2,999 Conversations

Sandesh: I figured instead of doing an intro, most of you, I figured there's a better way to do that. How many of you here are app devs or have a public app in the Shopify App Store? Okay, awesome.

How many of you were around during COVID when they changed the App Store pricing model and put more money in your pockets? Okay, you're welcome. Part of that was me.

Okay, you wanna save that though. How many of you have used collaborator accounts and/or have seen a collaborator request code? And how many of you hated?

You're welcome, that was me. So yeah, love me, hate me. I've been in and around Shopify for a while.

And I run stock. We do pre-orders for Shopify stores. Been doing it for four years.

And I'm pretty much a half good engineer. That's what I call myself. I run the business, I write code, I do whatever I can to help keep the app running.

And I worked for two years at Shopify, I worked on a bunch of things, App Store, Collaborator Accounts, Payments, things like that. It's been seven years in Shopify, which is kind of hard to believe, but still around still kicking, very happy to be here. And I have nothing to sell you.

I actually, I wanted to at some point in my life, talk to people and give them some ideas and things that I've learned in my time building stock and building pre-orders. So this chart here shows how many conversations we've seen over the months building stock. And if you look at October 2025, that was 2,999 conversations in that month.

That's a lot for a team that started with about 10 a day. So we went from 300 to 3,000 conversations And I burnt out. I burnt out completely.

I just couldn't get myself off the couch. I just used to play, I played video games for that entire week because that's all I could think about doing. I just couldn't deal with the amount of work, with the amount of problems, with the amount of issues that we had in the business, the number of merchants who kept complaining about different things.

And in that moment, I really wished that somebody was there to help me and I really wished that that I got a moment where I could tell people, hey, these are some things that I learned, and I wish you take it with you wherever you go so that you can avoid the problems and the mistakes that I made. So that's pretty much what I'm gonna talk about today. Just four lessons, four very simple, very straightforward things.

I don't even think it'll take 15 minutes, just 10. And hopefully, as you keep growing your business, as you keep building new things, as you keep hiring more people, you remember them. So, 2023, we ran stock.

I was previously called Restock Rocket, and we used to run it off of a WhatsApp group. So it was just me and my co-founder. Every day, I lived in Toronto here, and he lived in Bangalore.

So we would get four hours to catch up. And what we do is we would leave really long messages with each other, and that's how we used to run the app back then. So if you've been there, or you are there right now, that's just how it goes with building an app.

And one of the things that we used to do a lot, we spent a lot of time on, is thinking about how the app is actually structured. So we'd spend time on hours just debating about what the database models look like. And that has paid dividends in way more ways that I can describe.

And that's pretty much the first lesson for the things that we've learned over the years. Primitives compound and features depreciate. If you're taking the time to think about how your app is to be structured, if you're thinking about what a database model looks like, what are you even going to name it?

What are merchants even going to see when they look at your dashboard, when they access it for the first time? And you're taking an insane amount of time to do it? That's totally fine, because four years later, the words we decided on have stuck around.

The database models that we decided on have stuck around, they haven't changed. Now, you often hear contradictory advice as well. You hear a lot of YC talk.

You hear a lot of other people in the startup world saying, you have to move fast. These things can change. But let me tell you this.

After four years, I've seen it over the years. Every month, there is a new flavor of a problem. And it never stops.

So that's one point, your support is bad. Another point, your email delivery is bad. Another point, your webhooks are bad.

Thank you, Hooktech. And then at another point, you have to deal with LCP. Shopify changes something.

So the problems just never stop. And if you can ever get a moment to think about what your primitives need to be, or what are the words that you even use in your app. So for us, for example, we stuck with three words for how we represent pre-orders or back in stock alerts on the app.

And we stuck with those exact words four years later. We stuck with those exact models four years later. And we've just been able to build on top of it.

So if you're there thinking about how your primitives need to be, and you're taking time with it, please do, please take more time. Okay, 2024, we're no longer two engineers, we hired our first set of people, there's two other engineers there and somebody in support. Honestly, the workload got too much, and we were thinking about where do we go from here, how do we even get the time back to invest in the business.

For folks who have already built an app, this is pretty straightforward, like at some point, The problem becomes so much that you just have to invest in it, that you have to build a team. But I think that is one mistake that I still see happening a lot between app partners and app developers and even people who are just trying to build a business. You ideally want to think about your business just like your product.

Whatever is a system, whatever you can separate out, you should be going and doing it sooner than the need comes. and it pays off. And I'll give you an example of how we did that.

So we hired a support team in 2024, right? So it was just one person. I remember writing a 40 page handbook to them about how to do support.

Everything from how do you say hello to how do you ask for a review. And three years later, they've expanded on that in so many different ways. I can't recognize it, but anymore because I don't do support.

But as you can see in all their own ways, they've built on top of that system. And this only paid off because we decided to separate it into its own team at that time. We made a call that we're going to see and deal with a lot more problems in the future.

And the best thing we could do is separate that out. And we've done that even more. And I can just tell you, like three years later, we are in a much better spot, just thanks to that.

Moving on to 2025. So if you saw along the way, we were dealing with 10 conversations a day to 100 conversations. And now we're at 1,000 a month.

So we got built for Shopify. We were selected as one of the top 25 apps for built for Shopify. So we've been spending a lot of time building out our product.

And one lesson that we learned that was really painful was that nobody- people used our app, merchants loved our app, but they didn't know who we were. They didn't remember us, they didn't recognize us. And that's because in 2023, we made a stupid decision to name our app Restock Rocket.

And because the app store loves having keywords in the name, we just shortened it down to RR. So you can imagine that we built the best possible app we could. We had a great team.

We were really proud of everything we did, but all we could say is, "Hey, who are you?" We are RR. That's pretty much it.

So this is the mistake to avoid, that I think most people need to look into. Just as soon as you can, start thinking about what a brand looks like. Because that is something that nobody will ever be able to copy.

They can copy every feature they can think of, you can think of, they can copy every little documentation you've written, but they cannot copy how you think and how you present yourself to a customer, how you present yourself to a merchant. And today, we made that decision last year to change from that horrible name before to what it is now. I still think our logo is bad, by the way.

But I know that this is the direction that we want to go. And even more so with AI, it's so much more important, because what you're going to see in the app store going forward is every app being a clone of the other. It's almost too easy to tell Fable now, like, hey, just build me this thing, and it'll go ahead and do it.

If you've been around on YouTube for a long time, how many of you guys watch Linus Tech Tips or MKBHD or any of the consumer tech guys? So if you've seen, there's been an explosion of creators. And this is what the internet does.

It democratizes the crap out of everything. And the same thing is going to happen with the App Store. If you think there are lots of apps today, the hell is yet to come.

We're yet to see it all. And the only thing that will survive and stand the test of time, in my opinion at least, is brand. There's a reason why we still remember line of tech tips, MKBHD, all of these guys is because they've built something that stands at test of time.

It's the intangible. Nobody can tell you why they're good. You just remember that they're good.

And it's because they invested in that process of thinking about what the vibe is early on. And so we're doing the same thing on our end. We're changing how people perceive us.

we're changing how we want people to think about us. And I just think you need to do it when you can, as soon as you can. And now we're here in 2026.

We have 30,000 stores, we're dealing with 3,000 conversations a month, we have many different apps, we're a team of 20, and we made one big mistake along the way, which is that we did not adopt new workflows while things were growing. I think if you've been building Shopify apps for a while, if you show up to work every day, there's almost like a brand new flavor of a problem. It just doesn't change.

And so you end up spending a lot of time optimizing those things, right? I'm going to make it 10% better. I'm going to make it 5% better.

I'm going to think about how things need to change in this way or that way. But you never think about how am I going to just transform what work is like. And so here's a piece of advice.

Please adopt new workflows while everything is still on fire. It sucks, it's painful, but you should do it. You should invest or separate out the time to think about how the business can change.

And I'll give you a simple example of how I did that. I'm assuming there's a lot of devs here, so maybe you'll like this. I bought a little Linux box because I wanted to stream my movies.

And I set up a display on it, and then I set up Cloud on it. And then I was like, okay, you know what? I just need to be able to access it all the time.

And then I set up a remote desking software so I could just access it from anywhere on my computer, from any other computer. Then I was like, that's not enough. I want it on my phone all the time.

So then I just found a way to SSH into it. Then I was like, that's not enough. I need to be able to stream the terminal on my phone at any point.

Then I was like, that's not enough. I want to just be able to chat with it like an agent. And through doing all of that, I learned that agents are extremely powerful.

I know the lesson kind of sucks in this, But the point is that the agent, the workflow of how to get to building an agent was something I learned through spending time on the weekends outside of my work because I just felt like it was important for me to learn. And that led into us building an internal agent that's so good today that people just can't work without it. This didn't exist like two or three months ago.

It just happened because I was just sitting there thinking about what do I do with everything that's happening in the world with new technology. And so even with all the shit that you have to eat every day from every merchant, from every customer, from every problem, take the time out and it's totally okay to think about what new technology you can play with, what new technology you can add to your business and how you can change things because before you know it, the world will look radically different. Every screenshot you see here is things that our team used to do manually before and now they're just getting an agent to do it.

They're no longer triaging support tickets. They're no longer looking at app store ranking. They're no longer looking at very specific customer problems, no longer looking at how to migrate APIs.

Everything is being handled by the agent. And their job every day is to just make sure to come up to Slack and figure out how to get the agent to do more stuff. And this did not exist two months before.

So all to say, if you have the time and if you can, please spend it on figuring out how to transform your business. So that's it. Four simple lessons, primitives compound, separate your roles, build a brand, and adopt new workflows.

Thank you.

MC: Okay. We can take a couple of quick questions, but some real quick housekeeping. Folks in the back, hi.

If you're going to have ongoing conversations, could you go ahead and just do that outside for a little bit? Because it is pretty loud in here and we definitely want to be respectful for speakers. I've had already people texting me saying up front that they're having a lot of trouble hearing.

So, we're kind of living in with this space and that it's just one big giant room. So it's a really beautiful day outside. If you want to just step out, come back in.

We've got a security guard out there, but he's going to let you back in and it'll be fine. But we really want to make sure that we give speakers a good, respectful space so that way they can help communicate, especially the ones that have helped make this happen today. So we'll do a couple of quick questions and go from there.

So any questions? You can talk if you're going to ask a question. So sorry.

Audience: How did you recover from your burnout? Any tips for that?

Sandesh: Well, I played, have you guys heard of roller coaster tycoon? Yeah. Oh, Jordan asked me how did I recover from burnout or how did I recover from it at that time?

I played a lot of planet coaster for a week. That really helped. But the real thing that helped me was writing more.

So I tried to journal now a lot and I tried to remind myself that whatever problem I have today is just today's problem and it will be gone tomorrow. It doesn't help with the anxiety as much but it certainly helped with just writing it and seeing it and understanding that it's a frame of mind in a moment of time and that separation has helped. I still burn out often though so you know still learning.

Audience: Thanks, Sandesh, for the great presentation. I'm wondering, like, how do you decide your priorities, because you receive a lot of feature requests? What do you say yes, and how do you decide Yeah.

Sandesh: In the early days, everything was a yes, because that's just how it goes, you need your customers. Nowadays, we say no to most of the things, and I think it's the difference between building a product versus building an app. If you build an app, you're trying to service everyone who comes across to the door because you just need the business, but at some point, your app becomes a product.

It solves a real business need, it solves a challenge, and it solves it in the way that you think is best for the market. And it requires a real change in mindset that what you're building needs to stand, the test of time needs to stand long term. And so today we say no to a lot of things and we always think about if we're going to build something, is it something that will help everyone and help us accelerate our goals?

So when we do planning internally, we have five different initiatives. If you've used linear, you've probably heard of the word initiative. It's basically like, what is your plan for the year?

And so we say, this is our goal for the year. And we're going to do everything we can to make that goal happen. Whatever doesn't belong to that goal is not something we're going to do.

And we've got an increasingly more, let's just say we say no more often than we ever have before. There are still things that our internal coding agent does because it can. But most of the times we say no to whatever merchants want because we think we understand the market at this point better than they do.

MC: For one more.

Audience: Well, great lessons, thank you so much. I was wondering how do you actually do change stuff while everything's on fire? Because typically that doesn't just mean getting over yourself and maybe taking a week off, but that typically also affects your team. And just yesterday with the AI talk, we had the conversation of the team maybe not being ready to use AI or the tools that you would like to see them use. How do you get over that for yourself and the team?

Sandesh: I think a lot of it, and maybe this is a simple answer, I think a lot of it has to do with showing the way. Or rather, on top of that, also being very, let's say, fungible yourself. Like, I don't hold myself to any particular ways of working.

I just always think about how can I be better? And what can I do to really change how we work? So I might enjoy writing code today.

I probably do it tomorrow because that's just not what is important to the business. And I think when you have a team, and this happens across every business, the biggest to the smallest, people just won't change as easily. I think they will stick to what they're doing because it's what they know best, but they need somebody to show them the way to set an example and they will follow along.

So a lot of what I've done over the years is just been like I'm just changing how I work and people will see it and they will adapt it. Something as simple as the internal agent, I stopped writing and doing all the work myself. I just went to Slack and started pinging the agent to do things and people caught on and now they're doing it themselves.

So you really just wanna set an example for your people because that's the best way they will learn. They feel comfortable knowing that you're doing it and they will follow along.

MC: All right, thank you. [APPLAUSE]