02 - Case Study of Jimmy Brings (with Adam Watt)

Show Notes

A perfect example of how Composable Commerce is helping grow online businesses. Adam Watt, Head of Engineering at Jimmy Brings explains their success story. With guest Marvin van Wingerde, Head of Product at Deity.

What is The Commerce Composers?

Welcome to the Composable Commerce Podcast powered by Deity (Acquired by Ultra Commerce), the leading platform for Composable Commerce. In this podcast we explore the world of Composable Commerce: What is it? How does it work? And most importantly, how will it help businesses grow? We talk with online merchants, agencies and tech companies about their experience in Composable Commerce, including some of the biggest retailers in the world. So, do you want to know everything about it? Please hit the subscribe button so you won’t miss an episode.

00:00:00 Jamie: Hello everyone and welcome to this new podcast. Today we're going to talk with Adam Watt, head of engineering of Jimmy Brings, and Marvin Van Wingerden, head of Product of Deity. We're going to talk about is a very exciting project where we rebuild the new Jimmy Brings website to a full composable commerce environment. Adam, can I ask you to introduce yourself? Sure. Thanks, Jimmy. Hi, Marvin. Uh, so my name is Adam Watt. I'm the head of engineering at Jimmy Brings. And, you know, my role here is to really understand the technology stack and really drive where we need to be as a business, you know, from foundations for scale all the way through to flexibility. And with the real big focus on fast to market products to beat our competitors. Thank you. Adam. Uh, Marvin, can you give a short introduction about yourself? Yeah, sure. Hi everyone. I'm Marvin, head of product at deity. I've been with the company for about four years now, mainly concerning myself, of course, with the product management of the company and making sure that our strategic roadmap is in place. And of course was very lucky to be involved in the project from start to end. So I'm sure we'll talk about that quite a bit today. And I think the entire world of composable commerce is very exciting for us. So yeah. Great. Yeah. I mean, of course we want to know everything about that. But first we want to know what is Jimmy brings. I mean, there's a great story behind this. So, Adam, please tell us. Yeah. Thanks, Jerry. So Jimmy brings is a on demand alcohol service in Sydney, and it really came from humble beginnings, um, with two founders, David and Nathan Besser. Uh, and little known suburb called Bondi in Sydney. So at the start, at its core, it was two guys in a van delivering alcohol in the evenings. And it was that way for a very long time. You know, as a business, we really started to grow. You know obviously what a great service. Alcohol fast into your door. And it really was very groundbreaking for its time. You know, we're talking about an era when there was no Uber Eats or on demand delivery wasn't really taking off in a big way. And then you have these guys delivering alcohol in Sydney in less than thirty minutes. So of course it was a natural hit, especially during summer when you know, everyone wants their beers and wine and without leaving the party. And then today, we're now an Australian wine business in representing every major capital city and a most regional cities as well. Sounds really, really cool. I mean, I remember something during my student time here in the Netherlands called the Beer Taxi. It was called an Charmhaven knows what that is. Yep. And it sounds a bit like that. But here it never evolved into a business like what you have now. And I mean, we've seen the massive growth and I'm sure Covid was a big part of that, but not all. I mean, you did an amazing job building a brand, building a story around it and, you know, bringing that out there. And when you started to grow, you got obviously you're gonna have some technical challenges. Um, so you needed at some point you decided we need to do something different. Why? What what what were the the goals that you were facing or the issues that you were facing that you decided to, uh, change your platform? It's an interesting journey, actually. You know, when I look back at the technology that has existed at Jimmy brings, I think the first iteration was the founders, um, and van drivers with just plain walkie talkies. You know, that was kind of at its core. That's that's how they started the business. Then they moved into Excel spreadsheets, which is amazing that Excel basically ran an on demand alcohol company for a long time. And then we moved into, I guess, the iteration before, um, our new composable stack, uh, which is, you know, for its time, very groundbreaking and certainly benefitted the business in many ways, like it was cheap to build, fast to market, but for us it was leaving holes. You know, you touched on Covid and that was a huge growth period for us. And what we started to see is just the degradation of service. And because the system hadn't really had a major facelift for many years, the tech debt started to rise through that platform. So we really started to see, okay, we need to now patch hardware. We need to now patch our software. And we weren't actually building features, we were just patching holes. And that kind of led us down the journey of, well, how do we go about choosing that next generation of E-comm platform? I mean, we've seen that a lot, people patching holes and trying to stick together, but in a massive growth platform like yours, that's a hell of a challenge. So at some point you decided, hey, let's call in deity. I mean, let's do it different. You did your research and you decided to go for the composable commerce way. You then discovered what we are doing, where we bring all these services together and basically are the main glue inside the composable commerce world. Yeah. So then when we were called in, we this was a massive, massive project and we literally flew to the other side of the world and stayed there for a month. And Marvin then jumped into this project. Uh, my God, this must have been a massive. How did you get it over you? Where did you start? Yeah, definitely. It's a huge project, of course, with a lot of moving parts. So I think the first thing that we really did as a team, and also personally, is to get a really good grasp of the business. Was it a business all about what are their goals, what are their ambitions, and also what is their playing field look like? Because of course, uh, Adam, you already mentioned it's on demand alcohol delivery. Well, on demand delivery is, uh, kind of a different ball game than, let's say, regular e-commerce. So the first thing we did is talked with all the stakeholders, uh, of course, to get a good idea of, okay, what do the various business processes look like? For example, listing products, the key concept there, or one of the main important factors of the entire platform is, of course, that it's very precisely location based. Everything starts off with what is the location of the customer and based on that, what products do we show? What is the price of the products? ET cetera, et cetera. What promos do we show? So that was really the first thing that we did. So had a lot of chats of course with Adam, you know, how does this currently work. What do you want to change? What are the requirements to get a really good feel about? Okay. What what are we actually working with here. So yeah that was the that was the first thing we did when we moved from there. Of course, we uh, already had a few selections of particular third party services. So e-commerce engine for example, and search engines and etc.. And then when we have those processes in place, we started, you know, discussing with Adam and his team to, okay, how can we map that to these various systems that we're, that we're using and make sure that everything, uh, works. All right. So that's really where that journey started and how we got off to a good start. Of course. Yeah. And Adam, a question for you because Marvin just mentioned it like, hey services. And that is the main thing obviously in composable commerce. What kind of services do you use? What do you build yourself? What do you already have? What is working fine? And what are you gonna take from the market? Is it real? Best of breed approach. And Marvin just mentioned there are already some services which were chosen and were started to work with. But how do you choose your services and especially also for people listening to this who want to move into the composable commerce world, how do you choose your services? How do you pick them? Really good question, and certainly something that we dwelled on for a long time. Part of the reason that we went to composable was so that we we don't sit around figuring out how ways to rebuild our checkout and our cart management and, and these type of things. We really want to focus on features and how do we get features to market fast, which is to basically choose platforms that give us the best no code approach. So once the integration work has been completed, what are the tools that sit behind, you know, those APIs? That and gives us a really nice UI that allow our our marketing and our operations people to go in there and make changes that they can see reflected on the on the front end. And I think that that was probably where we really started, was to start looking at who has, you know, a a headless approach that we could utilize to really connect everything together and use data as the glue. And then what features do they have that extend our base e-comm offering? And I'm I haven't really touched on it like Jimmy brings is just not your general point and click e-commerce platform. It's, you know, we we attempt to deliver alcohol in under thirty minutes. And to do that, the shopping experience is very localized down to the individual suburbs that we service. So that complexity that we have, you know, we don't really have stock on hand, which is another big change that we have in, I guess, from traditional e-comm. Um, you know, we don't decrement stock and we can shift stores and we can route drivers around because we're, we're more like an Uber model. So yeah, it's a very big shift when you start talking about what platforms fit into that role, and it was really functionality. Do they tick the scalability box and do they have a headless approach? And I think one thing that you mentioned, that's one of, in my opinion, really key in what composable commerce brings you, great benefit is that you say our business is not a click and buy business. It's really complex. And these services out there, like search and product and stock services, are not made for cases like that. I mean, they're made for the more general cases and not for something very obscure. Let's call it bespoke. Maybe a better word, uh, as the business that you're building. And then still we have all these services out there, and they're gonna have to work together because you can use these, let's say, more simple services in a very complex environment. And that is, again, where Marvin comes in. I mean, the services are there now and the business is very complex. How do you manage all of that? Yeah, of course, I think that's one of the key elements actually, when we talk about composable commerce in general. And, you know, you have this set of services, and of course, you can work with them individually and you can, in most cases, nearly all cases, you can connect via an API and, and make it available to your front end. But really, the essence is about making those services work together and to make them interoperate with each other. And that's, of course, also what Jimmy brings needed. So you can't just have BigCommerce isolated and, you know, not make sure that it works with the internal ERP system, for example, you cannot also have just algolia running in isolation without it being aware of, for example, stock or availability or such things. So again, actually, that is why we started with, uh, making sure that we understand these business processes. Adam also gave us a very nice presentation, actually, when we were in Sydney about our current application landscape, because there's also quite a few, uh, internal microservices, I think, Adam, you built over your time at Jimmy. So we have a stock service, for example, we have a black listing service and such, and that really helped us in figuring out, okay, how do we make sure that these processes are reflected in the various systems so that the life cycle is actually accurate. So, for example, we really started out just putting things on paper and making sure that the data models are correct and that, uh, various updates, for example, placing an order or listing products is correctly done from the process point of view and not necessarily one isolated service. Yeah, it was great to get back to the whiteboard, actually. You know, it was nice to to be in a room with people, you know, post Covid and, and actually sit there and draw out architecture on whiteboards and, you know, move things around. So yeah, it was a really good experience. I mean, I was there guys, and it was amazing. And everybody who listens to this, I think it's really important we see these remote oil companies. And I think sometimes it's important to just get back to the drawing board together and just, you know, put your heads literally together in a room. And of course, at the gym it brings have a drink and enjoy, uh, working together. Okay, back to the questions. Because, Marvin, I think you touched some really interesting point that you say, hey, we have to make sure all these services work together. This data goes to the right place and at the right moment. And the more traditional way is where we're going to connect all the APIs together. And then, you know, we're going to get a spider web of drama when we have so many services with what we could provide them. Data was really a middleware in a more technical way, but what we call our Commerce Composer layer, where we can connect all these services separately and then have that data being orchestrated, managed, mapped and etc.. Can you tell a little bit more about that, you know, tweaking in that middleware and that commerce composer versus building separate integrations? Yeah, of course. So I think uh, especially in the more, let's say, monolithic e-commerce engines, is to basically build in all the functionalities into the commerce engine. Of course, that has changed quite drastically with the whole composable commerce movement, uh, where we actually pick best of breed services that are independent of each other. And I think with that, the requirement comes to have that composer making sure that everything basically works together, as I mentioned earlier, to make them interoperate. And that's actually also our point of view, of course, as a product that we have as deity, that we want to have that independent integrations that can work with each other over time. Because that's I think one of the key elements is flexibility from a more, uh, longer time period, because, of course, you know, we choose particular services right now, but maybe three or five years down the road, our requirements might change or better competitor for a particular service pops up. We want to make sure we want to use a new search engine, for example, or CMS or whatever. And if you basically build all these integrations directly from one system to another without a normalized composing layer in between, then you have that risk that basically when you want to introduce or replace a certain service, that you have to basically rebuild your entire integration and especially also rebuild the dependency in the front end. And that's where we come in, where we abstract away that logic on the front end side. So every data that the front end receives is completely normalized. That's why we always opt to integrate via our composer instead of doing direct system to system integrations. It's mainly for making sure that it's maintainable and that it's future proof. I mean, it's the essence of composable commerce, having a way to compose the data together and use that commerce composer to bring all that services together. And talking about that getting together, I mean, Adam, we're getting towards the end massive project years in the making. I think about eight months of developing eventually. What now? I mean, come on, what happened? How did you experience that? Yeah, it certainly was like the big crescendo, to be honest. Uh, you know, it was it was certainly my almost my entire career. If Jimmy brings has been working on a new approach to, to get something to market, you know, my predecessor, Elliott was also working on, you know, he came to Jimmy brings to release the new version of the app. Uh, so it was even going even longer than than my tenure. Yeah. Preparing leading up to it was kind of one of those moments where, you know, you are double checking everything. A lot, of lot of big days, late nights, obviously dual shifts coming into the international time zones and really just ticking all the boxes because I think when when we look at composable architecture, we, you know, everything expands like the complexity layer is there. And using that composer layer to really orchestrate the data contract relationships is a huge plus. Like if we can confirm, okay, we're talking wine here, then our clients don't have to worry. And so when we actually got to go live, it was a real funny experience. The first order we actually placed on the platform was sitting there and the delivery driver shows up to my house. The delivery driver has absolutely no idea that it's the first order on the new platform, and he's just here delivering the order to the front door. And I was like, come in, come in, come in. And he come into the house. I got the big screen set up and there's, you know, everybody's on the call and everyone's like, yay! Cheering and this guy has absolutely no idea what we're talking about. Just here's your beer. And then yeah it was a great moment to be able to. Did you even know who you were? No. He had no idea that I was. Yeah, I told him, because we tell our drivers that don't enter the customers houses, you know, for safety. And so, yeah, he came in, I showed him my little Jimmy brings ID and things like that, and he's like, oh, okay, great. Great. Comes in and gave a wave to everybody. But it was a fantastic moment to be honest. And something that I'll always remember. You know, my professional career is, you know, always that first transaction. And it's also the first transaction in the, uh, E-comm platform as well. So always be there. Yeah, it was a really great milestone. Yeah. Marvin I mean, the same here, obviously, you know, the launch on the other side of the world, we couldn't have beer delivered, but we did have some drinks. Tell us how you experienced that launch. Yeah. I'm still waiting on the drone to arrive from Adam that he sent me all the way from Sydney, but I'll. It's charging along the way. Yeah, yeah. Okay, okay. It's fine, it's fine. Yeah. No. Yeah, definitely. It was a it was a really great launch. I would say, uh, quite smooth actually, which, uh, which is of course really good. So yeah, as Adam already mentioned, you know, especially two, three weeks before the launch. It was really all hands on deck and making sure that everything works as intended. Of course, as with every launch, uh, is the case, but I think that worked out pretty well. One of the things especially, uh, given that time zone difference, of course, one of the things which is actually good is that the peak trade of Jimmy Binks is in some sort of a good time zone for us because basically, uh, when our day starts is when your peak trade starts. So we don't have peak trade in the middle of our night. But basically, you know, it's it's a regular thing. So, uh, we're on call doing shifts, making sure that twenty four seven, at least someone is available. And I think that worked out quite well. I think, uh, I got one call from Adam on the first night, but we resolved the matter within, uh, quite quickly. So I would say, uh, quite a nice launch and, uh. Yeah, absolutely. And I think, you know, touching on that, it's it's not something that we couldn't have done a soft launch or, you know, just migrated our web customers over because of the way that all the moving parts work. Obviously, as a business, we have loyalty, we have different discounts. And so, sure, we probably could have in a longer term, like if we expanded the project out for another year, maybe another sixteen months, we could have integrated our old E-comm platform into the new layers. But I think the delay in doing that would have still put us back into rebuild mode and not building features for our customers. So transferring everything over in one night, you know, four hundred thousand plus customers in one night and then trade the next day and have a really solid day of trading and not to roll back and not even contemplate it. It's a great experience. Exactly. And of course, then a beautiful cake arrives in the office from deity. It was it was amazing. So thank you very much. You're very welcome. So you talk about having a new architecture, you know, launching, not having to roll back. I mean, it sounds like a dream. And I've done some project launches and not all of them go that smooth. I mean, I'm sure that we really had to check all these moving parts and make sure that they work together, and that is how you can accomplish that. But more important is after the launch, you say, hey, before we had to patch the whole thing Constantly and now we can build features for our customers. So how do you experience this new architecture, this new platform, after the launch, did it improve your processes? Yeah. You know, for me, this has probably been the quietest summer that I've had in terms of tech. So that's a fantastic thing. You know, I get my weekends back. Obviously weekends are great. Trade in Australia for alcohol delivery services. And unusually like the last couple of years, Jimmy Briggs, we've had to have meetings about what do we need to buy more servers for the platform, who's going to be on call to scale in the servers, put them into the load balancer. And you know, all these things that cloud and you take for granted when you have them. And so when Christmas rolled around this year, it was very much like, oh, everything's nice and smooth. You know, it's it's working. Everything that we planned is working the way it should. So that was that was a great moment for us. And so it kind of puts us back into feature development mode now. So longer term for Jimmy brings there's some exciting things coming. We've got some big features coming out, some of our suppliers in the next couple of months, and we're really looking forward to unlocking all of the functionality that we have in all of our new products. Okay, so now I'm very curious, what are the new features? What are the new functionalities? Tell us what is next. We've got a signed NDA in place. Uh. I think the key for us is really what is you know, we want our customers. We've given them a new facelift, right? The apps got this beautiful new UI, you know, really easy to interact with. And we get a lot of good feedback from it. But then how do we take that to the next level? How do we help our customers in the best selection and get that best experience for them? So personalization is a big thing that we're going to start using. You know, all the products that we have all have personalization features built into them. So everything from better discounts, better loyalty programs, you know, all of those things will start to come into play for our customers in twenty twenty three. Amazing. Sounds like I need to get back to Sydney to play some waters there. Marvin, you know, more general question from a head of product perspective. Is this how you envision composable commerce? Is this how it should work? Yeah, definitely. I think Jimmy brings is a perfect case of how composable commerce should be done and when it makes sense, actually, to embrace this composable commerce. Uh, architecture, as we've discussed during this podcast, there's a lot of moving parts, there's a lot of different systems together. And the key thing is here again, they have to interoperate. And I think this is a really beautiful case. I think in total we were using sixteen different services, both, uh, web based SaaS systems, internal systems from Jimmy Brings group. So I think it's really a beautiful case. Uh, we even have a web app. We have native apps again, uh, a lot of moving parts that are working together. So I think it's a really great example. And if you look at it from a product perspective, from, let's say, data perspective, I think it's really great. This is really where our product fits the case and where it makes sense to use our product. So I think it's really, uh, really valuable case here. And I think from a project perspective in a composable commerce way. I think you did a really good job because I know I mean, of course there was a preparation and everything running, but the eventual coding took about eight months. And I think that is amazing. If you compare that to building a full platform. You know, when, when, when I see projects going over two or three years and still not being completed. And we certainly seen some curve balls. So it's okay. It wasn't just a straight, easy build. Look, anytime you replatform something that hasn't been touched in a number of years, there's always a hole here and there, but, uh. Yeah. And look, I think the composable architecture has already paid off. You know, we've we've already made changes to some of the internal microservices and not had to push a release out to any of the front end components or change anything in the PWA code base. So it's fantastic to be able to do that. And it just I think it builds confidence. If you're only changing something small in a microservice and it's unit testable and you're covered by your integration test and your Pipelines. Then the confidence grows and it reduces that stigma. And I think one of the other things to remember during the launch, we had to change something minor. And it was a hotfix went out like during peak trade. And it just worked, you know, like we'd never have been able to do that as a company. And, you know, deity, you know, says what it does on the box. You know, it solves these problems that we've had in the past and certainly is great to work on. Yeah, I remember you asking, so when do we plan the deployment? Yeah, yeah. And then let's just do it right now. It's a rolling upgrade. Yeah yeah, yeah, it's a big change. Um, and yeah, it was definitely like a another moment of happiness. Well, it sounds like a dream with these things. And of course, we're talking about the good things, but I know there were some struggles there as well and some swearing going on. And I think the Australian swearing is very much like the Dutch. So that was pretty, pretty funny to see. We've got some other good ones, but I won't say another podcast and I I've heard actually some of the Dutch guys here using the Australian words now, so I've been hearing them all around. So hey, we're grabbing a beer salvo, right? Yeah. That's good. Mate. I know this one. Hey, Adam, what would you recommend to anyone who wants to start using composable commerce? Yeah, it's something that's probably come up. Is a lot of questions from people looking this as a case study is where do you start? And I think if I had to have my time again, I think it's selecting that partner up front that can help you make your platform selections. Um, you know, we're very lucky to have a partner on board that we've recommended really good products, and we definitely dove into them and did our due diligence before just blindly selecting. But, you know, the recommendations were great. We kind of did things a little bit differently for our platform selection. We kind of looked for the E-com piece first and then went and chose a partner around that. But I think next time, if I have to do this again, I'd probably look at a partner first that can offer a solution. After doing a bit of a deep dive into the company and coming up with that. But, um, overall it's it's a great start and I think, you know, sit down and map your business processes. It's a very time consuming thing to do if you if you haven't got it. And many startups, you know, it's just build, build, build, get things to market and don't worry about documentation and workflow processes. And taking that time to do that will help you understand where the gaps in the platforms you're looking at exist. So you know, how closely does your loyalty platform align with, like for instance, talent. Once you know, if you tweak your process a little bit, then you don't have to try and rebuild that as a microservice. Or you can just use the platform that's existing. Um, and I think that that's probably another big call there is just to really sit down and understand your business and then look at the fundamentals of the platforms that you're selecting as well. I think that is the best advice. You know, get to the drawing board first And you don't have to do everything yourself. I mean, call in some help. We can even help with that and sit down and see what's going on. Get that broad overview and then start building. Okay, guys, I really, really enjoyed talking with you. Adam. Thank you for being here with us for sharing this great story. And Marvin, for your insights in how it is from a product perspective and how we work together for this project. We're not done here yet. We're going to keep on building and we're going to make this brand even bigger. And I hope one day I'm gonna see Jimmy brings vans driving around here in the Netherlands. Maybe. Maybe not. Who knows? But it would be amazing to see them here. Uh, Adam Marvin, again, thank you. And to all of our listeners, thank you very much for listening to this podcast. I hope you enjoyed. I hope you learned something. And if you're in Sydney, order a beer. Goodbye. Thanks, guys. See you. Thanks, everyone. Bye.