: Weekly-Data-Review Date: 2024-03-08 Meeting participants: Uttam Kumaran, Daniel Schonfeld, Bencohen
WEBVTT
1 00:00:13.790 ⇒ 00:00:14.690 Huh!
2 00:01:20.720 ⇒ 00:01:21.750 Daniel Schonfeld: Depth.
3 00:01:22.840 ⇒ 00:01:24.169 bencohen: How we doin!
4 00:01:24.580 ⇒ 00:01:27.459 Daniel Schonfeld: Come on, Oliver
5 00:01:28.300 ⇒ 00:01:30.260 Daniel Schonfeld: Biscay. Huh?
6 00:01:30.920 ⇒ 00:01:34.209 bencohen: You’re so mobile now. How’s it feeling
7 00:01:35.130 ⇒ 00:01:37.949 Daniel Schonfeld: me? Oh, good!
8 00:01:38.020 ⇒ 00:01:41.580 Daniel Schonfeld: My fucking Yale keeps doing this. Hold on.
9 00:01:46.610 ⇒ 00:01:49.320 Daniel Schonfeld: It’s hurting a lot I’m going to clean on Monday
10 00:01:51.010 ⇒ 00:01:55.849 Daniel Schonfeld: for a scheduled visit or non scheduled, unscheduled. I told them I’m in like
11 00:01:56.430 ⇒ 00:02:00.100 Daniel Schonfeld: a lot of pain on the, on the bone, on the on, the
12 00:02:00.730 ⇒ 00:02:02.270 Daniel Schonfeld: on, the
13 00:02:03.390 ⇒ 00:02:13.040 Daniel Schonfeld: on the leg. like the bone leading into the what’s the ball on the edge, out of your out of your ankle. Calculations. What?
14 00:02:13.270 ⇒ 00:02:14.610 bencohen: Calcaneus?
15 00:02:15.260 ⇒ 00:02:24.319 Daniel Schonfeld: Yeah, it feels like. It’s Bru. It’s been as bruised as this like someone hit it with a hammer. So when I touch it, it’s like very pender.
16 00:02:24.740 ⇒ 00:02:29.350 Daniel Schonfeld: and then I get these. Is it the taylor’s?
17 00:02:29.910 ⇒ 00:02:33.749 Daniel Schonfeld: The the outside bone that leads into your ankle.
18 00:02:34.400 ⇒ 00:02:36.990 Daniel Schonfeld: The very end of it, before it hits the ankle
19 00:02:37.350 ⇒ 00:02:42.019 Daniel Schonfeld: is is been very, very tender and sore since the surgery.
20 00:02:43.530 ⇒ 00:02:44.360 Daniel Schonfeld: Oh.
21 00:02:44.580 ⇒ 00:02:46.210 bencohen: so he’s gonna take a look at.
22 00:02:46.330 ⇒ 00:02:50.179 Daniel Schonfeld: And then I get these like super sharp pains
23 00:02:50.630 ⇒ 00:02:56.410 Daniel Schonfeld: right above that bone. It almost feels like, how much are you walking? How many hours a day?
24 00:02:56.540 ⇒ 00:02:59.190 Daniel Schonfeld: I don’t know. An hour or 2,
25 00:03:00.910 ⇒ 00:03:09.129 Daniel Schonfeld: maybe too much. Who knows? I took it off a bunch of times. Walk about it, I’m sure. Oh, dude, what do you? That’s retarded. Yeah.
26 00:03:09.150 ⇒ 00:03:15.780 Daniel Schonfeld: you can’t help yourself. You need to get into rehab. Yeah, next week. But it feels like
27 00:03:15.890 ⇒ 00:03:26.320 Daniel Schonfeld: every once in a while. Get this like unbelievable shooting pain where it feels like there’s an open wound, and someone pours like alcohol on it like that kind of like tingling, burning.
28 00:03:27.210 ⇒ 00:03:30.210 Daniel Schonfeld: It’s real quick. It happens like 4 s. But
29 00:03:30.900 ⇒ 00:03:32.370 Daniel Schonfeld: I don’t know what that is.
30 00:03:33.240 ⇒ 00:03:36.260 bencohen: Sounds like. You’re not listening.
31 00:03:36.930 ⇒ 00:03:49.099 bencohen: You shouldn’t be. You shouldn’t be walking without tech. I can’t believe you’re just inside, not outside. Just like to the bathroom once in a while doesn’t matter. Yeah, the thing is tightened up like,
32 00:03:49.190 ⇒ 00:03:52.999 Uttam Kumaran: if the rehab will fix everything. Hi, Tom, good morning, hey? Good morning.
33 00:03:53.900 ⇒ 00:03:56.019 Daniel Schonfeld: We’re we’re real this time. We’re not virtual.
34 00:03:56.190 ⇒ 00:04:08.209 Uttam Kumaran: I know. I know it’s uglier. It’s actually was an improvement. Dan’s falling out of love with the vision pro. And I’m falling more in love with it. I’m just running out of content.
35 00:04:08.390 ⇒ 00:04:10.879 Daniel Schonfeld: It’s like the the novelty is wearing off a bit.
36 00:04:11.690 ⇒ 00:04:14.419 Uttam Kumaran: It’s been one day.
37 00:04:14.770 ⇒ 00:04:20.690 Uttam Kumaran: that’s that’s all it takes.
38 00:04:20.769 ⇒ 00:04:35.250 Daniel Schonfeld: It’s like go to Amusement Park. You know. How many times can you ride the you know the upside down roller coaster? I need more roller coasters. It’s only I only been. It’s like going to theme Park with 2 roller coasters. That’s what it feels like. They need to make more.
39 00:04:35.260 ⇒ 00:04:43.110 Daniel Schonfeld: You gotta do abandoned. You gotta live on the roller coaster for a week. I tried. You know what? I also have different issues because of my eyes.
40 00:04:43.210 ⇒ 00:04:51.130 Daniel Schonfeld: So like, I see a dot. I have to see a.in middle of every screen. I have to like. Point my fucking nose at everything in order to make it work.
41 00:04:51.800 ⇒ 00:04:54.979 Daniel Schonfeld: It’s I think the battery life sucks.
42 00:04:55.380 ⇒ 00:04:57.580 Daniel Schonfeld: It’s the battery’s a problem.
43 00:04:58.430 ⇒ 00:05:10.580 bencohen: You have to carry this like Backpack or this back around with you, Dan, I’m telling it like it is. Honestly, I’m just going back.
44 00:05:11.060 ⇒ 00:05:20.389 Uttam Kumaran: I’m just gonna wear. I’m just gonna actually to prove a point. I’m just gonna wear this the entire time. I’m now. I have everything I have. You guys here and I have.
45 00:05:20.430 ⇒ 00:05:25.590 Daniel Schonfeld: I have the Tony’s, the the Tony Stark situation.
46 00:05:27.970 ⇒ 00:05:34.719 bencohen: Alright. So Utam. We had a lot we talked so there. So a lot went down this week. There’s a lot of Patrick
47 00:05:35.100 ⇒ 00:05:43.269 bencohen: Cody, a lot of Kim, a lot of me and you. We don’t need to make this one long. But let’s tie this in a bow.
48 00:05:44.120 ⇒ 00:05:45.610 Uttam Kumaran: So
49 00:05:45.850 ⇒ 00:05:53.670 Uttam Kumaran: also start with the warranty. So we kind of like. Pretty much did a deep dive into everything
50 00:05:53.980 ⇒ 00:05:57.410 Uttam Kumaran: warranties related. Basically.
51 00:05:58.100 ⇒ 00:06:05.210 Uttam Kumaran: we mainly found that there was that one driver issue that Cody found out. But I think, right. Apart from that, we have a good
52 00:06:05.280 ⇒ 00:06:15.609 Uttam Kumaran: like footing on just warranty data, and how warranties are being claimed across queues, so we can monitor that as that goes up and down. But basically.
53 00:06:16.330 ⇒ 00:06:23.739 Uttam Kumaran: the only issue that we found that when I talked to Cody was that there’s that driver issue
54 00:06:23.900 ⇒ 00:06:30.310 Uttam Kumaran: and so wanted to see if you guys were aware of that, or if there’s anything to to do there,
55 00:06:30.450 ⇒ 00:06:31.820 Uttam Kumaran: maybe we can start there.
56 00:06:32.460 ⇒ 00:06:43.779 Daniel Schonfeld: Yeah, I already spoke to John about it. and he was supposed to get me on with the factory this week. But we’re gonna have to get on next week. We’re gonna get a credit, II estimated, he said, around 150 units total.
57 00:06:44.210 ⇒ 00:06:47.830 Daniel Schonfeld: So far. So we’re just gonna get a credit. It’s like 65, grand.
58 00:06:47.990 ⇒ 00:06:48.860 Uttam Kumaran: Okay?
59 00:06:49.060 ⇒ 00:06:53.770 Daniel Schonfeld: Is that right? Cause? Cody said. On a call this week we had a team call.
60 00:06:53.820 ⇒ 00:07:04.840 Daniel Schonfeld: It was a total of like 348, if I think the number was correct units total. And then he said, about 30% or 35% or something like that were driver related.
61 00:07:05.720 ⇒ 00:07:09.700 Uttam Kumaran: Yeah. And and the thing is, II actually realize I don’t have the.
62 00:07:09.770 ⇒ 00:07:15.809 Uttam Kumaran: So I don’t want. The other question is, I don’t have the Ben. I’ve texted you about this, but the parts
63 00:07:15.980 ⇒ 00:07:17.559 Uttam Kumaran: related shopify.
64 00:07:17.600 ⇒ 00:07:30.080 Uttam Kumaran: So I. So what the calculation we did is we basically looked at the cost of goods sold, and then how many warranty claims there were so many replacements there were. We didn’t look at any like individual parts.
65 00:07:30.140 ⇒ 00:07:37.450 Uttam Kumaran: but even then the like. The Bdxba ones are, are much higher than the
66 00:07:37.610 ⇒ 00:07:50.769 Uttam Kumaran: the other ones. And did Cody not give you access to that site? I mean, I can do it. But I told him to do it. Not a problem. I will handle that right now.
67 00:07:51.440 ⇒ 00:08:03.410 Daniel Schonfeld: Is there anything else I’m missing? Been on this when I when I go back to the factory and say, Oh, that’s it’s 150 units. Obviously there’s ship additional shipping costs that that are tacked on. I think I actually included that
68 00:08:03.550 ⇒ 00:08:05.569 bencohen: I mean the the the.
69 00:08:05.650 ⇒ 00:08:09.209 bencohen: It’s it’s maybe unfair to calculate. But, like
70 00:08:09.610 ⇒ 00:08:12.890 bencohen: you sh! You know, product shipping.
71 00:08:13.370 ⇒ 00:08:14.710 bencohen: stocking
72 00:08:15.260 ⇒ 00:08:23.110 Uttam Kumaran: a cost around customer service back and forth, I think, yeah, mainly we just see the shipping and the and the replacement.
73 00:08:23.200 ⇒ 00:08:36.320 Uttam Kumaran: Outside of that we don’t have many thoughts. And again, part of our our goal is like, can we just see a signal and like what’s high? And then maybe you can just say any any warranty.
74 00:08:36.490 ⇒ 00:08:44.239 Uttam Kumaran: It’s like an extra 5% just given like logistics. yeah, that’s what I’m trying to figure out is like, what is that
75 00:08:44.380 ⇒ 00:08:50.950 Daniel Schonfeld: statistically relevant number per skew where we have to raise flags. And that’s what I was trying to get to with Cody.
76 00:08:51.020 ⇒ 00:08:56.320 Uttam Kumaran: he said. II really started tagging the warranties like 11 months ago, he was telling me.
77 00:08:56.970 ⇒ 00:09:01.070 Daniel Schonfeld: But we’ve obviously been selling that unit for 3, 4 years.
78 00:09:01.260 ⇒ 00:09:05.439 Daniel Schonfeld: so it’s hard to tell if it’s a problem or not yet.
79 00:09:05.470 ⇒ 00:09:13.699 Daniel Schonfeld: because I don’t know which batch. Also they’re coming from. It could. Maybe there’s a hundred 50 from one container, but I don’t know how yet
80 00:09:13.790 ⇒ 00:09:18.560 Daniel Schonfeld: to to figure that out by using serial numbers or model numbers.
81 00:09:19.740 ⇒ 00:09:23.680 Uttam Kumaran: Yeah, I mean. I would have to look and unleashed if we have
82 00:09:24.320 ⇒ 00:09:28.600 Uttam Kumaran: that cause. I also, we don’t have a through line into the supply.
83 00:09:29.180 ⇒ 00:09:31.390 Uttam Kumaran: We just know what the supply is.
84 00:09:31.710 ⇒ 00:09:44.779 Daniel Schonfeld: Yeah, but at the very least some order date is helpful to tie, to warranty date, to start understanding over time. What is that time period where we start seeing them come back.
85 00:09:45.010 ⇒ 00:09:50.869 Daniel Schonfeld: and also, and putting aside funds and or inventory
86 00:09:51.010 ⇒ 00:09:53.120 Daniel Schonfeld: to
87 00:09:53.340 ⇒ 00:10:08.940 Daniel Schonfeld: you know, to anticipate that. So if we start knowing that around the year mark after pumps go out, we start to see a 5% warranty reclaim. We know we have to have that in stock also as a separate for just warranties and parts.
88 00:10:08.940 ⇒ 00:10:27.590 Daniel Schonfeld: That’s more the probably the most important. Well, most important thing is getting to what the root issue is of the warranty claims. If there is, there’s always gonna be a delta, or or a number rather, of people who return things, but those outside of that delta. And then we’ll figure that out over time. That’s when red flags have to go off.
89 00:10:27.960 ⇒ 00:10:37.090 Daniel Schonfeld: But again it needs to tie back to the original order dates to see if it really is an issue. He can come in today and say, we just had 50 today.
90 00:10:37.230 ⇒ 00:10:39.620 Uttam Kumaran: But yeah, it might not be an issue at all.
91 00:10:39.670 ⇒ 00:10:49.009 Daniel Schonfeld: because the 50 might be extrapolated out over 4 years. Even exactly so, the things we’re trying to get to is like, what is the
92 00:10:49.350 ⇒ 00:10:54.730 Uttam Kumaran: when do we raise a flag as like warranties is a percent of cogs
93 00:10:54.760 ⇒ 00:11:07.039 Uttam Kumaran: went up. For example, if the reason why we didn’t look as a percent of sales is because you’re not actually losing any sales you’re just using. The replacement cost case is 3% like the barrier.
94 00:11:07.150 ⇒ 00:11:12.869 Uttam Kumaran: And and then basically, can we look at that per month and say, like, in any given month. If we hit this.
95 00:11:13.010 ⇒ 00:11:25.729 Uttam Kumaran: we’re basically like a little bit ahead of what we should be doing per month. That’s something we can get to. We noticed that on the cft, it’s around like 2 to 3%. But on the other pumps.
96 00:11:25.910 ⇒ 00:11:29.070 Uttam Kumaran: yeah. And it’s like it’s it’s much, much higher.
97 00:11:29.090 ⇒ 00:11:34.470 bencohen: I don’t want to do benchmarking.
98 00:11:35.520 ⇒ 00:11:40.890 bencohen: even goal setting of any kind. Yet what I want to do is slice and dice
99 00:11:41.330 ⇒ 00:11:49.130 bencohen: deeper. Figure out the time. How much time is, you know, per skew. How much time on average, is there a cluster where.
100 00:11:49.340 ⇒ 00:11:53.330 bencohen: you know? December of 2022
101 00:11:53.550 ⇒ 00:11:56.010 bencohen: we had a crazy amount where we know.
102 00:11:56.400 ⇒ 00:12:03.609 bencohen: Let’s slice and dice deeper. Find some information, and then then we’ll then we’ll know. You know.
103 00:12:04.000 ⇒ 00:12:12.280 bencohen: The issue in one way is like some of these issues are solved. The issue is that we still have stock of
104 00:12:13.780 ⇒ 00:12:25.940 bencohen: product that has not had all the newest changes. So like, for example, there’s a lot of changes made to the heat pumps, but we still have 6 months of sellable goods on hand, meaning
105 00:12:26.420 ⇒ 00:12:29.289 bencohen: that 2% will continue until those are
106 00:12:30.080 ⇒ 00:12:41.559 bencohen: cleared. And and there, and by the way, there are on on all products that is holds true where there’s been improvements made incrementally, and the factory is making good product.
107 00:12:41.630 ⇒ 00:12:44.930 bencohen: It’s just might not sell it yet.
108 00:12:46.380 ⇒ 00:12:52.509 bencohen: just something to take, just something to know. But I think before we set, because, like I don’t. What I don’t wanna do because we’re not like.
109 00:12:53.460 ⇒ 00:13:02.360 bencohen: you know, sound like a duck and quack like a duck. We’re not like a Vc. Packed, do you see? Like a lot of these other ones? So like when you see industry benchmarks around returns.
110 00:13:03.360 ⇒ 00:13:06.850 bencohen: I don’t put much stock in them because they’re selling socks.
111 00:13:07.080 ⇒ 00:13:18.140 bencohen: and their Vc. Is saying, we don’t really care if you make any money at all. We want you to get to a hundred 1 million dollar valuation as fast as possible. So returns galore all day. Baby.
112 00:13:18.500 ⇒ 00:13:21.270 bencohen: we just don’t do that. So
113 00:13:21.520 ⇒ 00:13:33.119 bencohen: what we need to do is get all information, know everything possible. And then, once we have it, we can say, Okay, we can live with a 1.6% return rate on these skews.
114 00:13:33.260 ⇒ 00:13:38.560 bencohen: Only then can we do it, though. So we’re close. I think you’re gonna have all your information.
115 00:13:38.760 ⇒ 00:13:43.979 Uttam Kumaran: Okay, yeah, III think you’re totally right. And then again, we mainly want to also see.
116 00:13:44.190 ⇒ 00:13:49.060 Uttam Kumaran: like are is the war. We mainly also looked at the warranty process itself.
117 00:13:49.090 ⇒ 00:13:53.080 Uttam Kumaran: So what happens when Cody gets a warranty claim. And
118 00:13:53.370 ⇒ 00:14:00.870 Uttam Kumaran: like, what’s the process of us tagging? And then that’s when we realized the warranties are coming in as as 0 as 100 discount sales
119 00:14:01.000 ⇒ 00:14:11.040 Uttam Kumaran: instead of like just not being counted at all. So the the thing that we’re working with him on is, can we count? Can we create new skews that are
120 00:14:11.130 ⇒ 00:14:13.829 Uttam Kumaran: 0 cost skews that way. It doesn’t affect
121 00:14:13.870 ⇒ 00:14:17.489 Uttam Kumaran: discounts or sales.
122 00:14:17.850 ⇒ 00:14:22.380 Uttam Kumaran: because that’s that’s likely. The reason why our discount amounts are are
123 00:14:22.450 ⇒ 00:14:23.540 Uttam Kumaran: much higher
124 00:14:23.650 ⇒ 00:14:35.720 bencohen: like we talked about pro. A lot of the process was created by me, and I don’t know anything. I just did logically what I thought A to B for the company that was doing 0 in sales to get to a mail just
125 00:14:36.090 ⇒ 00:14:43.919 bencohen: moving, fast, breaking things, and then Cody adapt, you know, adapted some things a little bit better. But there’s still
126 00:14:44.340 ⇒ 00:14:48.140 bencohen: a lot of process that’s not great. So
127 00:14:48.890 ⇒ 00:14:54.869 bencohen: you know, like I said, you and Cody can decide what to do. But we need him and his team logging thing. So we have
128 00:14:55.060 ⇒ 00:14:56.570 Uttam Kumaran: full capability.
129 00:14:57.000 ⇒ 00:14:59.220 Uttam Kumaran: Okay, okay, great.
130 00:15:01.480 ⇒ 00:15:17.909 Daniel Schonfeld: Yeah. Let’s Cody also dig into Ben. It’s good for him to Bootam. Main function right now is to set these things up so other people can identify them like I don’t. I want him to be able to also learn how to read the data.
131 00:15:17.930 ⇒ 00:15:21.929 Daniel Schonfeld: So it’s, you know, I don’t want boot time to always have to do this. It’s really.
132 00:15:22.070 ⇒ 00:15:48.729 Daniel Schonfeld: you know, I want to attempt to help build the the vehicle for them to get on and figure all these things out. So Tom’s gonna be stuck in data, become a data analyst full time on this stuff. That’s not what we want. So let me send him. We really put all this data in just a quick dashboard. I’ll just send it to him. And maybe we do quick. 30 min is like, here’s how you can click around pretty basic data. And then we can kind of have him kind of drive.
133 00:15:49.220 ⇒ 00:15:50.600 bencohen: Yeah.
134 00:15:50.910 ⇒ 00:16:01.240 Daniel Schonfeld: not very disciplined. So he he needs to get on a cadence of like a weekly thing where he is forced to look at these reports or run them
135 00:16:01.400 ⇒ 00:16:15.539 Daniel Schonfeld: and start to at least just eyeball it, and then bring it to a meeting. And just say everything will storm all 3. Here’s the warranty rate. Redemption rate
136 00:16:15.660 ⇒ 00:16:23.269 Uttam Kumaran: customer service. What’s a quick report? I mean, he had some data on the warranties look like it’s coming from Google sheets. So I think there’s some tracking
137 00:16:23.510 ⇒ 00:16:25.359 Uttam Kumaran: going on. But
138 00:16:25.520 ⇒ 00:16:32.709 Uttam Kumaran: maybe, like, I’ll just pop on a call and just say, like, Hey, here’s here’s a lot of data we have. Here’s how you can explore it, and I think should be
139 00:16:33.870 ⇒ 00:16:34.930 Uttam Kumaran: good to go.
140 00:16:35.010 ⇒ 00:16:36.430 Daniel Schonfeld: Okay, good.
141 00:16:37.170 ⇒ 00:16:49.439 Uttam Kumaran: The other thing that we wanted to talk about was these discount or price tasks. So we spoke with Kim.
142 00:16:49.990 ⇒ 00:16:53.590 Uttam Kumaran: So we were like, Okay, what’s the what were 2 things?
143 00:16:53.620 ⇒ 00:17:05.180 Uttam Kumaran: Ben was like, okay, we can change price. We can change marketing. And so we’re like, let’s just look at how we’re doing discounts. Basically, we found that everybody is using like of the codes used.
144 00:17:05.290 ⇒ 00:17:10.350 Uttam Kumaran: All of them are pretty much the welcome 5 discount code
145 00:17:10.569 ⇒ 00:17:16.000 Uttam Kumaran: And so what we want to understand is like, is that effective or not? Basically
146 00:17:16.030 ⇒ 00:17:25.190 Uttam Kumaran: and so one test that we want to try to run is just removing that first subset of users and seeing how that impacts sales.
147 00:17:25.839 ⇒ 00:17:34.820 Uttam Kumaran: The second thing we want to test, and there’s not a ton of other discounts that aren’t like really seasonal discounts. We’re working with him on like segmenting that. But
148 00:17:34.890 ⇒ 00:17:51.259 Uttam Kumaran: where I think it’s welcome 5 is a good introductory test to try to run. The second thing is, we want to adjust the price of some products, and then also pretty much get a gauge of like elasticity like, what’s how does that change? How does that impact?
149 00:17:51.460 ⇒ 00:18:00.500 Uttam Kumaran: the people like buying patterns, basically just getting a sense of, Hey, is this count more important than the price.
150 00:18:00.760 ⇒ 00:18:08.159 Uttam Kumaran: apart from like what we’ve been doing on the copy and messaging side, actually just running a test on how high or how low should the price be?
151 00:18:08.500 ⇒ 00:18:22.719 Uttam Kumaran: How high or high, low should the discounts be? And the other thing that we would have typically done is looked at competitive products. But because there’s not a ton in market. Or maybe you guys can suggest some we could look to get a gauge of like how far we could test
152 00:18:22.900 ⇒ 00:18:29.440 Uttam Kumaran: So we all just stop there. And then I can talk a little about like how we’re gonna try to do that. So I think
153 00:18:32.410 ⇒ 00:18:38.299 bencohen: II don’t think comp competition will be helpful because we’re if you look at like
154 00:18:38.340 ⇒ 00:18:45.359 bencohen: the graph of where we are price quality brand name, like, you know, you know, like all of those business school squadrons.
155 00:18:45.600 ⇒ 00:18:48.539 bencohen: we don’t fall anywhere near any of these piers.
156 00:18:48.600 ⇒ 00:18:54.210 bencohen: Yeah, we’re just kind of a different outfit. So I don’t think that that will be fruitful. I think
157 00:18:55.190 ⇒ 00:19:07.700 bencohen: the exercise that we talked about was just starting with Welcome 5, and seeing if we remove it, what kind of change there is if we keep it, what kind of changes? If we raise the price, we’ll kind of change if we make that
158 00:19:07.750 ⇒ 00:19:11.069 bencohen: we can. We can slice and dice it a million ways, but
159 00:19:11.340 ⇒ 00:19:15.830 Uttam Kumaran: we have a good baseline could have done the same thing for 2 years.
160 00:19:16.320 ⇒ 00:19:23.139 bencohen: Only difference is that the price is moving constantly. cause we’re manually moving it. But
161 00:19:23.540 ⇒ 00:19:28.409 bencohen: we can do anything. The biggest limitation we have is that shopify
162 00:19:28.840 ⇒ 00:19:32.989 bencohen: at least our theme doesn’t have some of like fancier.
163 00:19:33.570 ⇒ 00:19:37.040 bencohen: Optimizely, kind of stuff where you can have
164 00:19:37.680 ⇒ 00:19:53.720 bencohen: different people landing in a different experience. We did have a an A like a, you know, an optimizing type of agency working for us for 6 months they were playing with different things. They actually did do a good job, but we didn’t go there.
165 00:19:54.590 ⇒ 00:20:10.389 bencohen: They were doing more. We wanted less disruptive things. We were. We were looking for smaller optimizations on pdps, and they did. They did come through. Pretty nice. But what? You’re what you in the best case that you were talking about yesterday or the day before was like.
166 00:20:10.830 ⇒ 00:20:15.630 Uttam Kumaran: we have 2 landing pages. One is no discounts. One is this discount?
167 00:20:15.780 ⇒ 00:20:20.879 bencohen: There might be a plugin of some kind that can accomplish this. I don’t know.
168 00:20:21.150 ⇒ 00:20:23.569 bencohen: we just gotta take a look. Kim can do that.
169 00:20:23.880 ⇒ 00:20:36.069 Uttam Kumaran: Yeah. So II pretty much. What I do is I just text like a bunch of people I know that are running these tests. And II go on Twitter. And a lot of people recommended this intelligence. Basically, it’s like, I think of
170 00:20:36.090 ⇒ 00:20:43.030 Uttam Kumaran: pretty like, easy to use, like, kind of optimizely tool optimize. It’s kind of a gigantic tool.
171 00:20:43.110 ⇒ 00:20:50.899 Uttam Kumaran: So I was gonna try to get a demo from these guys. Maybe I can record it and then just see if they can give us like
172 00:20:51.280 ⇒ 00:20:54.909 Uttam Kumaran: a week for free or so, and we can try out one of them.
173 00:20:55.050 ⇒ 00:21:00.720 Uttam Kumaran: But the thing, the thing that’s tough here is like, I want to really test like apples to apple. So I want to have, like
174 00:21:00.880 ⇒ 00:21:06.619 Uttam Kumaran: what is welcome 5 versus not having it, and then get a statistically significant amount.
175 00:21:06.990 ⇒ 00:21:12.980 Uttam Kumaran: same thing with the with the prices. So I think, Ben, maybe on the price side, if you have ideas on
176 00:21:13.940 ⇒ 00:21:16.729 Uttam Kumaran: which use would be good to move
177 00:21:17.160 ⇒ 00:21:28.799 Uttam Kumaran: basically in which direction? Again, I can put some. I can propose couple of things that way. It’s some. We maybe do it by percent or something. But since you’ve been modifying it, you might have a good idea
178 00:21:28.940 ⇒ 00:21:32.680 Uttam Kumaran: what’s been working or what hasn’t, or what we want to figure out.
179 00:21:34.440 ⇒ 00:21:43.079 bencohen: Let’s see about this software. First, let’s see if we can use it, and then I will advise. Yeah. One note is that it’s
180 00:21:43.830 ⇒ 00:21:45.230 bencohen: March eighth.
181 00:21:45.490 ⇒ 00:21:49.389 bencohen: And I’m noticing on Amazon that are.
182 00:21:49.870 ⇒ 00:21:51.290 bencohen: Things are getting hot.
183 00:21:51.680 ⇒ 00:21:55.060 Uttam Kumaran: We are about to hit the season.
184 00:21:55.380 ⇒ 00:22:01.290 bencohen: I wanna experiment. It might even be better to experiment with more traffic
185 00:22:01.820 ⇒ 00:22:03.690 bencohen: people more apt to buy.
186 00:22:04.020 ⇒ 00:22:06.470 bencohen: But I also don’t want to fuck around
187 00:22:06.740 ⇒ 00:22:16.419 Uttam Kumaran: my yeah, I coming into March. I was like, I wanted to try to close it all out. March for one weekend. So I wanna try and get
188 00:22:17.050 ⇒ 00:22:21.750 Uttam Kumaran: things going by like early next week. and at least have something to try.
189 00:22:22.120 ⇒ 00:22:37.989 Daniel Schonfeld: Yeah, there’s a certain acceptable amount of traffic, though, Ben, like we with like optimizing. And some of these other programs, I’m sure this other one. intelligent, whatever it’s called would work the same where you can bifurcate, and then just actually say, only use 10% of the traffic as it comes in and
190 00:22:38.160 ⇒ 00:22:40.079 Daniel Schonfeld: use it.
191 00:22:40.130 ⇒ 00:22:58.590 Daniel Schonfeld: you know, and kind of multivariate testing where it just kinda switches off and random. II don’t think it’s the type of thing where we have to like, take half of it. But obviously, again, you have to statistical amount. So it is better to do it. Even now I’m okay, losing 10% for a week. To figure this out and boost conversions
192 00:22:58.670 ⇒ 00:23:00.990 Daniel Schonfeld: and pricing and all that.
193 00:23:01.300 ⇒ 00:23:03.010 Daniel Schonfeld: it will gain it right back
194 00:23:05.380 ⇒ 00:23:07.619 Daniel Schonfeld: as long as it’s a smaller
195 00:23:07.640 ⇒ 00:23:16.460 Uttam Kumaran: sample size that we’re using. Yeah. So I’m just gonna get a demo. And then before we turn anything on, I’ll just confirm and like what the capabilities are.
196 00:23:16.720 ⇒ 00:23:20.659 Uttam Kumaran: And then what’s the goal, too, is the goal to maximize sales profit.
197 00:23:20.800 ⇒ 00:23:36.959 Daniel Schonfeld: You know, increase conversion rate. You have to obviously just pick one goal first and test it and then move on to the next. There’s only so many things you can test at a time. Yeah. And again, I think to get to get us into habit is like, I’m hoping these guys are good because it would be great to be able to run.
198 00:23:37.120 ⇒ 00:23:54.010 Uttam Kumaran: you know, have tests running where we’re we’re testing things out. And again, they don’t have to directly affect discounts or pricing. They can be content tests or other. Otherwise. Again, I think the test that we’ve been running on the marketing side has been even Kim modifying things on the very, very top of funnel.
199 00:23:54.020 ⇒ 00:24:01.209 Uttam Kumaran: like email messaging. And so II think there’s probably a ton of room within the
200 00:24:01.340 ⇒ 00:24:03.390 Uttam Kumaran: like shopping experience itself.
201 00:24:03.520 ⇒ 00:24:08.059 Uttam Kumaran: Now, I’m just having one at least, having a goal like, we have one test running.
202 00:24:08.260 ⇒ 00:24:11.969 Uttam Kumaran: And then hopefully, we can put that into Kim’s hands to to be able to execute.
203 00:24:12.080 ⇒ 00:24:23.019 Daniel Schonfeld: Yeah, there should always be test. You’re test we running 24, 7 yearout all the time all the time. I agree and thank you for for stepping in there and doing this, taking the initiative on it.
204 00:24:23.260 ⇒ 00:24:28.650 Uttam Kumaran: The the other thing I wanted to, I put in the email is, there’s a software called Microsoft clarity
205 00:24:28.690 ⇒ 00:24:37.120 Uttam Kumaran: that provides heat maps free of, like the entire. Another tool that was was like recommended by a lot of people.
206 00:24:37.530 ⇒ 00:24:53.039 Uttam Kumaran: I think we should just try install that, and we can just make sure that it doesn’t affect performance, and I’ve read a lot about their docs. It shouldn’t do anything. Is it? The same as like a hot jar. It’s it’s basically a hot jar. Yeah,
207 00:24:53.250 ⇒ 00:24:57.369 Uttam Kumaran: that way again. In this testing process, it also will record sessions.
208 00:24:57.490 ⇒ 00:25:20.890 Uttam Kumaran: So for the people that do order, or people that don’t order, and pretty much for everyone that orders. We can walk through. What they’re ordering. Process was how they got how they from the from the Gothic site. What do they click on the other things that I noticed of reading some case studies. People are clicking on things that aren’t clickable. People are looking at stuff that maybe we didn’t know they were looking at. That’ll just give us a pretty much, a really.
209 00:25:21.190 ⇒ 00:25:31.830 Uttam Kumaran: a lot of what happens in web traffic analysis is they do like session viewing. So they pretty much record every session of every user, and then kind of looks at that I think this is like a little bit of a
210 00:25:32.020 ⇒ 00:25:38.790 bencohen: an easier step to manage just looking at where people are clicking and their flows. One thing we’ve done this before.
211 00:25:38.940 ⇒ 00:25:48.909 bencohen: Yeah, I can’t remember which which application we use, but I don’t wanna have 2 of these running concurrently. II would probably removed it from our stack. But
212 00:25:49.230 ⇒ 00:25:58.670 bencohen: these are heavy things on the site. They slow down. So just make sure. Make a note, please, like. let’s just verify that we do not have another one running, because 2 is
213 00:25:58.910 ⇒ 00:26:06.499 Uttam Kumaran: okay. Okay, yeah. And then I’ll before we install anything, I’ll try to find out where we can look at the site speed.
214 00:26:06.750 ⇒ 00:26:11.780 Uttam Kumaran: analytics that way. We have that.
215 00:26:12.330 ⇒ 00:26:21.150 Daniel Schonfeld: I got all that. I’ll I’ll help you figure that out.
216 00:26:21.980 ⇒ 00:26:26.060 Uttam Kumaran: The other things are.
217 00:26:26.850 ⇒ 00:26:30.039 Uttam Kumaran: yeah, I’m I’m just gonna keep pushing Pat to make
218 00:26:30.180 ⇒ 00:26:36.810 bencohen: those changes. We talked about. Yup, Dan, I know, had different priorities. He thinks that there was
219 00:26:38.030 ⇒ 00:26:39.690 bencohen: overload.
220 00:26:39.750 ⇒ 00:26:41.090 Uttam Kumaran: So yeah.
221 00:26:41.130 ⇒ 00:26:46.179 bencohen: like, let’s have more of a importance hierarchies. So we see the most important things. First.
222 00:26:46.360 ⇒ 00:27:02.049 Uttam Kumaran: Yeah. And so I’m I’m pretty much. I pretty much told him that I said, Hey, you’re gonna get 2 perspectives, but I like that. He has like a he has, like some principles in the way that they’re translating information like, lean on that to explain your decision making. But
223 00:27:02.430 ⇒ 00:27:07.239 Uttam Kumaran: move quickly on like some of these dashboard updates, and let’s try to get an update by
224 00:27:07.370 ⇒ 00:27:11.199 Uttam Kumaran: Monday or Tuesday. Of some of the changes.
225 00:27:11.650 ⇒ 00:27:25.960 Uttam Kumaran: and then I also we’re talking to Kim. Also, I know we were talking Kim a bunch about trying to get her onto using some of the dashboards and things like that. So hopefully, some of the efforts that we’re doing on the marketing side for discounts.
226 00:27:26.090 ⇒ 00:27:30.349 Uttam Kumaran: We’re looking into how she can start using white dash.
227 00:27:30.510 ⇒ 00:27:31.550 bencohen: Yep.
228 00:27:31.870 ⇒ 00:27:42.049 Uttam Kumaran: And then the last thing is we, everything’s moved off of Mexla, basically. So I saw you told them to fuck off permanently. So I think we’re
229 00:27:42.170 ⇒ 00:27:43.930 Uttam Kumaran: that there.
230 00:27:45.410 ⇒ 00:28:02.490 Uttam Kumaran: Good. Alright. Well, it was. I think it was a good week. Are you on our slack yet? Or did I forget to invite? I’m on the slack. No, no, no, let me. I’m so I thought II no, II thought about messaging, and then II was just like busy with some stuff, but
231 00:28:02.520 ⇒ 00:28:04.640 bencohen: one sec. One sec. Let me.
232 00:28:05.760 ⇒ 00:28:12.449 bencohen: I don’t even remember I don’t. I don’t use I don’t. I used to use stuff like add coworkers. Alright.
233 00:28:21.880 ⇒ 00:28:30.939 bencohen: alright! I just invited you, and here’s a link to join. Let me know when you’re in. We don’t really use a lot of channels. We more just have a few group chats.
234 00:28:31.040 ⇒ 00:28:32.610 bencohen: but this way. You can
235 00:28:32.900 ⇒ 00:28:44.109 bencohen: just more easily talk to Kim and Cody just tech you know, message them. I don’t like email particularly. It’s sort of too formal when most things are just a yes or no.
236 00:28:45.200 ⇒ 00:28:55.499 bencohen: But I gotta run to actually get 1030. So let’s let’s circle back later. Let me know if any issue with the invite it was Utam, and pull parts to go
237 00:28:55.610 ⇒ 00:28:57.480 bencohen: alright. Good! I’ll talk to you later.