Confessions Of Supply Chain Executives | What Has To Be True For E-Groceries To Be Profitable?

For years, grocery executives have been asking the same question in boardrooms across America and really across the world. When will online grocery actually make us money? The answer has been uncomfortable to date. It hasn't. It doesn't. At least not yet. Most grocers lose money on every single e-commerce order they fulfill. They've been subsidizing convenience with razor thin margins, hoping that scale will eventually solve the problem. But here's what changed last month. U.S. e-commerce grocery penetration hit 19%. That is, according to Brick Meets Click, 19%. That's not a rounding error anymore. That's nearly one in five grocery purchases happening online. And if you can't figure out how to make it profitable at 19%, what happens when it hits 25% or even 30%? Today, we're going to answer not the $64,000 question, but the billion-dollar question facing the grocery industry. What needs to be true for grocers to actually make money on e-commerce? Welcome to Confessions of a Supply Chain Executive, the podcast where we get brutally honest about the challenges, failures, and also where we can celebrate the victories in retail supply chains. I am your host, Chris Walton. Today's episode tackles the most pressing question in grocery retail, profitability in e-commerce. We're not talking about growth metrics or customer acquisition. We're talking about actual unit economics, making money on orders, not just fulfilling them. My guest today is Richard McKenzie, the CEO of Valak, a division of the Rolik Group. For those unfamiliar, Rolik operates one of Europe's most successful online grocery models, and Valak is tasked with taking the lessons learned from their operations and to help other grocers achieve e-grocery profitability. Richard brings a unique perspective from seeing what works in mature e-grocery markets and understanding what American grocers need to check the profitability code. Richard McKenzie, welcome to Confessions of Supply Chain Executives. Thank you, Chris. Good morning to you. Yes, I'm excited to have you. I mean, I think you bring a very unique perspective to this conversation. Why don't we get started too? Why don't you explain your background and how you came to be a quote-unquote grocery e-commerce expert? I'd say I started this journey, I spent a decade in China. And in China, just everything went online. and the grocers who were there just lost literally 120% of their profit to the likes of Alibaba, JD.com and so on. And I just kind of started to look at it and say, okay, the West will be different, but the same thing's going to happen. And that's why I got into, that's why I got interested in this. And then I joined Ocado. I was their chief commercial officer for four years, looking at their solutions business, looking after their partnerships. And then that led me on to working with Rolick, where I kind of thought, well, actually, Rolick's got something pretty special from a customer point of view. And what's the technology that underlies it? So here I am. What years were those that you were in China? So I was in China 2010, 2019. So it went from zero e-commerce to, you know, I left Shanghai, I reckon there was probably 40% of grocery online. It was just total transformation. Right, right. And so, gosh, yes, so 15, 16 years ago, now, you started to see this, You had this epiphany that this was coming. And that 40% number is interesting, too, because I want to start with the elephant in the room today for this discussion. U.S. e-commerce grocery penetration, as I said at the outset of this podcast, it hit 19% last month. And it's approaching what previously was the magic 20% threshold where e-commerce starts to, for lack of a better way to put it, generally lay waste to retailers. So from your perspective, working with grocers on both sides of the Atlantic, is this 19% milestone or is this 19% penetration, I should say, is it a milestone tipping point that should start to worry traditional grocers? Is it a validation that e-commerce grocery is essentially mainstream at this point? For me, e-commerce has been this kind of elephant in the room that grocers have wanted to ignore. And I'm sure we're going to talk a lot about it. It's less profitable. it's difficult, they don't know how to manage it, and they try to ignore it. We're just getting to this point now where you can't ignore it. You know, 19% is a lot. And if you're not doing it, that 19% is going to somebody else. And I think it really, you know, a lot of the managers said, okay, it was a COVID blip, we want this to go away. But actually, now it's just keeping on growing. And you've got to take it seriously, whether you like it or not. So it's become this problem that I think can't be brushed under the carpet anymore. Yeah. Yeah. It's interesting. I went on, just funny anecdote. I went on Amazon this week and I tried to order an air filter for my furnace. And as I'm ordering the air filter, it popped up like literally 12 to 15 groceries in a tile format that it just wanted me to add to my order. So it's definitely becoming more part of the consumer experience every day too, as those entrants come into play. And the consumer expectations are going up and up with this. Amazon is driving that. This need to deliver now, and this need to deliver in three hours, it's just getting more and more. And I think that's what's driving this penetration. Yeah. And the survey that I referenced at the outset, the 19% from BrickMeetsClick, they said that the Amazon push into quick delivery of just the grocery items you need or maybe forgot is definitely having an impact on that number. All right. So let's talk bluntly then about the unit economics. When a typical U.S. grocer fulfills an online order, are they making money, losing money, breaking even? And what's the typical loss or margin on an order in the U.S.? The brutal truth is most people are losing money. And the reason they're not investing in this, they've got themselves into this vicious cycle, is they're losing money and therefore they don't invest in the proposition. Therefore, they don't get growth and therefore they lose money. And I think that's kind of that's where most grocers are today. How much? You know, I think the good ones are kind of operating close towards break even. But I think honestly, you're seeing kind of minus five, minus ten dollars in order in places. It's pretty brutal. Yeah, that's not good for anybody. You know, and everyone also has a different theory about why that is. So break it down for us because you've got experience in China, you've experienced Okada, you now have experience at Valak and Roller Group. Where does the money actually disappear in online grocery? What are the biggest cost drivers that kill profitability or lead to that outcome that you just mentioned? There's two obvious ones which everybody talks about, which is the cost of picking and packing the order. And that's just typically that's not being done terribly efficiently. And then there's the cost of the last mile, which typically kind of starts to fall apart. If you're doing it out of a store and you've only got a handful of orders, you can't optimize that route very well. And I think those are clearly the big ones. I think the one people tend to forget, though, is the basket size. If you suddenly, if people aren't doing a full basket shop with you, you know, if you're not getting $100 and you get, what, $30 a margin, suddenly you'll never pay for that pick and pack and deliver. So I think it's, for me, picking, packing, delivering is really important. But I do think we tend to forget about actually you just need to also make sure you're getting as big a basket size as possible in that. How do you counter that? I mean, e-commerce doesn't seem built for that in a lot of ways. You know, and this is, I think, where Rolick has done a really good job because Rolick has said we will deliver quickly, but we will not compromise on range. So, you know, we're doing typically 25,000 orders or 25,000 SKUs rather that we're offering in our markets or there or thereabouts. And it's the widest. But it's not just, you know, your fourth type of Tabasco sauce. We're making sure there's fresh bakery. we're making sure that you can do you know you can want three apples picked for you and you've got that choice so what i say is we offer the full complexity of grocery to make sure that people can get everything they want on their order and therefore they're comfortable to take and make a full basket and i think that for me is kind of is important that makes sense too because like if i'm a regional or a local grocer like i almost have to do that because if i don't do that then it's going to be death by a thousand cuts is like amazon cherry picks you know orders with their air filters like i just talked about exactly and there's no mistake about it amazon is offering a pretty full range now themselves yeah um and you know they're going to be thinking the same thing they're not just going for one banana right right one banana yes well said that might be the quote of quote of today's podcast already all right so so let's let's talk let's talk about let's go back to some of the issues then the picking problem so you know yeah we you know i've had a number of people on the show you know eight years of doing this show you know in-store picking seems very inefficient at the end of the day but then building dedicated micro fulfillment centers is also you know relatively capitally intensive and expensive so so what's the right answer how do you know when you've reached the scale to justify dedicated infrastructure for in-store picking? You know, there's a couple of points. In-store picking is pretty, is very slow. And actually kind of, I think some of the US numbers, the stores are bigger, which makes it harder. But actually some of the numbers you see in the US are really pretty, pretty slow. And I think, you know, that does mean actually it's kind of the tipping point of when you would automate probably comes a little bit earlier. You know, and we're seeing business cases down to kind of 500 orders per day in terms of when it makes sense. What I would also say, though, is not just a case of do I automate this one store now? I think it does make sense to start to aggregate the volumes. Because if you kind of start to say, actually, I'm going to do 1000 orders out of this store and automate it, you're suddenly going to end up with a much more efficient system, and a store that actually knows the processes of how to do this. Part of the problem, part of the reason we're kind of running a you know 50 uph or whatever it is in the pick rates and why they're so low is because the stores just don't have a lot of practice doing it um and i think that problem of lack of lack of volume and lack of aggregation really hits got it so so so said in other ways so you're saying like take a look at the market take a look at where you have the potential order density that you need and then put the infrastructure in you think and you put the infrastructure into the store to help with that is that and but i think it depends okay in in the u.s you've got big you have a lot of big stores you've got a lot of space i think actually there's there's you can create that room in the store to put in put the mfc in mfcs are more difficult because you've got to work you've got to work all of the store processes um actually typically in europe i'd actually say we tend to do it more out of the store but actually i think in some of the in the case of the u.s mfcs are probably more relevant i want to touch on the last mile part too a little bit more too so how does the last mile part come into this because you know what i always hear from gosher's too is like you know i've got the picking problem i can't make my employees do it so i'm going to leverage the third party providers like in the cart and doordash to come in and do it and then you know they'll charge me what they're going to charge me for the last mile so so how do you how does the last mile factor into this equation you know in your mind typically the last mile is the single biggest cost unless and actually it's kind of also the one we tend to ignore because we can build shiny automated fulfillment centers you we've yet to get the robot to drive the car although autonomous vehicles will change these this these economics in the future it's one way that aggregation makes a big deal is a big deal because if you've only if you're running a couple of cars out of a store that you will just have no opportunity to optimize those routes if you're running 20 cars you'll probably optimize 19 of them if you're running 50, you'll probably optimize 49 of them. And actually, we've done some work admittedly in Europe, where we were taking store pick, moving it, centralizing it and creating, in this case, three hubs in a big city rather than 30 that they were doing it from. And we were getting the last mile costs down by about 40%. Not actually because we were doing anything different in terms delivery times purely because we're aggregating the volumes and you can just optimize them better even though on average the stem time the drive time to the first order was a bit longer got it that kind of thing about last mile is just the more you fragment it the more difficult it is to manage got it so the load factor on the delivery cars going out is better is what you're saying exactly if you if you're running 20 you'll have 19 of them full if you're running two you'll have one of them full. Right. Right. That makes, that makes a ton of sense. All right So there been a lot of debate over the years and I feel like I feel like you and I had this discussion too you know in in in passing as we seen each other in person at different conferences and whatnot there been a lot of debate over the years And I feel like I feel you and I had this discussion too you know in in in passing as we seen each other in person at different conferences and whatnot There been a lot of debate about the right infrastructure to run online grocery profitably You've got you got and there's so many different names for these things. Like you got dark stores, you got MFCs, micro fulfillment centers, which you already said. You've got the spoken hub models, too. You've got automated warehouses, which could be a part of any of those we just mentioned. So everyone's got a different answer. You work with Valak on this and a subsidiary of the Rolex Group. What infrastructure model do you believe is the right one that actually drives profitability? And why do you think that is? I hate all the labels we tend to put on things of MFCs, CFCs, whatever you call them. To my mind, the right answer is creating the right size FC for the demand you've got locally. And there will be some cases where you can fulfill to a small city beautifully from one FC. And actually, we've got quite a lot of those. We've got a city of kind of 500,000, 1 million people. We've got one FC. It delivers. It can get anywhere in the city in a couple of hours. And we just have a superb service. I think when you've got volumes that are lower, it still makes sense. And you've got big stores. It still makes sense to be going into doing MFCs inside of stores. And I do think that for retailers looking to use their existing infrastructure, that can make sense. But what I think you've got to do, though, is not just to say we're going to put one in every store and do 200 orders out of every store. because then you're just running a lot of inefficient processes, both in the picking and the last mile. So to my mind, the right message to get out there is it's not a one-size-fits-all. It's building the right size of fulfillment center for the demand you've got in the catchment area that you're trying to serve. That makes sense. Is there a minimum daily order threshold or volume that you need to at least hit to even be thinking about doing anything like you just described? Yeah, I think so it's a build to do something that's really standalone, which is the most efficient. I think you're talking about kind of the order of 3000 orders a day. If you can kind of get to that kind of volume, it makes sense. Or if you think you're going to be in there in the next 12 months, it makes sense. If you want to build something the side of a store, you can probably be making sense at 500 orders per day or above. And I think, you know, you can, now clearly your level of efficiency you're getting down to of those is worse, but you can do it. It takes the difficulty, frankly, of those is the whole management layer of getting the automation to work together with the install processes, because you've got to make sure that you're running the automation you're running you're picking the tail of the product from the store they're both being brought together at the same time and you're not suddenly ending up with bottlenecks or choke points in that process which is where a lot of the mfcs fell apart in the last few years and i think where a lot of people have been burnt wow that's it's so big delta though like you said three three thousand right to five hundred that's the that's yeah i think for stand that's 3,000 is standalone. Yep. 500 will be inside a store. Yeah. Wow. Interesting. And you, but you can still get, you know, 30, 40% of your pick going through automation in those, in those, in those small facilities. And you can still get that. And they can be done efficiently. As I say, the difficult bit is not actually the automation is making sure you can, the store knows how to run the entire, or the guys in the store, make sure you can run the entire thing. Right. That you can operate, you can actually operate it. Exactly. To the efficiency that it's designed to achieve. And that's absolutely what went wrong. Some of the previous ideas on this, I don't think were bad ideas, but operationally, they just fell apart. And the software that was provided with them was just not adequate. Right, right. Yeah, and I want to get into that more, too, throughout this discussion, too, especially as you talked about basket size being a key component of probability as well. I'm curious, though, because the delivery expectation around speed also plays into this too, right? It's not just how many orders am I taking a day. It's also how fast do I want the order as well and where can it come from that gives me the speed that my customer wants from me. How does that factor into this? Yeah, and honestly, I think this is absolutely critical to growing your volumes. and if you look at every i think i think almost anywhere in the world you look at who's winning they are typically the people who are now delivering you know on time in full that's kind of the standard bit but quicker um and i you know you look at where the growth is coming from in the us it is most definitely um it's most definitely same day and getting quicker and actually i'd say if you look at what we've done in rolig you know we've been operating a three-hour cutoff so you You get an order three hours after you press click, you press buy. And we've been doing that three-hour cutoff for 10 years. And that is honestly just from a customer point of view. When customers are used to that, they'll never go back to saying, okay, I'll wait eight hours or I'll wait till next day. That convenience is absolutely essential to driving it. And you've got to build the technology and the software and everything behind it to deliver that. a lot of the software just doesn't do that. Right, right. And for the US listeners, maybe that are unfamiliar with Rolex, I should have done this at the outset too. But just as a reminder, like how big is the Rolex operation? So we're about $2 billion. We operate in five countries in Europe. We operate out of 12 fulfillment centers. So it gives you a rough idea of the scale of the fulfillment centers. we are growing you know year on year at 30 40 percent at the moment um you know in germany our kind of which is our biggest growth market we're growing kind of 40 to 50 percent um and the one thing i the one plug i've got to give rolig the nps numbers we get are outstanding you know we are talking 90 plus npss which i just don't see anywhere else no which which is typical When you have a digital model that works, that provides people that can, especially when it provides, I was just hearing Uber's CEO talk about this on a podcast yesterday. When you have a digital model that works, that provides the convenience at the right price, the general love of that product is very, very high. So that's why you're seeing what you're seeing, right? Customers love it. And it's not a public number, but we spend a lot less on marketing to achieve that growth rate than a lot of other people are spending to get a much lower growth rate. Yeah. Right. All right. So let's bring it back to US then. So the best way I could describe the US market is it's kind of like a hybrid approach right now. It seems like it's got a lot of tentacles. You've got some people placing orders from stores, some maybe doing dedicated facilities. Many, many, many, many, many, as I've already said, are using third-party services. Is this like kind of hybrid approach where we're kind of just diversifying our bets, running a portfolio of options. Is that the right long-term approach or do you think it's just a short-term stopgap? I'm guessing you think it's the latter based on how you've been approaching this conversation so far. This is no longer an optional play. Your 19% number that you opened with just tells you this is something you've got to make as a core part of your strategy. You've got to own your customers. And if you're going to own your customers, you've got to offer them a really good proposition and you've got to be able to deliver it in a way that is the way that is economic. And to my mind, that means you've actually, you've got to start to say, okay, how do I, how, what do I have to do from a technology point of view to be able to deliver that? In some cases, it will be standalone FCs. I think in many cases, it's going to be starting to automate the size of the store. But I think you've really got to, you've got to start owning that as a retailer. And if you're not, let's be honest, there's nothing that stops the third parties building their own fulfillment centers in the future and disaggregating you. And let's be clear, Amazon are the ones who are winning in this market at the moment. So where do you want your customers to go? Yeah, and Amazon has said overtly that they're going to do that. I mean, they've made announcements to that front. Yeah, the way I use your banana analogy, yeah, I mean, it seems like at best it's a stopgap because you're going to end up the second banana if you don't start taking this approach in the long run. All right. Before we shift gears, I do want to ask you, though, because one thing that's been absent from this discussion, I think it's particularly important in discussing the U.S. market, is the real estate angles to this. So, you know, location matters enormously for last mile economics. How should grocers be thinking about the real estate utilization when it comes to designing and deciding where these automated facilities should go? I don't think there's a single right answer to this. It's going to differ hugely depending on the size of the city, the level of demand you've got, and the traffic profile you've got in the city. I think what I would say is aggregating does bring benefits. We talked about that particularly with the last mile, but also in terms of the speed of the picking. So I don't think you want to just be saying we want to do a little bit out of every store I've got. You do want to kind of start saying we're going to aggregate, we're going to run automation at scale, we're going to run last mile at scale. If you've got the volumes and you don't have that space in store, I do believe kind of setting up fulfillment centers that are standalone makes a lot of sense. And it's the model we've operated very successfully in Europe. And then that's quite simple. You start to say, if you take our oldest city, we started with one fulfillment center. As our volumes grew, we just started ringing the city. We now have three. And it just means we can get to customers quicker and quicker and quicker. But I do think don't underestimate the value of actually bringing some of the volumes together so you can actually start to set up something that is efficient. Which brings up another topic. I remember you and I were talking about this a little bit at grocery shop at the end of last year, too. Because part of this, too, is like you've got grocers that are, you know, delivering from their own sites, you know, and controlling that delivery themselves. And then there's you've got grocers that are on the third party marketplaces as well, where then those marketplaces are handling that. you have to be thinking about all that volume in aggregate, right, for this to work. Like you don't want to set up a fulfillment center and then have, you know, all your volume running through, you know, Instacart or DoorDash. I'm not picking on any of them, but just like that's going to factor in and make your utilization a lot less efficient. If you're not thinking about that and designing that as part of your strategy too, right? I agree. And I think if you're a regional grocer in the US, you don't want to be too dependent on the aggregators. or they're third parties. But I do think there's no reason you can't be serving them as well out of these larger facilities. And I think that's kind of the key for me is actually let's just start to set this up at a scale where you're efficient and then actually start to get all of the volumes to that. And if you set up correctly, that's fine. There's no reason you can't do that. Yeah, right. That's the key though. You have to be thinking about that in advance. Otherwise, you're going to be running a distribution. You're going to be trying to do two things to accomplish the same thing with no success. Absolutely. Absolutely. No success. All right. So to that point, so if people are listening and they're kind of buying into what you're throwing down here, which I definitely am so far. So what are the operational metrics that matter most as people start to think about this? Are there KPIs that grocers should be obsessing over to drive the most profitability that they can? Yeah, I don't think this is too complicated. Ultimately, what do you need? You need a unit per hour or a cost per pick or whatever you do. It all multiplies up to the same thing that's efficient. I think the key thing actually, and particularly as you start to automate, is actually there the forgotten cost within that as in the overhead So as you start to automate you tend to forget about the management the shift leads And so you can have a really fast pick right at the point of pick but actually all the stuff that goes on around the edges, you've got to watch. And I think that's kind of, again, one of the things that's fallen apart in the past is people kind of put in some super duper automation, but actually everything else just tends to fall apart and there's a mess around it. So for me, from my point of view, and one of my big learnings in this industry is it's not just about the pick rate of how fast you can get something in a bag. It's about everything else that goes on. And as we get faster and faster robotics, there's more and more of that is the case. Right. You still got to manage it. That's what I hear you saying. You still got to manage it. And frankly, that, you know, what often gets called other in the P&L just is actually becoming the biggest line. And actually, that's what you could, because you can also, and particularly now you've got robotic pit coming into these centers, actually a lot of the actual manual labor is getting really quite efficient. I think the second thing is drop density. And I think that is probably the one thing that probably people forget about most. and I think it is one of the things actually we've put a lot of effort into and have actually made most progress over the last 18 months really trying to understand how to make that work and it works differently in different markets just depending on the labor model you're using and then I think the third thing is basket size which I keep coming back to you've got to make sure you've got a big enough basket that there's enough margin you can afford to provide this service to the customer. And that ultimately, you've got to that means you've got to be offering a wide enough range. And people trust you on the fresh products and the handling of the products and that it's all going to turn up in a with you with your eggs not broken. Is there a basket size number that tells you like, yeah, you've got a good healthy, you know, kind of starting point here to go in this direction? I always I always have a rule of thumb of $100, just because I think you've got $100, you've got you know you've got kind of 30 dollars a margin and then you can do you know if you've got it gives you as a target ten dollars to do your pick ten dollars to do your delivery and you've got some you've got some money left over to pay for everything else but that's important to think about too because like the merchandising strategy then of how you get to the basket is also fundamentally important to make e-commerce grocery profitable too to get the automation to work harder for you in the background so yeah how do you think how do you think about like what are Are there merchandising tips or tricks you've seen that work well to get that basket size, you know, staying consistently at the level you need it to be? I think so. You know, clearly there's some tactical things in terms of types of promotions you drive. You know, you don't want to, you want to drive, you want to drive multi buys because you, you don't want to drive just money off. So there's a kind of, there's some pretty tactical parts around that. There's the good old tactic of when do you do free delivery thresholds or, you know, when do you have, when's the delivery become cheaper to drive that, which is classic. I also think, and this is clearly a bit of a strategic question, but if you have automation, the incremental cost of adding a new SKU is lower. Now, you get no value from adding your fourth tomato ketchup in there. but if you can start to put genuinely exclusive product in and if you think that you know if you're serving an entire city there's things that make sense to range for an entire city that might not make sense to range for a store so you know classically you know expat communities you know I remember a good example you know if you've got a if you've got a French community you they may not may not be worth putting that French product in any one store. But I've seen some great examples where you put that in a fulfillment center and it flies because it's the only place is the only place you can buy that product that you really want. So I think that there are some few things there that you can do within that. And I think, frankly, it's one of the things which offline grocers have not been terribly good at because they think about stores and how to arrange for a store, not about things that think about the work for an entire city. yeah preach it man you are you are speaking to my choir on that one because i have i've long been on my soapbox about how digital is the key to unlocking localization i hear girls all the time talking about we're going to make our assortment more localized i'm like no you're not you can only put so many products on your shelf digitalization is the key to making that happen invest there because then you get all these ancillary benefits that we've just talked about interestingly one of the things we do in europe is we source a lot from farms that are local to fulfillment center um which again makes no sense if you're kind of if you're trying to get those go from the farm to every to every shop you've got in right wherever but because you've got farms that are local to it and then that can just be distributed to all of munich pretty efficiently um we can actually we probably have more local products than does the standard the standard grocery store right yeah no that's a great point for those listening especially those in the u.s that play into that market. I can think of retailers particularly. Substitutions. Substitutions in out-of-stocks, like those always seem like the big things that we hear about is the pain points for e-commerce, grocery. How should they be thinking about managing the trade-offs around those? The beautiful thing about moving into automation is you control almost 100% of your stock. If you've got product on a shelf, you're competing with your customer. even if the system says it's there, the customer might have picked it up and put it in the next aisle. The more you can move into a controlled environment, and I think you do need to have the supply chain segregated so the product's coming in separately because you don't want to go onto a shelf and then also somebody in the back of the store trying to manually split the product. But if you can do that, you know exactly what goes into it, you know exactly what comes out of it, you end up with very, very low out of stock and very low substitutions. The worst you get, frankly, is when a supplier lets you down in terms of the product not arriving. But even then, you know it's happening. You actually know about it. And it's not by surprise. And when you then are doing same day delivery, you're not really worried about did the supplier turn up? Because actually, typically, the supplier did or didn't turn up in the morning. Therefore, the product is or isn't available for delivery in three hours time. So if you can get this right, actually, in our experience, you can drive that problem down to being very, very low. Yeah, and that actually just happened to me where I got a cancellation on an order same day from a store pick. I won't name the retailer, but yeah, it was very, very frustrating. But yeah, and it's not theory. It's proven because warehouses run at greater inventory accuracy than stores do for all the reasons you just described. Yeah, and we have 99.5% availability. It is doable. I don't know. This is not rocket science. This is not because we're geniuses. It is because it is just easier doing it like this when you've got segregated inventory. Right, right, 100%. So that's how you have to – segregated inventory is a key piece of this too, which we kind of danced around probably 10 or 15 minutes ago, but that's good to call out now as a piece of this. All right, so last question on this topic before we move on. And, you know, slot management and customer behavior shaping. Like, I feel like particularly overseas, there's more attempts or more success at potentially doing both of those things. Can you actually influence when customers want their groceries delivered to them too? And if so, does that actually move the profitability needle as well, Richard? Like, is that something that U.S. grocery should be thinking about? Or should they just be like, no, we just got to get it the same day? Like, how do you think about that? I think you end up with quite heated debates on this topic. Okay. You know, at Rolick, we offer basically an unconstrained customer offer. So we are basically making sure that you will almost always be able to find a slot in the next few hours when you place your order. And that for us is an article of faith that we just say we want to be offering our customers that high that high degree of availability because we don't want them ever to go elsewhere. And that's one of the reasons we've grown our business so successfully is because we just offer that level of convenience. You know, you absolutely can start managing slots. and you absolutely you know there's lots of examples of people out there who will price differently to drive people away from away from a Saturday order and to make sure that you are filling up the Wednesday afternoon orders and I think that absolutely can work and you absolutely can do that I'd say it's actually kind of one of the things that I see has gone wrong quite a lot in e-grocery is because what you haven't done is the guys who are doing that don't always talk to the supply chain teams don't always talk to the merchants so you end up with enormous peaks coming at a time when the supply chain team is running the fc or the store pick doesn't know it's coming um and yeah and let's be candid if in a store you get 20 more customers than you expect you have a few queues but it's kind of manageable if you suddenly end up with that you just max out and there's just a, there's a, you can't push the picking any faster. So the system does break unless you manage that. Which goes back to the idea we were just talking about, like the idea here is segregating the inventory, leaving the store inventory for those that are coming into the store to purchase. Correct. Segregating everything out to give you still the flexibility to demand shape if you want. If that's part of your business model, you can, but it'll make it more efficient too. Exactly. And I think this is the whole theme, you know, when you start talk about how you think, how you think about kind of setting up your e-com, you've got to actually make sure that you are being joined up because actually the system is less forgiving when you aren't joined up than it's, I think, than stores are. All right. So let's shift gears a little bit now. So, so there are ways to monetize e-commerce beyond just the selling of groceries too. And specifically, I think about membership models. So I'm going to put you on the spot a little bit with this question. So you've got things like, you know, everyone knows them, but Walmart Plus, Amazon Prime, they essentially are helping, you know, with the profitability question because customers are prepaying for the convenience that the retailers provide them with these membership programs. So let me ask you, should every retailer have a membership program? I kind of think the answer is yes. I'm honestly a big fan of them. And I think there's a, you know, one is just the response, and I've seen this in a few places, including Rolick, the response you get in terms of frequency of shop and loyalty is just huge. The Pareto curve we see of just people who you end up with driving people from being marginal customers to being loyal customers because they've already paid for their deliveries in their mind, and therefore that extra delivery is free. And I think the second point is really interesting, though, because as soon as you start to kind of own a customer much more, you can understand their behavior and guess what in terms of selling advertising that advertising becomes a lot more valuable so you know suddenly you are no longer just a box shifter you've actually become the person who understands the customer behavior and when you're talking to procter and gamble that becomes a really valuable asset you've got yeah well that that's funny as i was going to ask you next i mean like at what scale does does that retail media play then particularly become important when you start talking about e-commerce profitability? I don't know the answer how low you can go. You know, I would say, you know, we, we, Roelig operate in a pretty challenging way, as in we have five different countries for our two billion of sales. And we do a good retail media business in each of those because we understand the customer behavior. Now, clearly there's benefit. There are some benefits of scale, but I think you can, it is not the case. that you can't do this when you're at 300 million. Actually, I think you can. Because I think that ability to understand and really offer advertising to a customer at critical points in a customer's decision-making process, I think is super valuable. How too does private label play into this, Richard? It's part of the merchandising question. It is part of the margin question. I a big fan of private label but then again I was brought up in UK grocery which kind of pioneered a lot of this We no longer see it as the cheap substitute So I a big fan of it I think you require scale to deliver it It takes time to develop it But if you can get it right, it's a big margin driver. And also can then become a loyalty driver as well, because you own those products, nobody else can. Right. Yeah. It kind of plays back in what you're saying, like why you're kind of a big proponent of membership models too, because like you get to know the customer, you can then serve up the items to the customer that you want and then get the retail media dollars, but also you can serve them up the items that are uniquely yours, which is also an important factor on why they choose to shop with you. Absolutely right. You know, if you, if you love that, um, that brand of noodles or that, you know, you're not going to change or it's a real, real friction point for you to change. Right. Right. Okay. So we've covered bananas. We've covered noodles. Um, and let's also, yeah, right. Right. I think I do. Yeah. Um, very, very high in starch. Um, all right. All right. So let's rewind a little bit. So, so, you know, just to give a lay of the landscape for everyone listening that's still with us. So we've covered the, we've covered the general costs that are involved here. We've covered some of the infrastructural questions. We've covered some of the operational dynamics that go into thinking about this question. We've also covered the e-commerce side from the revenue generating side, because it's not all about eliminating costs. It's not all about efficiency. There are revenue gains that can be made here too, which we just touched on. So now I want to talk about the organizational side of the e-commerce question of profitability, because like you, you actually mentioned it already to a degree, but let's go deeper into it. Like, you know, organizational, you know, organizational dysfunction for lack of a better word can have an impact on this. And I've seen this happen at retailers, particularly over, you know, my almost 30 year career. So what are the organizational barriers that you think prevent grocers from building profitable e-commerce operations? I think that I've seen, like you, I've seen this happen many times. And it's almost quite sad. There's one example I can think of where the marketing was not owned by the e-commerce business. And they'd have this difficult conversation with saying, the marketing team would say, oh, but you're less profitable than the stores. so we're not going to give you any marketing budget guess what they never grew you and then as a result you ended up in this vicious circle which you couldn't you couldn't start to become efficient because you weren't growing so you do you i think in my mind you need to set this up as as a business in its own right um oh interesting okay and i think you know you do you know i do think you need to say you know the commercial teams the people who run the fulfillment um need to be thinking need to be thinking in a joined up way um and to the earlier example you know if you suddenly dump in unpredictable um third-party orders but the supply chain team or the team is running the fulfillment centers or the running the fulfillment in the last mile don't know about it you've just got no hope or they will just staff it up to the entire day as if they're visit the entire day to that peak and you end up you know yes maxing out and at some points but then you know running at 20 utilization for the other times so you you do need to kind of think about the everybody needs to think about the demand and the customer profile in the same way um so to me that is absolutely bloody critical wow okay so you just you just hit me over the head with a hammer on on that one actually that was not something i expected to hear or to talk about and it quite frankly flies in the face of how at least particularly in the u.s most people are organizing themselves because they're organizing themselves around the omni-channel idea but as i heard you talk about it i'm kind of like yeah if you're really and i think this may be unique to grocery. So I'm curious about your perspective on that. When I hear you talk, I'm like, yeah, grocery though, grocery is a different beast. You probably do need to separate this out as a standalone operating model. And Chris, I'm a big fan of Omnichannel, but I think what we've also got to say is, yeah, what does that mean? I think the customer needs to be owned centrally. So if we talk about loyalty, loyalty needs to be owned centrally. There's no, that's possible if you're just doing it in one channel. So the ability to kind of offer customers incentives across the channels, central. Ownership of data, central. Ownership of suppliers, buying, central. But you still need commercial teams because we talked about earlier that you might want to change the ranges. You certainly want to change some of the promotional. On my view, there's a good opportunity to change some of the promotions. the marketing you need to be thinking about in a jointed up way because otherwise you're just going to end up with demand that the guys in the fulfillment central the stores can't fulfill so you do need that kind of vertical from commercial marketing operations that i think all sits in this kind of econ vertical yeah and then the other part is the incentives too right As you're talking, I think you're right. This is what you have to do. But you also got to think about the incentives on the store side. Because the one complaint I always hear is, the digital is stealing my sales in my store. You put this new fulfillment center in and now I'm not selling anything. You've got to think about that. I totally agree. And I think you've got to make sure whatever the store is incentivized on, whatever KPIs the store manager gets, he is not hurt by having if you're doing e-com sales out of his store he's not hurt by that and even frankly if you choose to if his customers are doing e-com he probably should get some credit because you probably do want to be advertising both channels in both is this an organizational structure that is unique to grocery like i kind of think it might be i think it might be And I think the challenges of fulfilling grocery are kind of unique. Yeah. And the margin structures are different too. Margin structure is different. And it is a much more difficult thing to fulfill well than is apparel. So I think it probably is unique. I've not spent a lot of time thinking about it. But yeah, I think it probably is. That's a really great – wow. I might have to write an article about this, Richard, when I get off this podcast. Do I get a reference? I think you're going to. Yes, 100% you are because you flipped the script in terms of how I've ever thought about this. And I imagine how a lot of our listeners have thought about it too. All right. So let's start wrapping it up here. So if a grocer comes to you tomorrow and says, like, let's give an example, a theoretical. Like, well, we're at 15% e-commerce penetration. We're losing money on every order. What should we do? What's your answer to them? Yeah, you need to understand what the hell's going on. You know, if you're a 15%, you shouldn't be losing. You know, if you're a decent-sized grocer of 15%, you should be getting towards making money. So what is it, your basket size? Is it your fulfillment cost? Is it your last mile? But I think, yeah, ultimately, you need to be setting it up so that you've got a customer proposition that customers will shop a full basket and that we're fulfilling their needs. You can pick it efficiently and you can deliver it efficiently in both of those to have answers around aggregation and also automation. Are there grocers in the, I mean, I know you'd say Rolik too, but like, are there grocers outside of Rolik in the US or Europe or anywhere in the world really that you think are getting this right and that people should study? where do i go i'd say there's some interesting models out there you know i've always i've always got a little bit of a soft spot for for picnic in the netherlands which is a milk crown model super efficient very different model um you know we actually compete with it at times but it kind of but it's it's milk crowned um and actually to my mind it's a little bit like what i'll you know what aldi offers you know it's the cheap it will be it's a very cheap way of getting a narrow range of goods to customers to customers doors and actually i think that's kind of interesting um i also think going back to some of my roots you know what you see in china and some of the kind of some of the immediacy some of the kind of fresh that you're getting delivered in china some of the merging of food service and grocery that you get in the chinese operations is really interesting um you know you can now kind of get your lobster cooked put into a put into a meal and delivered with your bananas and that's kind of interesting or you can order the lobster fresh um and that kind of that's kind of interesting as well because you know margins in food service are higher than in grocery so if you can crack that that's kind of an interesting option that's kind of interesting as well okay well great job bringing protein into this discussion finally Not so starchy after all. At the end of the podcast, right, right. All right, so my final two questions for you are on the confessions angle, which is why we call this podcast Confessions of Supply Chain Executives. So my first confession question for you is, is there an uncomfortable truth about grocery e-commerce that most grocers still just don't want to hear? I think e-commerce, so people talk about e-commerce being unprofitable, but actually sometimes e-commerce is hiding the unprofitability of stores. and if you take the volume out of the store you've legal you're left with a store problem and actually some people are kind of misdiagnosing it as an e-commerce problem actually the problem is actually you when you take the volumes out you're left with some stores and that's awkward yeah and people don't want to admit to that wow i'd say the second one is this is a major strategic problem for these guys but the returns often go beyond the incentive programs of the management and you've got to have somebody who's buckling up to say i'm okay to i'm okay that this is going to really be something that's driving my business in five years not necessarily in one year and that's actually that's a challenge you know when chief executives last an average of what three to five years right which which actually should in theory be easier for the regional grocers to actually you know get their heads around because they tend to stay in position longer exactly i think this should be something that suits family businesses down to the ground right sam interviewing you again in say three to five years time is ee grocery going to be profitable for most players are we still going to be having this conversation and say you know 20 29 2030 2031 like what's your take richard what do we know it's going to be bigger this is definitely a one-way street I have a feeling what you're going to get to is a haves and have-nots and you're going to end up with a bunch of people who've invested worked out how to get a model that they own and they own their customers who are going to be on a pretty strong trajectory and you've got a bunch of people who have chosen to stick their heads in the sand and they're going to be the ones who struggle and that for them I just don't see how they're going to reach a point where they're profitable and if they're working with some of the third parties, frankly, the third parties are going to get more and more power over them. And I think the pie is going to go more and more to the third parties, which actually kind of goes back to what I said in China. You end up giving all your profit away. Yeah. And the retail media angle too, because they get a lot of first-party data in terms of making their customers continue to use their platforms too. That's a really interesting point. Correct. Man, Richard, I wish I could talk to you for longer, but unfortunately we got to end it at some point. So Richard McKenzie of Volok, a division of the Roller Group, thank you for bringing what has just been an outstanding level of clarity to one of retail's toughest challenges. Thank you, Chris. As always, I love being here. So I'd love to do this chat again in 2028 or whenever you mentioned. Yeah, right. I'll book you right now. Hey, if people want to get in touch with you, they're listening to this conversation, what's the best way for them to do that, Richard? My email is richard.mckenziemck at volok.com. And our website is www.thelockdeck.com. So love to hear from anybody. All right. Well, there you have it. And of course, today's podcast has been produced with the help and support of Ella Searyord. I am Chris Walton, and this has been Confessions of Supply Chain Executives. Never forget, OmniTalk fans, confessions are almost always good for the soul. Be careful out there. We'll see you next time.