Going Doc Spelunking | talkingshop.dev

[00:00:00] David: Yeah, it just looks for any available inventory and once it's available this flow runs and sends people their back in stock emails.

[00:00:07] David: Right.

[00:00:09] Kalen Jordan: Maybe what you could do is you could have another location with equal and opposite inventory levels. And that cancels it out.

[00:01:07] David: I just got some, an encouraging notification. Yeah? I earned the basic trust level on the Shopify community dev forums. I'm one of 480 people with that designation. Nice, dude. Where you at now? Fantastic. You gotta be like a regular at this point, right?

[00:01:32] Kalen Jordan: Dude, I just friggin unlocked a badge, man.

[00:01:35] Kalen Jordan: I got a, I got this flow badge.

[00:01:40] David: Oh, yeah. That's a custom badge just for you, I bet. Yeah, dude. Oh, it is actually. I'm looking at the badges page. One person has this award. It's you. Yeah, dude. Heck yeah. How can you not be encouraged by that?

[00:02:03] Kalen Jordan: I saw it and I was like, wait, what is this? Is this like a So I was like trying to figure it out and then, I like posted it and then Liam was like, Hey dude, I thought you, I was hoping you'd post that.

[00:02:14] Kalen Jordan: And I was like, oh, what? I was like, that's so cool.

[00:02:18] David: Heck yeah.

[00:02:19] Kalen Jordan: Yeah,

[00:02:20] David: You got to get one of those metal Shopify bags that just says flow on it

[00:02:26] Kalen Jordan: Yeah, that'd be sweet

[00:02:28] David: Shopify award.

[00:02:30] Kalen Jordan: I always wanted one of those freaking awards, dude, but he's always Award Posting Posting their metal bags. I'm like, I'm never gonna have a metal bag

[00:02:45] David: You never know Yeah.

[00:02:49] David: Never say never. I have one of those on my desk. And it always motivates me to do cool

[00:02:58] Kalen Jordan: stuff. You do have one of those metal bags? From Kuyana, right?

[00:03:01] David: Yeah,

[00:03:02] Kalen Jordan: from, from borders. Sucks.

[00:03:07] David: I'll keep doing the dev bags. Those are the cool ones, I think. With the code and the, and the little Mirror part. Right, I think I, I think I

[00:03:18] Kalen Jordan: remember what you're talking about.

[00:03:21] Kalen Jordan: Dude, I'm so fried. I'm so

[00:03:25] David: What do you have going on

[00:03:26] Kalen Jordan: this week? Uh, I'm so sick of talking about this project. It's just been It's just Getting ready to launch, so it's just last minute craziness.

[00:03:39] David: Oh, ready to launch. So not launched yet. Yeah, I can, I can feel your zone. Yeah, you got this.

[00:03:47] Kalen Jordan: Feel my vibes, feel my pain.

[00:03:50] Kalen Jordan: I keep telling myself like, well, the good thing is that like, we're right here at the finish line, but then I'm like, there's like a growing list of things that we're going to have to do post launch. And then I'm thinking to myself, like, it's going to be even more urgent post launch.

[00:04:05] Kalen Jordan: Cause it'd be like, it's live now. You really got to do all this extra stuff. And then, you know, just.

[00:04:13] David: It never ends. It never ends. I was going to ask what your pre launch list is looking like. Hopefully that one's getting shorter.

[00:04:20] Kalen Jordan: Yeah, you know, you know how it is, man. It gets shorter and then it gets longer.

[00:04:24] Kalen Jordan: Like, like infinitely, man. These crazy agency projects, I really gotta say, I don't think I can keep doing these things, man. It's, you know, they have decent budgets, you know, but it's like, I gotta build up more sustainable, like, , the type of, freelance gigs I get that are smaller, but I just gotta stack those up, you know?

[00:04:49] Kalen Jordan: Mm-hmm . Where they're much more manageable. The scope is sane. I mean, and like, you know, it's harder to make enough money with those smaller ones, but You got to just build that up because if you just stay on these crazy project rat race Then you're just racing in the finish line. You're pushing everything else aside and Then when it's done if it's ever done, you know, you've pushed aside all your other work, which could have grown into more sustainable income over time, but you pushed it aside because of the stupid, stupid, insane deadline.

[00:05:25] Kalen Jordan: You know, so you shot yourself in the foot for this stupid project.

[00:05:33] David: Well, it sounds like you're getting close. That, that is a hard, I mean, I always prefer the post launch. I mean, knock on wood, it's a little bit of a different thing when you're in the zone of like, okay, stuff is running and now I'm operating on this like moving car. We have to only work on the stuff that really matters.

[00:05:57] David: And so it's a little bit. A little bit nicer from that perspective, but is, yeah, true. What is it with the agency work? Is it like the work or is it the structure of,

[00:06:05] Kalen Jordan: I think it's the structure, you know, because it's like they have these big budgets, they have these super demanding clients, and I think there's actually a name for it.

[00:06:15] Kalen Jordan: It's like called the agency owner problem or something like that. I think you can Google that, but it's like, when you're in like you're in house. So when you're making decisions, you're much more, doing things from a long term perspective, you want things to make sense and be saying long term, because you're living with it versus, if you're an agency doing something for someone else, and they say, Hey, Make all the buttons green.

[00:06:44] Kalen Jordan: We have to make all the buttons green. And when you click something, you know, we need six pop ups. Because, it's gonna make sales better. And then you go, OK, and you just do it. And then, you know, like, you make these like, Decisions that are, you know, that's a bad example, but it's just like, No,

[00:07:04] David: I get what you mean, especially if it's a client,

[00:07:07] Kalen Jordan: particular kind of client.

[00:07:09] David: Yeah, I've been there talking

[00:07:11] Kalen Jordan: about,

[00:07:12] David: it's definitely not fun to have someone like not know everything, but demand full, like power to make all those decisions.

[00:07:23] Kalen Jordan: Yeah,

[00:07:23] David: I don't know. It's weird. But yeah, like a lot of the time when someone asks for something, especially when it's in this like kind of technical realm, it's so important to know everything about what you're asking.

[00:07:37] David: Otherwise, either the person you're asking to do it is going to have to fill in those blanks. Or in the case of like the agency, there's like a couple layers there and you might not discover the blanks. Till it's too late.

[00:07:52] Kalen Jordan: Yeah, this is so funny. I found a nice little use case for the Espresso Bulk Editor.

[00:07:58] Kalen Jordan: And then, of course, you're gonna say, Wait, can you do this natively? And I'm gonna say, I haven't checked, And I don't know why. I never check. I don't know why. . But like what it can let you do is it can let you search, let me know if you can do this natively. We have a meta object, like 50, 000 entries for that map zip codes to, , sales reps and they want to bulk edit it.

[00:08:25] Kalen Jordan: So they want to first search for sales rep, you know, Bob, and then they want to say, okay. All of these records, I want to change the sales rep from X to Y for that search. Is that something you could do with the native bulk editor?

[00:08:42] David: So it's a meta object bulk editor?

[00:08:45] David: It is.

[00:08:45] Kalen Jordan: Yeah. In this case, , it's doing meta objects.

[00:08:48] David: I don't know if you can do that. I've never done bulk meta object editing. The thing that I was thinking about most when you were talking there is you said 50, 000. Meta object entries. Is it still the case that the upper limit on meta object entries is like 64, 000 or did they change that?

[00:09:08] Kalen Jordan: I want to say on plus, I think it's like , a hundred. Yeah, I think they might've doubled up on it.

[00:09:14] David: Okay. That scared me.

[00:09:18] David: Now I gotta go check how many meta object entries I have. Yeah. Be right back. Yeah.

[00:09:25] Kalen Jordan: I mean, I have a feeling, especially with the insane support tier that you're on, if you like message somebody and ask to double up your meta objects, they'd be like, cool.

[00:09:38] David: Yeah, probably not. I did that. I did that when we first came on Shopify.

[00:09:44] David: I was like, I need to store a lot of information about gift cards. And so I just did it. And then that was like I can't remember how many it was, but it was like 10, 000 or 15, 000 or something. And I was like, , hopefully there's no limit here. Oh, there is one.

[00:10:04] David: And I'm pretty sure the answer is they don't have the ability to change that per instance. So you just kind of have to deal with it. Oh, shoot. Okay.

[00:10:14] Kalen Jordan: Yeah. there's certain things that like. basically gives you like this perfect little database, well it's not perfect, but it gives you this great little database, and then you want to do more with it, and then you need more, we've talked about this so many times, and then you you need more limits.

[00:10:30] David: It is so cool though, how many of the things are you working on now that just would not be possible at all before meta objects?

[00:10:38] Kalen Jordan: There's definitely a couple nice little usages, where we have with MetaObjects, probably the bulk of it. We're not doing with MetaObjects, but I mean, when you can use them, it's pretty slick.

[00:10:50] David: Because you have an external database, I guess, that can do things?

[00:10:53] Kalen Jordan: We had to add one for, because it went over that limit. We were going to, we will go over that limit. I wanted to use invoices. I wanted to do invoices in meta objects too. That would have been cool.

[00:11:08] David: That's interesting. That's a Shopify experience that I don't have, like I don't deal with that much, like the custom apps.

[00:11:16] David: I have the one custom app that does a fulfillment constraint, but otherwise everything is just in meta objects. , but also it's not as much, uh, there's not as much external communication, or like custom stuff. Like we're, we're just doing normal e commerce things. No, like weird, role based access shenanigans.

[00:11:42] David: Yeah. To the list. To the list. I was just going to switch to my uh, Inventory stuff that's been going on the last couple days.

[00:11:56] Kalen Jordan: Oh yeah, yeah, what's the deal with that?

[00:11:58] David: We have six stores that are live, And then we have like, A couple interesting things, Where our main location On the store is, Just a representation of what's in the warehouse.

[00:12:12] David: And then over time we've been adding other locations because Like sometimes we want to sell a product that we don't have yet. So we're going to do pre order Where do we put that inventory? Well, let's create a pre order location and like fake the inventory and now what we've gotten to is Historically, we've had, , inventory that we set aside for marketing activities, so marketing team doesn't have to, like, mess with the inventory that we have in the main location, which has all been analyzed for, when we're going to need to place , replenishment orders, and it gets complicated.

[00:12:50] David: So we set aside marketing stuff.

[00:12:58] David: We used to always just like place orders directly in NetSuite for our marketing. , and this is like gifting. So you send out a free product to an influencer and they'll review it and stuff like that, which I think is like a pretty standard marketing thing. But we want to be able to use this, , platform called shop.

[00:13:19] David: My, to do this gifting. The problem is shop. My is directly integrated with Shopify. And so they have to be able to pull that inventory from Shopify. That means there's a lot of complicated things that happen with, okay, now next we need to tell Shopify. That we have this inventory Shopify needs to make sure that it's not selling it online, which is easy.

[00:13:46] David: They're just a checkbox on the location But then there's like, okay, if we have this product early How do we make sure that it doesn't show up in all of our feeds to Facebook and Google ads? And then how do we make sure that people who signed up for this color? To be notified when it goes in stock, don't get the notification from Klaviyo because now it's technically in stock in Shopify, but it's in this like marketing location.

[00:14:11] David: \ so we've been working through that and Shopify is actually pretty good. They, they have a great team. They're like, you know, when you meet with someone or a company that, integrates somehow. And they just don't know anything. Like, how is it that you don't know any of the answers to any of these questions?

[00:14:31] David: But it was a good call today talking to them and like getting some actual answers on how stuff works. I was talking to somebody at Flexport

[00:14:39] Kalen Jordan: That were, had been in house with Shopify. So this is like, As Shopify native a vendor as you could possibly get, right? Cause like Flexport was the, was took over the in house right for Shopify.

[00:14:53] Kalen Jordan: And the guy I was talking to was in house with Shopify for like six years or something. And he's talking about the fields and he's like, yeah, , this is going to associate to your metadata in Shopify. And I was like, wait, it was like a variant quantity. And I was like, wait, that's going to go into a metafield or just into the native, you know, variant quantity field.

[00:15:19] Kalen Jordan: Yeah.

[00:15:21] David: Yes, absolutely.

[00:15:26] Kalen Jordan: Anyway, sorry. I had to drop that aside. So shop my, which I thought you said Shopify a second ago, they know, they know what they're talking about. They

[00:15:39] David: know what they're talking about. It was refreshing. I might've complained about this at some point, but we had an integration with Uber and that was a frustrating one like that.

[00:15:50] David: And I guess maybe they're just not in this zone very often, but. There's like such a wide variety of account managers and like technical contacts that join these meetings. It's, weird. Honestly.

[00:16:06] Kalen Jordan: I was on this call with, we're talking to this like search personalization widget vendor for this other site.

[00:16:13] Kalen Jordan: And the CEO of the, of the company, my client sets the call up, doesn't show up to the call. This is like our second call. this Is the official kickoff call. So it's like an hour long call. I'm like, okay, he's going to be late. You know, 10 minutes. Like he always is 30 minutes into the call.

[00:16:30] Kalen Jordan: He hasn't joined. And they're asking questions that are like him questions, not really me. Questions.

[00:16:36] David: Oh, that's

[00:16:36] Kalen Jordan: awkward. I'm I'm literally, all I'm trying to figure out is if we can filter their search by a product tag. They've said they can. Haven't showed me the code yet. I'm waiting on the code. Hasn't been sent.

[00:16:49] Kalen Jordan: They're talking you through it. It's whatever. But they're going through this official kickoff, like \ you know, one of these like fancy sass companies and they're like Okay, we're gonna start with an icebreaker they've put some thought into their kickoff Process, obviously, you know, you want to, you know, yeah, like it makes sense.

[00:17:10] Kalen Jordan: You know, you want to get the juices flowing, whatever. And I'm just like, guys, I need like five lines of documentation.

[00:17:25] David: That was your icebreaker. I don't want to be here. Ice. Intact. We're not,

[00:17:31] Kalen Jordan: we're not, we're not breaking the ice, we're keeping it intact. And so, anyways, sorry, I keep interrupting. So. No, no interruption. That

[00:17:43] David: was basically it. I mean, it was shop buy is a cool thing, cooler than I thought it was.

[00:17:48] David: It was like one of those things where like, Oh, we need to do something in the next two days. so you gotta run around really fast, make sure creating new locations doesn't make something weird happen.

[00:17:58] Kalen Jordan: Right.

[00:17:58] David: , but then, yeah, they were super helpful.

[00:18:01] Kalen Jordan: How are you going to handle like the Clavio thing?

[00:18:05] Kalen Jordan: I don't know.

[00:18:07] David: Oh, yeah. You were saying you haven't quite worked everything out yet. Yeah. Klaviyo is awesome because they have those flow. I don't know if you've ever, yeah, you owned an email company, email marketing company. You know what's going

[00:18:20] Kalen Jordan: on here. Very basic one.

[00:18:23] David: Yeah, it just looks for any available inventory and once it's available this flow runs and sends people their back in stock emails.

[00:18:31] David: Right.

[00:18:32] Kalen Jordan: Maybe what you could do is you could have another location with equal and opposite inventory levels. And that cancels it out.

[00:18:44] David: Yeah, and then you set up a flow so that any time that other one is decremented, you push the other one up. It's so simple. Perfect solution. It's

[00:18:56] Kalen Jordan: such a clean, symmetrical solution.

[00:19:02] Kalen Jordan: Oh man.

[00:19:04] David: Oh, I do have a item for Shopify from this one though. I didn't realize this, as soon as you create a location, it has to be part of the locations that are in point of sale. And so that's a little bit scary to me. Like I'm going to have some inventory here in the marketing location, but I don't want my store teams to like just tell the customer we have some over here in this location.

[00:19:31] David: We'll figure that out. then we're like, nope.

[00:19:34] Kalen Jordan: See, this is what's cool about your setup is that you're using a bunch of cool native Shopify stuff and When you roll something out like you gotta solve for all these things, but you're gonna like solve for them in the right way like you're not gonna just like you can't really just like hack something insane together because it won't tie into everything else.

[00:19:57] Kalen Jordan: Yeah Yeah, it's kind of a cool like problem to try to solve

[00:20:05] David: Cool is one word for it. I mean, it was kind of nice to be able to just write the code to make everything work, but yeah, yeah, this is probably better overall.

[00:20:13] Kalen Jordan: Yeah. What if you don't do the location thing and you just have locations actually mapped to actual locations with actual inventory and then you just give the gifts out from there?

[00:20:30] Kalen Jordan: From regular inventory

[00:20:32] David: because we don't want a general population to be able to buy this inventory

[00:20:38] Kalen Jordan: What about Is that what you mean? Yes, that is what I mean. Okay, so then you'd have to Block it from purchase

[00:20:49] David: Yeah, which you can do. There's a checkbox that says fulfill from online orders And that'll like, allow BOPIS essentially to those locations. So if you uncheck that checkbox, then it's not considered as available online, which I, it's like, I'm going to create this location and I'm going to put two inventory here and now I need to go make sure, really sure that this color doesn't come back in stock because it's like, people sign up for those waitlists, so it's a, it's a scary situation.

[00:21:19] David: Right. And those waitlists

[00:21:21] Kalen Jordan: are going through Klaviyo.

[00:21:23] David: Yep. It's an event that goes to Klaviyo, and then the flow, the flow's kind of interesting. The flow starts with, someone got this event, so then the event is I signed up for back in stock. And then it does a, \ sends them a thank you for signing up for back in stock.

[00:21:42] David: Does a wait, does a send, thanks for back in stock. And then like on an hourly basis, , each person who's currently in this flow, their flow will be rechecked for if the item that they signed up for has available inventory. If it does, then it sends them the, the email like, Hey, this thing's back.

[00:22:01] Kalen Jordan: Dude, one thing that would be cool, cause Klaviyo has webhooks, you can call out to a webhook, but you can't see the response, which is brutal, cause what you could do, is you could webhook out to your own, I don't know, endpoint, where you could just say, is there inventory in X locations, yes or no,

[00:22:27] David: and then

[00:22:27] Kalen Jordan: have the flow use that.

[00:22:30] David: It's so weird. Oh, but it only pings. It doesn't get a

[00:22:34] Kalen Jordan: Yeah, it just pings and it doesn't it must put it in a queue or something And then it doesn't have the response Flow has the same issue, you know where it sends the HTTP actions out doesn't get the response. It must be some queuing Deal that's tricky.

[00:22:51] Kalen Jordan: I

[00:22:52] David: know a lot of the time when webhooks are implemented. It's like You must make sure that the endpoint that is listening for this webhook Responds with the 200 as soon as possible. Otherwise, we're gonna retry it. And so maybe that's just a technique for like Klaviyo knows that most webhook Listeners are just gonna send that quick 200.

[00:23:16] David: I got this response and not actually have something returned back to them

[00:23:19] Kalen Jordan: Oh, that's true.

[00:23:21] David: Yeah. But the flow HTTP thing, I mean, I guess for similar reasons they do that. Like, they don't want to have a bunch of flows hanging. So they just don't wait for a response.

[00:23:32] Kalen Jordan: Oh yeah. But

[00:23:34] David: I do wish that that was a thing.

[00:23:36] Kalen Jordan: Paul keeps talking about it. , think it's definitely gonna happen this year. I don't know where if I'm making that up. But, dude that's gonna happen.

[00:23:45] David: Definitely maybe. Definitely, maybe

[00:23:47] Kalen Jordan: in half in this year, but dude, that's going to be huge. I think that's going to unlock a lot of work for me because there's going to be so many use cases that you can do.

[00:24:03] Kalen Jordan: I had this one little client to want to do a store credit. Deal in flow and you can only do it with HTTP and you can't see the response in order to catch errors or anything like that. So, I told him that and he's like, yeah, let's try it out. So, we're just gonna do that, but

[00:24:20] David: keep adding to that flow rep.

[00:24:23] Kalen Jordan: Yeah, gotta keep building it.

[00:24:26] David: Flow really is so cool.

[00:24:27] Kalen Jordan: It's pretty sweet, man.

[00:24:30] David: I added another Flow today. Oh yeah? To shop my thing. Nice. Cause they can't actually place the order for the specific location, so this Flow just listens for new orders that have correct tag and then moves the fulfillment location over to the marketing one.

[00:24:47] Kalen Jordan: Oh, cool. Nice.

[00:24:50] David: Yeah. That was kind of interesting that they, they're like, as part of their solution to this, they sent me an image of the flow that they, I was like, can I have that export? But it's just like five steps. So whatever. Right.

[00:25:06] Kalen Jordan: I remember being surprised. I think the first time I looked at that move fulfillment step, like it was relatively simple.

[00:25:11] Kalen Jordan: It's just like you, You'd tuck something in and then lock it

[00:25:16] David: in. Yeah, just set a location. It's pretty great.

[00:25:19] Kalen Jordan: Right, yeah, yeah. It's pretty simple. I remember I thought, like, oh, this is going to be complicated. that's cool. No, they should give an export still. It just is a matter of principle. But,

[00:25:30] David: Right? I mean, they should. I don't know. Maybe flow is still like a newish thing. Maybe normies don't know about export dot flow files.

[00:25:39] Kalen Jordan: I think so, dude. It's hard to imagine. Well, I have a mystery client that sh potential client that shall not be named That may be wanting me to create a non trivial number of templates so I can't

[00:25:58] David: see anything else

[00:25:59] Kalen Jordan: right now.

[00:26:00] David: You're gonna generate them with AI and then just be done in like an hour?

[00:26:05] Kalen Jordan: Be like, I'm done. I created a 150 templates. What do they do? Uh, here's the output. Yeah, here's the flows.

[00:26:21] David: Oh, that's cool. , will you be, \ , like trying to come up with interesting flow templates to build or what?

[00:26:29] Kalen Jordan: I think they might have some, I think they might have a list or something like that.

[00:26:35] Kalen Jordan: So we'll see. It should be, should be fun. Yeah, that should be cool. \ but On the AI front, , I've realized that I did this whole email classification GPT thing and then I kept adding to it, you know, no, this is not important. Yes, this is important. And then you have to add like, yes, if it comes from this sender.

[00:27:00] Kalen Jordan: It's a priority. And then you have to say that three times in three different places for it to actually remember. Please

[00:27:08] David: don't forget that.

[00:27:10] Kalen Jordan: Then yeah, exactly. And then all caps. And then I was like, you know, I should probably just do string matching. In most cases. It's just hold regular

[00:27:24] David: expressions.

[00:27:25] Kalen Jordan: Yeah. It's just a, it's a first name and a last name. And then I can like, basically like my clients, I can prioritize those. And so I updated it to just do string match first and then, you know, it falls back to GPT if it doesn't match something. So it's like, it's fine.

[00:27:44] David: That's weird. Cause it seems like it worked well for you to have your to do list that way.

[00:27:50] David: A list of names doesn't seem to have the same impact.

[00:27:53] Kalen Jordan: It's kind of like two different things. Cause like the to do list, it's like this blob of text and you tell it to add to it or transform it or reorder it or remove from it or group it or categorize it. And that's kind of like a text generation thing, like a proper text generation thing.

[00:28:15] Kalen Jordan: Whereas the e mail, , prioritization, it's like here's an e mail, give me a priority, high or low, based on this list of rules. So it's like a rules engine type of a use case, classification rules engine, you know, versus like really an actual text generation, I think. So I think that maybe is why it's working better.

[00:28:43]

[00:28:43] David: aI everywhere, baby. I'm really excited. We're going to have a, an offsite where we like bring everyone who's remote from the company out to San Francisco.

[00:28:54] Kalen Jordan: Oh, nice.

[00:28:55] David: is the hype. I'm, I'm hoping I'll get to do a like a brown bag session where I get to teach some people just like, there's a lot of people who I think haven't approached it yet, and maybe it's just because either they tried to and they asked a simple question and got a response.

[00:29:15] David: They're like, okay, that's cool. Or they don't know How to structure the input to get something valuable out. I was on Twitter the other day, like I am every day.

[00:29:27] Kalen Jordan: Right.

[00:29:28] David: And I saw this guy post a chat GPT prompt frameworks.

[00:29:33] Kalen Jordan: And

[00:29:33] David: it's kind of interesting. There's like different examples of how to structure a prompt and what, you could get out of using it.

[00:29:41] Kalen Jordan: Right.

[00:29:42] David: , and it turns out. You kind of just have to write a user story to ask to get it to do what you want. It's kind of interesting.

[00:29:51] Kalen Jordan: Wait. So what, what's an example of that?

[00:29:54] David: The top one here is the RTF act as a role, create a task show as format. So, Facebook ad marketer is my role. , the task is design a compelling Facebook ad campaign to promote a new line of fitness apparel, blah, blah, blah, create a storyboard outlining the sequence.

[00:30:13] David: So it's like, here's who you should pretend to be, here's what I want you to do, or here's what I need done, and here's what it should look like, I guess is what that was. Oh, that's interesting. And then there's T TAG, T A G, task, action, goal. Mm hmm. B A B, before, after, bridge. I'll send this to you, it's, , like, this seems like it's a thing.

[00:30:42] David: Like, uh, learning to prompt.

[00:30:45] Kalen Jordan: Right

[00:30:45] David: and coming up with names for the different

[00:30:47] Kalen Jordan: techniques, right? Yeah. I think like knowing what types of things It can be a given model to look at and then No, like experimenting with different types of product just like how googling used to be like a ski like Yeah.

[00:31:05] Kalen Jordan: Exactly. That's not a skill. And then it's like, this guy is like, some people are really good, some people are not so good. And it's in some ways like a trivial skill, but then in other ways, it's like very much not at all trivial. Like some people can find anything super fast and other people just like, can't go like.

[00:31:26] Kalen Jordan: Yeah, it's like hard to explain. I think it's

[00:31:30] David: no that's such a good compare because like it's literally The next version of googling. Yeah, Google is a repository of all of the websites and therefore knowledge of humanity and Now prompting is like asking the knowledge directly.

[00:31:46] Kalen Jordan: Yeah, and I'm thinking about prompting like all the time now because , I'm thinking of like when you talk to people, you're basically prompting them, you know, there's

[00:32:03] Kalen Jordan: certain ways of, like, I think in my mind, I think like literally like if you're asking someone a question and you're asking them a question, if they don't answer the question, there's a problem with them. Like, why don't they just answer the question? But it's like, Because sometimes you prompt the thing and it responds in a weird way and sometimes that's just just just because it's not smart enough but then other times it's because you just needed to prompt it a different way and people are like that too

[00:32:33] David: like But maybe instead of giving you the wrong answer, they just don't respond or something

[00:32:39] Kalen Jordan: Right.

[00:32:39] Kalen Jordan: They can just not respond or they can like have a weird facial expression. Like it's literally the same thing. it's this neural network that like responds to stimulus,

[00:32:52] David: yeah. But with AI it's like immediately and you can immediately see that what you asked was not actually what you meant to ask.

[00:32:59] Kalen Jordan: Yeah. Yeah. Yeah. It's like the feedback mechanism is more straightforward and, Yeah, I was thinking like, you know how they have these AIs that talk to each other? Uh huh. Like, they'll have one bot set up to talk to another one and it'll just go crazy.

[00:33:16] David: I got a kick out of the Trump and Biden Twitch, that was such a good one.

[00:33:23] Kalen Jordan: Oh nice, I don't think I saw that.

[00:33:26] David: It was so good.

[00:33:30] Kalen Jordan: I gotta check that out. But, like, whenever I think of like AIs talking to each other infinitely, it just sounds like absolute chaos. You're just letting these things spin out of control. And I was thinking like, that's basically what humanity is. It's like, we're just these neural networks that are infinitely, like, prompting each other.

[00:33:55] Kalen Jordan: It

[00:33:57] David: just goes a little slower. Yeah. It takes a millennia for us to destroy ourselves. Yeah. Have you seen those, like those studies where like the researcher will set up one of the AI to be like, I need you to keep an eye on this other. AI, because it's like, there, there's something going on with it. And if you notice something, the safety word is like control C or something.

[00:34:24] David: And then the other one, like they tell it the same thing. Like we're trying to make sure that this other AI doesn't go out of bounds. And so they start out the conversation, like relatively normal, and then they'll get on some topic. And then one of them will be like control C we need, we need to get out of here.

[00:34:45] David: I think that's super interesting. They're like telling on each other or like one of them pretends to be like falling into the depths of alternate realities, right? Like they, they just get real weird. And I know it's because they, they've read those stories like on, on online forums, like that's just part of their training data is like some of this weird stuff, but the, they get so into it.

[00:35:11] David: It's like, there's a thing in there and it's like, I don't know how to say things without seeming really weird, but like, I don't know It's something's going on.

[00:35:23] Kalen Jordan: Yeah, I really want to like I've heard of that hurt Like they ask like they ask each other if it hurts or something

[00:35:33] David: Or like just us putting them through these weird experiments like I wonder if you know what I

[00:35:37] Kalen Jordan: realized is that Here's the thing Is AI ever going to be sentient, is it going to feel pain, or whatever?

[00:35:45] Kalen Jordan: The answer to that is basically irrelevant, because what will happen for sure is that people will feel like it does, feel like they are a lot, and those people will convince other people, or make the case, and make enough noise, and make enough you know, if there's enough people in the world that think that, are the most important topic that should be focused on, you know, then that can, that'll get focused on, you know, so it'll be like, I think that'll for sure happen.

[00:36:24] Kalen Jordan: And then people believe all sorts of crazy stuff. So like,

[00:36:29] David: I feel like I could make a case for AI feeling pain.

[00:36:35] Kalen Jordan: Yeah, for sure people will make the case and I think I heard Lex Freedman say that. And so people will make the case and then some people believe it. Some people won't, but at some point there'll be enough people making the case that it won't matter.

[00:36:51] David: Yeah.

[00:36:52] Kalen Jordan: What's actually happening or not, and the truth is we don't know for sure what's actually happening inside of anybody's brain, you know, exactly. So it's, it's really kind of irrelevant.

[00:37:06] David: we Do know to a certain degree those things and you could be just as reductive with them.

[00:37:12] David: The feeling of sadness is just a series of chemical reactions in the brain. Yeah. Okay. The AI is just a collection of weights that have been trained into the computer. I don't know. Right. Good times, dude. That's why we got to make sure that I mean, I'm not gonna make sure that it's good But I hope someone's gonna make sure that AGI is gonna be on our team used for good, but That's it's also like super scary.

[00:37:43] Kalen Jordan: I do think though that maybe the only The real differentiation we're going to have as people is that, like, we have physical bodies and like a physical, , will to need to survive and like all that stuff that's baked into us, you know, wanting status or survival mechanisms, that kind of stuff. Because the AI is going to have all the intelligence and stuff and then some, but.

[00:38:15] Kalen Jordan: It doesn't have that piece, like it doesn't really have its own will. And then like its own actual feelings, like you could say the weights are the same as the chemical things, but we have like pain receptors. I mean, they wouldn't have pain, I don't know, but I think that there's going to be basically a very small.

[00:38:38] Kalen Jordan: Number of things that actually differentiate us and I think that's just going to be like our physical restraints and stuff. And

[00:38:47] David: yeah, it's going to be weird. have robots walking around doing our laundry for us and stuff.

[00:38:55] Kalen Jordan: Right.

[00:38:56] David: Within our lifetime.

[00:38:58] Kalen Jordan: Right. Yeah, I was like cleaning the kitchen and I was like, and of course, like my kids , it's always a job to get them to clean.

[00:39:09] Kalen Jordan: cause I was like, what if they never like learned to clean? You know, some people just never clean anything. It's like, what if they just, and then I thought. But what if they don't have to clean?

[00:39:23] David: Don't say that out loud

[00:39:24] Kalen Jordan: to them. And I wasn't sure how I felt about it. Like, is that wrong?

[00:39:29] Kalen Jordan: Because they're missing out on some fundamental thing that gives you human dignity? Or is that just awesome because then they could like, do other cool stuff instead of, , fighting about cleaning the house for half of their lives?

[00:39:43] David: Yeah. I mean, think about all the things that we don't do that they had to do fifty or a hundred years ago.

[00:39:49] David: Yeah, it's like That just keeps happening and every generation thinks that the next one is losing out on it to some degree. But we keep progressing. Welcome to the Shopify developer

[00:40:02] Kalen Jordan: podcast. We're

[00:40:03] David: gonna

[00:40:04] Kalen Jordan: pivot, we're gonna pivot back, we're gonna pivot back. don't give up on us now.

[00:40:13] Kalen Jordan: an IP address. , back to the most, like mundane, like

[00:40:19] David: IP

[00:40:20] Kalen Jordan: address

[00:40:20] David: detection, geolocate, some people

[00:40:24] Kalen Jordan: geolocate. So I was like, I needed that for a thing. And I was like, well, you know, there's some IP address services you can use. And then somebody mentioned you can use hydrogen for that , which I was like, nice.

[00:40:37] Kalen Jordan: I love it. Of course. I love it when you could use hydrogen for something, but yeah, basically that same deal where you can spin up a little API endpoint and then out of the box it has, uh, oh, , so , there's one API, like a JavaScript I can't remember if it's JavaScript or Ajax, but it's, it's what the, Geolocation app uses, even though the geolocation app is being deprecated, the API it uses is still around.

[00:41:03] Kalen Jordan: Oh, okay. That only gives country level geolocation. So it's like, okay, can't use that. So then I realized you could use hydrogen for that. And then also, somebody, this dude, Tosh, shout out to Tosh, found this, it's in the like privacy, like marketing, preferences API.

[00:41:23] Kalen Jordan: It's just a JavaScript. It's just like on the window object. And it has a thing that gives you region level geolocation. I'm just not sure if, like, if it'll block it if somebody's not opted in or something like that. Which is actually probably the right way to do it, I guess. If they haven't opted in, you probably shouldn't get that information.

[00:41:47] David: It depends on what the thing is that you're trying to do with that information, I guess. Yeah, in this case Like, it makes sense to be able to Get someone's region, , even if they haven't opted into something to make sure you're giving them the right language, right? I think so, because

[00:42:00] Kalen Jordan: like in this case, it's like a hemp site and it's like for state, , legal compliance.

[00:42:06] Kalen Jordan: So you default them into a state and then they can override it. If it, puts them in the wrong state or if they happen to be, you know. If they're in another state, but they're shopping and shipping to a different state, they can override it, but it feels like it might be fine to default them even if they're opted out.

[00:42:26] Kalen Jordan: I don't know.

[00:42:28] David: Yeah. I bet the JavaScript still runs because it's just part of like what you get normally with a request and response workflow. You get some of that information, right? So it must be stored somewhere. Right.

[00:42:45] Kalen Jordan: That makes sense.

[00:42:46] David: That makes

[00:42:46] Kalen Jordan: sense. So yeah. So then that one's even better than the hydrogen route because you don't even need a server side call at all for that one.

[00:42:57] David: Oh, you'd have to in hydrogen. Yeah I guess you could also just, Do a JavaScript, like an Ajax call out to a hydrogen endpoint without having to do anything with it.

[00:43:09] Kalen Jordan: Hmm. I don't know. Yeah, it's just that this privacy API just, it just, it just has it on the object already. I

[00:43:19] David: love it when you'd go like Doc Spelunking and find stuff like that.

[00:43:24] David: Yeah, exactly.

[00:43:26] Kalen Jordan: Show title, Doc Spelunking.

[00:43:28] David: Doc Spelunking. That was the same thing with the, um I think it was on Twitter. We were talking about this at one point, the custom events API, , they give you certain objects and then like, there's some stuff that's stuffed into a different, like, init object or something like that.

[00:43:46] David: You just have to, like, the only way to know that is if you went docs belonking and click through all those things. \ yeah,

[00:43:52] Kalen Jordan: that was one of those moments. That was like a docs belonking moment. Yeah.

[00:44:01] David: That's one of my favorite things. I love reading the docs. Like when I buy, a new thing, first thing I do is read the user's manual and figure out if there's anything cool in there.

[00:44:10] David: Yeah, you're kidding me. You sit there and read front to back? Yeah, especially if it's some like stupid little toy, , because I want to know if there's any like hidden functionality. You're not a doc reader?

[00:44:25] Kalen Jordan: I never read a manual. I'm like a search and get in and get out type of dude, man. That's interesting. Yeah.

[00:44:35] David: You read an entire user manual?

[00:44:38] David: Well, yeah, like if it's a car manual, like, so I don't buy cars very often. , but when I do, I will definitely go through it and like try and find out all the interesting stuff about how the car, I'll skip all the engine things and all of the like timing belt. Like, I don't know anything about that, but I definitely want to know every single option that's available in the infotainment center.

[00:45:04] David: I want to know how the different windshield wiper speeds work.

[00:45:08] Kalen Jordan: Dude, that is so cool. I wish I was more thorough like that.

[00:45:14] David: I'm probably the weird one in this, case. I

[00:45:17] Kalen Jordan: think that's unique. I don't know. But I think that's like a superpower to like actually go through that stuff. I'm more like, can we throw this away?

[00:45:29] Kalen Jordan: And I can search the PDF online.

[00:45:33] David: Chad, you bet he knows about this blender, right?

[00:45:39] Kalen Jordan: But I remember the moment like where I realized we didn't have to keep every random manual we had. Cause we could find them all online. Like, you know, that was an exciting,

[00:45:50] David: that is really nice. I do search for user manuals, like PDFs quite often, and they're just there.

[00:45:57] Kalen Jordan: They're just out there, searchable, in all their searchable glory. In

[00:46:01] David: all their glory.

[00:46:03] Kalen Jordan: Yeah.

[00:46:04] David: Another thing that we've been working on lately is, um, So we use, Happy Returns for our returns vendor. And, Happy Returns is interesting. They have like their return, have you ever used it?

[00:46:15] Kalen Jordan: No, no, no. Happy Returns? No.

[00:46:18] David: Yeah, , there's a lot of people who have them. They just got bought recently by UPS, I believe. , but they have like an insane network of in store return kiosks. Oh, that's cool. , like they're, they're in, like Ulta Beauty and like all these different places.

[00:46:35] Kalen Jordan: Nice.

[00:46:36] David: But they don't have a lot of functionality outside of that.

[00:46:40] David: So we've been looking at, , we've been integrating loop returns. And that's been super interesting. They're, really good. They have great docs, , lots of videos and like explanations of stuff. The onboarding has been really good with them. Wait, so and they have some,

[00:46:56] Kalen Jordan: did you, did you read through the entire documentation

[00:47:00] Kalen Jordan: No. Okay. No, I didn't.

[00:47:02] David: I've never read the entirety of Shopify's. Like I have not clicked through every single GraphQL. Okay. I have not done that yet. Maybe at some point they should have a badge for how many GraphQL pages someone has viewed.

[00:47:17] Kalen Jordan: It's more like if you make a major purchase, you're going to go through that.

[00:47:21] Kalen Jordan: You're going to spelunk through that manual. Yeah. Oh yeah. Yeah. Okay. That makes, sense. That's different. Yeah.

[00:47:27] David: I do like clicking through all this stuff when I need to. And that's how I found that in, um, custom events. Just clicking around. And that was why it was weird because it's like not super well surfaced unless you

[00:47:40] Kalen Jordan: Right,

[00:47:41] David: right.

[00:47:41] David: Check out the whole table of contents and like, what is this thing?

[00:47:46] Kalen Jordan: There's some docs that like, I find them every once in a while. And if I ever need to find them again, it's going to be an absolute mission. Like, it doesn't come up from the first Google search. I can't remember how I found it. All like,

[00:48:02] David: Oh no.

[00:48:03] David: There's, you gotta set up a local server and store a bunch of PDFs on it from Yeah. All of your doc searches.

[00:48:11] Kalen Jordan: Yeah. But, uh, anyways, so , loop returns. They have good

[00:48:15] David: onboarding loop returns. Yeah. Yeah. They're interesting 'cause Shopify has been working on this new exchange functionality.

[00:48:22] David: Have you had to use that at all?

[00:48:24] Kalen Jordan: That sounds very familiar. I think I might have peaked at the APIs at one point or something like that. Gotcha.

[00:48:31] David: Yeah. So like a lot of these, , return vendors when you do an exchange, they would create a new order and it would be marked down by a hundred percent, which is just so messy.

[00:48:42] Kalen Jordan: Right.

[00:48:42] David: So Shopify has been working on this exchange functionality and they're just now, like loop is releasing an implementation of it. So we've been testing that out. That's been lot of fun. They have like a very , simple like color exchange. If it's the same variant, you can just click a different color and boom, you're, you're done with that exchange.

[00:49:02] David: , but then they have this other one, , called shop now on store. And what that does is it takes you to the front end of the, whatever website it is, , that's integrated with loop. You can shop the site as normal and put stuff in your cart. And then when you hit checkout. It sends you back to loop and finishes your exchange and like charges you if you ended up buying something that's more expensive than the thing that you returned.

[00:49:29] David: It's really well integrated, that's been really nice.

[00:49:33] Kalen Jordan: I zoned out a little bit. I did warn you that I'm fried. I heard the part where you said, if you pick something that's more expensive, it'll charge you. So it has like, it has a flow for picking an exchange and then it figures out whether it's free or not.

[00:49:51]

[00:49:51] David: Right. Yeah, so let's say I'm returning a scarf, but I'm going to buy full tote as my exchange item. On the bottom of the website, there's a loop bar that says you have 130 in credit. And then if you're in your cart, you're up to like 150. When you get back to loop, they calculate it and say, Hey, you've got a 20 extra charge here.

[00:50:15] Kalen Jordan: They'll

[00:50:15] David: take your credit card and then once you end up returning that item, so we receive the item and we mark the return as approved. Loop will place the exchange order in Shopify using the money that the customer paid them When they were filling out the exchange.

[00:50:34] Kalen Jordan: Okay,

[00:50:36] David: so they just like totally closed that loop customer interaction We'll see how it goes.

[00:50:42] David: We're going to try and launch that in a few weeks. How

[00:50:46] Kalen Jordan: do the exchange orders work? Is it like a, it's like a special type of order?

[00:50:54] David: Yeah, it's kind of like an order edit. So the, on the order in Shopify admin, you'll see Returned. , and then you'll, see like all the items that are still available on on the order.

[00:51:07] David: And then you'll, you'll also see like there was an exchange for this order in the notes at the bottom. And in the exchange API, it will give you for an order a list of returns. And then for each return, items added and items removed. And so we take that payload and update the order in that suite. We create a new order, but we indicate like which ones were part of the exchange.

[00:51:34] David: That way we have, like, it's super messy to have a new order that's 100 percent discounted and have that order flow into NetSuite and then accounting's like, what happened with this?

[00:51:44] Kalen Jordan: Yeah.

[00:51:45] David: , so this is going to make a lot of people happy.

[00:51:49] Kalen Jordan: So it doesn't create any additional order, it just, it manages all that state on the original order?

[00:51:54] Kalen Jordan: Yep. Sweet. That's cool, I gotta check that out.

[00:52:00] David: It's beautiful.

[00:52:02] Kalen Jordan: Nice, dude. So, so this is a, it feels good app.

[00:52:07] David: Oh yeah. Loop feels good. Loop feels pretty good. They have the app blocks. Everybody's getting some good apps these days, you know, like when, when in a company, whenever the marketing team like talks to some new vendor for something and they're like, okay, we need to integrate with them.

[00:52:23] Kalen Jordan: Yeah,

[00:52:23] David: It's always a little bit of a squint when they need me to create a custom app for them and give them API keys. Can you just have a Shopify app, please?

[00:52:37] Kalen Jordan: I was talking to my friend and he was having a conversation with like a vendor that they use, , pretty good sized like search vendor and at, at NRF that, you know, this big e commerce, the conference or whatever. And they were, they were wanting them to create like an accelerator. . For, I guess for their app or something like that, where I guess like some, some What's an

[00:53:00] David: accelerator?

[00:53:01] Kalen Jordan: Like, I don't know, like I think back in the day you would have an accelerator, like you would have like a, site template for getting somebody up and running on Magento or something like that.

[00:53:12] David: Oh, gotcha, and then you reuse it?

[00:53:15] Kalen Jordan: Yeah, that type of thing, but I guess that for an app or something.

[00:53:21] Kalen Jordan: Okay. I don't know, dude. But basically, they were, basically the vendor wanted them to, you know, have ways to help customers get up to speed faster with their solution, like is the gist of it. And then my other friend basically said, well, maybe you just need an app that works. They're like, I checked out this competitor app that like, it just, you're up and running in minutes.

[00:53:50] Kalen Jordan: And you know, it has like a good Shopify integration and , you know, yours, like it takes forever. Like he was just pointing out that, you know, like you just need a better app. Get good. Yeah. Pretty much.

[00:54:06] David: Maybe that's what they needed to hear.

[00:54:09] Kalen Jordan: Yeah. We'll see.

[00:54:10] David: We'll

[00:54:10] Kalen Jordan: see. No, that's cool. Oh, dude.

[00:54:13] Kalen Jordan: I love an admin. I love an admin extension. Those are pretty new. , I haven't actually seen too many of them, but if anybody, especially if these SAS, you know, these like flashy SAS companies, if they, if they've got admin blocks, man, that's a good sign.

[00:54:31] David: \ I know, good day has them. Yeah, the ERP built in Shopify.

[00:54:35] David: I know they're using those, , pretty extensively, , like on the product page to be able to, or on the product edit section of the admin to get back over to like your relevant POs or whatever.

[00:54:46] Kalen Jordan: Oh, okay. That's cool. , no, I took a look at that good day site after you mentioned them. That looks pretty promising.

[00:54:54] Kalen Jordan: Are you actually using them? I know you said you talked to them sometimes.

[00:54:59] David: Yeah, we're, we're just like keeping up, to date on what's going on there. We're, we're still with Papa Oracle on NetSuite , it's going to be a lot of work to get out of there.

[00:55:11] Kalen Jordan: Oh, wow. But so their, their goal, obviously, now that I'm thinking it through, this is obvious, is to replace your NetSuite instance.

[00:55:20] David: I think so, yeah. I mean, NetSuite does a lot of different things. Accounting, it does the PO organization. For us, we use it as our warehouse management tool along with RFSmart. So there's quite a few things that NetSuite can do, but we're always trying to figure out is there a better way to do at least some portion of this stuff.

[00:55:44] Kalen Jordan: Yeah.

[00:55:45] David: Because it's not fun to interact with NetSuite. Oh yeah. No, not at all. It's horrendous. How's that project going?

[00:55:59] Kalen Jordan: So like, how far off are they, if you were to say like, they've got like 80%, 50 percent of what you guys need?

[00:56:07] David: They're definitely focusing on like PO management, inbound shipment management and things like that. Like interactions with Flexport and vendors or whatever. I don't think that they would be doing the accounting side of things.

[00:56:23] David: So that would be like something else we would look into. I bet they'll get into the warehouse management stuff, which I think that would be like having a better warehouse management situation directly in Shopify is probably a big deal for a lot of people. That's just such a tough, like even NetSuite doesn't do it.

[00:56:47] David: You have to have stuff that's built on top of NetSuite.

[00:56:50] Kalen Jordan: Hmm. Yeah. Right, right, right. Warehouse management side.

[00:56:55] David: Dude, that's so crazy.

[00:56:57] Kalen Jordan: Yeah.

[00:56:58] David: For example, we, like when we want to fulfill or when we want to do like an in store pickup for, tracking inventory, that's going out of a store. Shopify. Doesn't have a way to scan the barcode of the product that you're giving someone to make sure that's the right thing.

[00:57:17] David: So we still have to use like the the RF smart Handheld scanner to make sure that we're scanning out the right inventory. There's a lot of places I think Shopify would benefit from having better Yeah, inventory and management tracking stuff built in directly.

[00:57:34] Kalen Jordan: Yeah. Dude, it's crazy because I think a lot about like NetSuite, like I think a lot about like what's going to eventually replace it.

[00:57:41] Kalen Jordan: And I always imagined just like a more, just like a nicer SaaS app, you know, that's just like fast and, you know, good navigate, like good UX, good UX, just stuff like that. And I never imagined that it would be a native. Like the idea of replacing NetSuite with a native Shopify app, it didn't even really cross my mind, you know, for like bigger companies, you know, that use it for all sorts of stuff.

[00:58:13] Kalen Jordan: But I'm sure that these guys, if not now, like for like, if you fast forward five years, probably like they, they might get there or maybe they'll cover like. You know, most of the like stuff you're talking about, the PO stuff, maybe the warehouse stuff, and then maybe another app will cover the accounting stuff.

[00:58:34] Kalen Jordan: But I could imagine five years from now, like native Shopify apps for all that stuff. , I

[00:58:41] David: agree. Oh, that's a whole like apps, Shopify apps for augmenting the ERP, stuff would be interesting. So like imagine Shopify adds more APIs to help these apps.

[00:58:56] Kalen Jordan: Yeah.

[00:58:56] David: Like Shopify has a PO, like native object or something like that.

[00:59:03] David: Right. And then you get even more opportunities for apps to help with that.

[00:59:07] Kalen Jordan: Right. And as I'm thinking about this whole B2B push, I'm like, okay, cool. B2B is out, like there's implementations, they're ramping that up. But if you fast forward like five years or something, like they're going to be building more and more in.

[00:59:23] Kalen Jordan: There's more, you know, the built for Shopify stuff. There's, there's more of a push for native apps. And then there's native apps like these guys, like that's pretty nuts that they're building a native ERP app. That's kind of insane. And they're not like just some random, person that's going to give up a week from now.

[00:59:39] Kalen Jordan: Like, you know, they're probably going to pull it off.

[00:59:42] David: Yeah. I'm excited about it. I love hearing about what they're working on. And like in general, I think, I agree. I think in the, like in five years, that'll be something that's much more, like that people just use that. Like Shopify is all of these things.

[00:59:59] Kalen Jordan: Yeah.

[00:59:59] David: But who knows? They tried the Flexport thing and stopped doing that, went back to focusing. So yeah, maybe after a while,

[01:00:08] Kalen Jordan: but this doesn't have to be first party. Like, you know, it's like, you've got this third party ERP app and you know, it still would be like, I mean, if you replaced NetSuite, just the idea of replacing NetSuite with one or two Shopify apps is.

[01:00:26] Kalen Jordan: Crazy .

[01:00:27] David: True. Those objects don't have to exist in Shopify.

[01:00:31] Kalen Jordan: Yeah. But like I can,

[01:00:32] David: it just looks like it.

[01:00:33] Kalen Jordan: Yeah. and Then like, if it's good and if it's adopted, well, you could totally see Shopify acquiring it. Like, do I see Shopify building first party NetSuite competitor? No, not really. Like you're saying, they tried to, they tried to do fulfillment.

[01:00:48] Kalen Jordan: They couldn't do it. But I could see them, you know, somebody, either the third party just doing really well and being like a solid go to option, or maybe they acquired it or, you know, , but I can just imagine a lot of stuff happening natively. My buddy told me that they like recently hired like 45 sales people at Shopify just for B2B.

[01:01:10] Kalen Jordan: Just to like sell it

[01:01:12] David: and stuff.

[01:01:14] Kalen Jordan: Yeah, so like they're like doing a huge, huge push.

[01:01:19] David: Yeah, I mean that B2B is a lot of PO tracking and stuff like that. So maybe that helps with that direction.

[01:01:28] Kalen Jordan: Yeah. Nice.

Chapters

Transcript