Trail Appliances: Bridging showroom and screen: AI, data quality, and the senior-heavy team

[00:00:00] Josh Johnston: What's made us successful is precisely the lack of that. We're just gonna keep doing things the way we're doing them. That's where that curiosity power has to be really strong in terms of the people that we're hiring.

[00:00:13] Kailin Noivo: Welcome to another episode of The Ecommerce Toolbox Experts perspective. Joining us today, we have Josh Johnston from the West Coast of Canada, Senior Director of Online Experiences for Trail Appliances.

[00:00:23] Kailin Noivo: Welcome, Josh.

[00:00:24] Josh Johnston: Hey, Kailin. Thank you for having me. Excited to be joining you today.

[00:00:27] Kailin Noivo: Awesome. So, yeah, we wanna talk a bit about how retailers and ecom leaders are thinking about digital experience. You and I were just chatting a lot about how AI is gonna impact this stuff, but let's maybe start off by telling us a bit about your career journey and how you ended up in your current role.

[00:00:42] Josh Johnston: Well, I'll try to do it as quickly as I can here. It's been a roller coaster. Back in my mid 20s, actually, the plan was to be a historian. That didn't quite work out, mainly because I realized history was a hobby, not necessarily a career. But while I was doing my schooling for that, I actually worked on the fulfillment side for London Drugs when I was there, literally going out, packaging boxes up and handing them off to Canada Post, and that was really my first foray into ecommerce. And I thought it gave me a good perspective of what the ground floor looks like, how logistics work and all that. In 2012, moved into the ecom manager role at London Drugs. I got my first taste of replatforming projects, and that's been something that's been fairly consistent throughout my career: coming in, building the product roadmap out for an organization, assisting with the major technological transformations on the ecommerce side. And I've worked at Best Buy Canada, more recently at Article Furniture, and have been at Trail Appliances now for 4 years. It's been great.

[00:01:47] Kailin Noivo: That's awesome. Well, there's a lot of topics for us to cover today. I know you and I initially kind of got intersected when you were coming over and trying to figure out what the digital strategy was for the company you're with now. But yeah, very, very curious, obviously, Trail Appliances, those that live on the West Coast of Canada, in the West Coast understand this. You guys have showrooms. You have product experts. You have consultants. Talk to us a bit about your online experience. How do you kind of fuse that together with what some of your clients are expecting offline, for lack of better words?

[00:02:22] Josh Johnston: Yeah. And that's ultimately the challenge we need to solve for in the appliance space. This is a longer customer journey. And if you've been in any of our showrooms, particularly in the BC market, you'll know that it's a very elevated experience. So, for us, we wanna make sure that shopping is easy and as compelling as possible. And, you know, I think, as a regional appliance retailer, the first step to doing that is just making sure that we're getting rid of as much friction as possible in the user journey. And also understanding that the website's role in our overall business strategy, is making it easy for customers to really execute that research phase knowing that you know, I think the latest study I saw in the United States, it's been more of a mature ecom market there is, like, at the height we're gonna see, like, 20% or 25% of appliance sales going directly through ecommerce versus stores. So, stores are gonna still play a massive role, but digital plays an incredibly important role in terms of ensuring the customers have all the information they need. We're generating strong leads for our stores, so that it's a seamless baton pass once the customer goes to the store. So, what that's looked like for us is a lot of focus on making sure that those core pillars of a strong digital experience are in place. And as you know, it's never-ending. Once you kinda get to best practice, then you're looking at your data and ensuring you're optimizing based on the feedback you're getting from customers, both quantitatively and qualitatively. So, we've taken things to a certain point, I'd say. And as I was saying off the call, we're literally gonna be launching a brand new composable website this Sunday night. So, that'll be another evolution of our digital experience and, again, will be a step upgrade in terms of aligning that in-store experience with our digital experience.

[00:04:25] Kailin Noivo: Yeah. No. That's great. Obviously, you have a ton of experience. You worked at another one of our customers, London Drugs, Article, a lot of West Coast great companies. Obviously, you've managed different teams as well. When you're starting to look at the funnel, does it really depend the industry the person came from, because some industries might be more impulse purchase, shorter funnel, some are longer, or more sessions is good, conversion rate's different. So, yeah, maybe talk to us a bit about your industry-specific. Are you seeing certain teams, like, over-invest in certain areas when it comes to funnel analysis and underinvest in others? Curious about your take there.

[00:05:01] Josh Johnston: Yeah. It's a bit of a longer funnel, and I would say you get more signals. So, you know, the user journey, let's say, at a lot of drugs, particularly for, like, Tide Pods, bad example, but for the sake of an example, I mean, that's more, get as much friction out of the way as possible. Make sure that you have inventory, and make sure that the price point is compelling enough for the customer to pull the trigger. With appliances, you do have some customers that, you know, their dishwasher broke, and they need something ASAP. So, they're similar to that Tide Pods journey, but they're still gonna be checking multiple other retailer websites. And you've gotta be able to bring to the table a competitive advantage around the experiential side of that, like, producing cognitive load in particular is where I'm going with that comment because I don't know if you guys have shopped for appliances, but even some of the really good websites, like, I love what the AJ Madison team is doing. There's a lot of information being thrown at the user. So, if we can make that a bit more of a seamless experience for the customer, that could give us an edge, along with those core vitals that are, you know, part of a great ecom experience. But to to level that back to your where you go with the question, I'd say you need a lot of curiosity and a lot of high agency out of, you know, the team to be able to problem solve where there could be gaps in the user journey. We have so many different user journeys in the space. You've got that replacement customer, then you've got the renovating customer. They're on, like, a potential 2 to 3 months timeline before they're gonna pull the trigger. You've got customers who maybe have already ordered one appliance, but they're looking for another appliance and making sure that we're providing what they need there. So, compared to other places that I've worked at, though, I would say, like, you need to get your hands on all the data that you have access to. You're dealing with a lot of signals versus a lot of, like, okay, that, yeah, everyone's gonna tell you probably that the funnel isn't linear, but I'd say in this space, it's even more kind of all over the place a little bit. You need to understand what signals are important and which ones aren't. And in order to do that effectively, you'd better be looking at a broad dataset. So, for us, it's like live chat logs. You're literally talking to the sales staff at the store level. What are they seeing? What are they hearing? It can deviate based on the regional market. That can be dependent on, like, the housing market, those sectors. And then aligning that with what we're seeing, Google Analytics, what we're seeing in Noibu, what we're seeing, you know, for me in particular, I love session camp footage just being able to align. Okay. This is what I'm seeing in the chat logs. This is what my final data is telling me. You know, my Looker dashboard. I have some hypotheses of what's happening here, and then I can align that with the session cam footage. And it's like, okay. Now we've got some ideas that we can build a feature around or a test that we may wanna run that, you know, continue to optimize from there.

[00:08:00] Kailin Noivo: You know, that makes a lot of sense. And for you guys, I'm curious, like, what most surprised you when you looked at the journeys? Where were people actually getting most stuck, and how did you find that insight?

[00:08:11] Josh Johnston: One of the big things in our space is the way the appliance industry does promotions. So, a lot of these offers are buy more, save more. Brands are very picky around what products qualify, which products don't qualify. And in terms of the type of offers, they could be like, you know, free haul away on this one, but not on this model, same brand. So, being really explicit in communication is essential in terms of really capturing conversion for that replacement customer and being super explicit around what the offers are. I think it was something that took me a couple of months to really appreciate. And we were seeing that in our chat logs, customers going in on confused, like, is this on sales and not? Or why is this one on sale and this one's not on sale? And, you know, for us on the ecom side, particularly our team, we're a bit biased, I'd say, to minimalist design, you know, particularly coming out of the furniture space. That's been the direction we've been going. There was a bit of a wake-up call there quite quickly that we were seeing in our data where it was like, no. For certain situations in the appliance conversion path, we actually need to be really in the customer space. They're really explicit around what the offers are. Each step of that, like add to cart, cart fly-out, and then to the cart page, around what the offer is. If there's an additional step they have to take, like what they have to do, making that super simple for the customer.

[00:09:40] Midroll: If you're listening to The Ecommerce Toolbox, you're entitled to a podcast exclusive website audit. Go to noibu.com/podcast-audit for a free scan that uncovers the hidden friction blocking your conversions and shows you where you're leaking revenue.

[00:09:54] Kailin Noivo: Not to give away too much, but do you ever envision, like, taking some of those landing pages or the high consideration purchase kind of behavior and actually dynamically changing what the customer sees next, like segmenting the user kind of early on in their journey? Is that something that you guys have thought of? Where do you think the technology is for something like that?

[00:10:14] Josh Johnston: It's definitely the direction I think we're going. You know, I, not to get too futuristic, although it's kind of here now, I'm probably gonna butcher the name of the fashion brand, but Brunelli Cucino (Brunello Cucinelli), I can't remember. But they have that AI front-end experience, and I do think that's going to be a version of that, anyway, will be a core component of the ecommerce shopping journey, probably in the next three years. For us, we're just getting the infrastructure in place to be able to do that. But we do have these core user journeys. Like, there are a couple of other ones that I haven't mentioned, and we do have a subset of interior designer shopping sites. So, what should a landing page look like for them relative to our retail customer? And then we've got B2B customers. What does that experience look like? I mean, we've got landing pages for them, and we're looking to optimize leads to send to the teams internally to be able to action those. But I think we're on the precipice of some really robust functionality in that regard, both in terms of making this easier to manage and maintain and set up and also for those experiences to be very generative based on a bunch of data signals that we could send to these AI tools and platforms.

[00:11:30] Kailin Noivo: It's very cool. So, at that point, like, what becomes the constraint? Do you know what I mean? Like, is it the ability to sift through the insights? Is it capturing the insights? Like, what becomes the biggest constraint in that vision?

[00:11:42] Josh Johnston: Data. Data quality. Particularly in our space, like, our product data is insane. We have a partnership with MerchKit out of Vancouver here to help us out with that, and then we use KINYO as our PDP; they're a great partner as well. But you got that side, then you got all your customer data, and then you got all your product, your behavioral data, then you've got your work qualitative datasets, like your component library, the CMS side, you know, all of your search data, anything that you could, you know, your brand style guide. Sorry. We're just trying to go there. Like, all this stuff that's probably more of a lengthy prompt that you can then expose to these AI tools to provide guardrails around the type of experiences they can construct, the type of components they could pull in. But at the end of the day, it's the fuel for these tools to create these custom, somewhat personalized experiences is the quality of your data.

[00:12:38] Kailin Noivo: Yeah. That makes sense. We think data capture was a bigger problem than it is now. Most people actually have the data now, but your point, like, they're capturing it from multiple different sources. So, kind of like transforming it, synthesizing it, storing it in a readily, like, queryable way is very problematic. Because at the end of the day, you also don't wanna spend $8,000 of tokens, right, to make an extra $600.

[00:13:05] Josh Johnston: So, I think there's a cultural dynamic to it, too. We did our first foray into AI with the AI tool widget on the BC site. That's on our PDP pages, and we built in tandem with a partner, an orchestration layer in LangChain that's referencing product data and a few other things. We've got some agents in there as well. In my conversations with some of my peers, I think there's a bit of a challenge and anxiety in some companies to just get going or just trying some of this stuff and beginning to build up that cultural competency internally within existing teams of how this stuff works and what to do, what not to do. I think there's this dual FOMO happening right now, but then also a lot of fear and anxiety around like governance and security and how to manage that. And so you get this kind of paralysis that I think is starting to go away, but definitely last year, I was seeing a lot of that. And I think as leaders in the ecom space, we have the opportunity to just grab the flag and run with it, you know, get a couple of case studies out the door, showcase the value that these tools can provide, and then, you know, leveraging that, start to build out the wider strategy for the organization with other internal stakeholders and partners.

[00:14:22] Kailin Noivo: Makes sense. Where have you been most surprised in a positive way and then most surprised in a negative way when it comes to the integration of these tools?

[00:14:28] Josh Johnston: I would say on the content side, just the amount of orchestration up until recently, you needed to really produce good content was annoying. But that probably says more about me than it does, you know, the hardcore specialists out there that are able to get what they need out of the video and imaging tools. But there's definitely been leaps and bounds done, I'd say, in the last two or three months, where, you know, even for myself, I'm able to produce some decent imagery where I couldn't before. Where I've been really happy with the tools is data analysis and market research. Particularly, Gemini for market research has been something I utilized quite a bit. Copilot's been super good. We're a Microsoft shop internally, just helping prep for meetings and stuff like that. It's got some limitations, but it's great for just being able to go through your emails and go through meeting notes. Just make sure that, you know, that administrative stuff, you can save some time there. But, yeah, for me, I like to, particularly with Gemini, just, okay, what are we seeing in Nanaimo, for example? What's going on with that housing market? You know, and it's fairly up to date. So, it'll go up to, like, you know, a new story happened two days ago. It'll include that in its analysis, and then I can align that with all the internal data I have. Okay, I'm starting to see a trend here. But that data analysis support's been super helpful.

[00:15:48] Kailin Noivo: It's been, like, pretty well-documented, maybe less for your industry, that, like, the weather has a pretty big impact on conversion for online stores. Do you ever think we're gonna get to a point where it's getting that specific? Like, Nanaimo's, like, raining today, so pull back on your ad spend, or is that just a bit too, like, Terminator?

[00:16:10] Josh Johnston: I think the challenge for us in the appliance space is just the lead time to transaction. Right? So, given that for some customers, it might be a month before they're willing to pull the trigger, it may not be the best decision to pull ad spend in that moment. But I think what that type of analysis is gonna help us with is when we're looking back, why did we see the results that we saw and was weather a factor in that? We'll have more confidence in saying yes because of the analysis these tools can do, or not, if there was a bunch of impact.

[00:16:42] Kailin Noivo: Makes sense. How are you thinking about how this intersects with your headcount planning moving forward, like, the distribution of your team? Like, yeah, how do all these things kind of intersect?

[00:16:53] Josh Johnston: Yeah. I think for the way I view it, it's an opportunity to be a productivity multiplier for the existing team. So, it's looking at the great people that we have on staff at Trail and, like, how can we make life easier for them and how can we, you know, 2x, 3x their productivity. And, of course, everybody wants that. So, I think for us, it's looking and then staying on top of what's possible with these tools. For some roles, you know, the impact is gonna be a bit minor. For others, it could be quite substantial. But we wanna make sure that we're doing that analysis first before we go to HR with a headcount request, just to ensure that we've explored that option. And I think we've seen that with some companies. I think Shopify had their memo where Toby was really pushing that. Makes a ton of sense. Is AI gonna, you know, there's still things they can't do well, right? And it's tough in those situations, but I know those team members wish it was able to support them more. And then for others, it's more if it can, just making sure they're trained, in order to use those tools effectively. And that's a tough one. Right? Because it's evolving so quickly. You know, what was cutting edge 6 months ago isn't necessarily the case now. So, you're always, that's where that curiosity power has to be really strong in terms of the people that we're hiring. Right? Like, they have to have that, like, motivation to wanna learn stuff at that comfort level with not always knowing or not feeling like you're always up to speed with the latest and greatest. You have to be consistently learning.

[00:18:26] Kailin Noivo: Very cool. As we look to wrap up, a question for you is, as you're recruiting for your team, and I'll speak super candidly, like, we've changed our engineering persona, whereas, like, it's kind of, and I hate to call it engineering, but it's pretty black and white. Some engineers are like, these AI tools are stupid. They're wrong, blah blah blah blah. Some potentially lean in too much. But, like, have you changed your persona for hiring around this as well? Like, we saw in our business probably 18 months ago, like, kind of the flock started to split where it's like some people wanted to, like, hang on to how things were done 10 years ago or in some cases, actually, like, 3 years ago. But, yeah, I'm curious. Like, has this influenced, like, how you look at hiring?

[00:19:15] Josh Johnston: So, on our dev side, we've got a really good core team, so we haven't done any hiring recently. But what's made us successful is precisely the lack of that we're just gonna keep doing things the way we're doing them. What I'm seeing in the broader market right now, and this is how we're structured, we see a smaller team of senior talent that's leveraging these tools versus a large team of junior to mid-level, that maybe isn't as well equipped to leverage these tools. And, you know, I think Lulu, Aritzia, others, like, hearing similar vibes there. I'm not sure it's the same for you guys. But for us, it's, you know, you need somebody who understands development enough to get the most out of the AI tool, somebody that doesn't have that experience or skill set, and again, they have to have the right attitude, too. Like, you can have a senior dev that wants to continue to do things the way they've done things for 10 years, that's not success either. If they have that curiosity and they wanna stay on top of things and they've got that experience, like that's super valuable because now you're like, you're getting 2x, 3x, 4x productivity from a senior dev. Like, think from a salary perspective, what the cost would be to try to get that at a junior mid dev level, you know, the math becomes very favorable. And that's not to say, like, we're against junior devs and all that, because the junior devs that are coming out now, right, like, they're native to a lot of these tools and experiences. They just are our products; they just need the experience, right? So, in certain situations, that can absolutely make sense. For us right now, though, we're built around a core team of senior devs.

[00:20:58] Kailin Noivo: That's a really interesting insight, and I also think, like, it's dependent. And what I mean by that is, like, ideally, you have a senior person that's bought in, but in some other and I'm not talking dev now, just in general. A junior person fully bought in is definitely better than an intermediate or senior person not bought in in certain roles, right? Like, for example, cold calling team, like, yeah, guys, I get it. 2019, the glory days, you just sat and cranked dials, but, like, now you have to use an auto dialer, which, like, only connects you once someone's answered you. You know what I mean? Like, you have to kind of adapt, or you end up in this strange scenario where, like, you have someone who's just trying to, like, pick all the vegetables instead of using the tractor. You know what I mean? And, like, maybe, like, certain vegetables you need to handpick forever, but, like, you should probably use the tractor.

[00:21:48] Kailin Noivo: Josh, this is a great combo. Really, really appreciate you taking the time, and, yeah, it was nice catching up.

[00:21:53] Josh Johnston: Yeah. Appreciate you having me on, and let's do it again sometime.

[00:21:58] Kailin Noivo: Absolutely. Thanks, Josh.

[00:21:59] Outro: The Ecommerce Toolbox Expert's Perspectives is brought to you by Noibu. To find out more about Noibu and how we can help you debug your ecommerce site and rocket your revenue, visit www.noibu.com. That's n-o-i-b-u.com. And then make sure to search for The Ecommerce Toolbox Expert's Perspectives on Apple Podcasts, Spotify, or anywhere else podcasts are found, and click subscribe so you don't miss out on any future episodes. On behalf of the team here at Noibu, thanks for listening.