Chipotle Mexican Grill, Inc. (CMG) Earnings Call Transcript & Summary

Good afternoon, everyone. It's John Glass, restaurant analyst at Morgan Stanley. Welcome back to our conference -- our global consumer conference virtual edition 2020. Before we're getting into our next fireside chat with Chipotle, just a couple of housekeeping items to remind you of. One is those listening on the webcast can ask a question. [Operator Instructions] I'll do my best to get to them at the end. We do have a lot to cover, but please ask them early and often. I'll try to weave them in. Secondly, from my perspective, for important disclosures, please see our disclosure website, which is www.morganstanley.com/disclosures or you can just contact your Morgan Stanley sales representative with your questions. Well, with that, it's my pleasure to introduce our next fireside chat with Chipotle Mexican Grill and its Chief Executive Officer, Brian Niccol; as well as its CFO, Jack Hartung. Gentlemen, thank you so much for making time today, and welcome.

Yes. Thanks for having us.

That's good. Brian, I wanted to maybe start with some strategy questions and talk about -- as you talk about the pillars you've mentioned in the past, and forgive me if they're not exactly the way you phrased them, but making the brand more visible and relevant, a disciplined approach to innovation, digital loyalty, and of course, running great restaurants. I wonder if we could just sort of take each of those as you sort of think about where you are in 2020. And I wanted to just ask about making the brand more visible, right? I think at the time, you've talked about that. There's just -- there's a lot of innovation you need to make around advertising, et cetera. 2020 though was a very unusual year, right? Media consumption changed. The strategy had to change. So in terms of that piece of the strategy, what really changed in 2020? What did you learn from 2020 about making the brand more visible and what are the things you think in that perspective carry on into 2021?

Yes, sure. So yes, on the more visible front, I think it was 2 things: one, being more visible for what makes Chipotle unique and Chipotle's purpose, and then the actual physical process of making the brand more visible. So Chris and the marketing team, I think, have done just a really terrific job. 2020 questioned traditional media vehicles as well as new media vehicles on what the right approach was to communicate. But one thing that I would tell you that has helped us be more visible is the fact that now we've got 17 million, soon to be 20 million people in a rewards program. And the reason why that's important is we've been very visible to the people that probably want to be the most engaged with Chipotle on the things that matter most to them. Because we've now gained insight on what do you want to know about the brand. And so we've been able to communicate our safety practices all the way to, most recently -- I just got an e-mail this morning talking about our Farmlink program, where we're taking waste and figuring out how to get food waste -- correction, it's called food waste, but unused food to food banks, right, and at a time where it's really important. So the brand has done a nice job, I think, of using traditional media vehicles. Obviously, the sports world shut down and then it's restarted. We figured out how to use the places where eyeballs are to Chipotle having the right communication in the right vehicle. And you've seen that happen in television all the way to digital. And I think probably the biggest new vehicle for us is this rewards database of, call it, 17 million-plus very active and engaged customers.

That's great. I want to come back to digital and loyalty maybe in a moment. And just finishing on media, as you think about the transition, right, we've gone from mass media to a digital world, maybe to a social world. Is that really saving the company money in the long run or is it just -- you spend the same amount, but you can get massive and better reach because you've got these channels that are cheaper and more efficient? How do you think about the brand awareness and the marketing spend going forward? Is it just more efficient or Is it less? Or how do you view that?

Yes. I think we've done a nice job of thinking about the return that we're getting on every dollar that we're spending in marketing. We've not locked on, this is the absolute dollar you need to spend or this is the absolute percentage, frankly, that equates to that spend. What we'd rather spend our time on is, what is the return that we can get based on every dollar that we invest in each of those vehicles that you mentioned, right? Like, just over Thanksgiving, the guys did a really cool thing with Chase Young and what, initially, you would have thought was just an interesting Twitter video, ends up getting well over a couple of million views because people are really interested in Chase Young. And then all the other NFL players and frankly, athletes that commented on it reposted it. So a very efficient buy that had a great return that we'll probably do more of versus -- I think the guys did a great job of figuring out what television shows we want to be in. And then the return associated with that. And then we also couple it with what do we want to be saying. It's like if you want to be talking about how our approach is to our fresh ingredients, it's a better -- there's a better vehicle to do that than we do it in that vehicle. So I would definitely say we feel really good about the percentage and the absolute dollars that we're spending right now. If we see a need to change that, it will be driven by belief that we've got evidence that we can get more return out of an incremental marketing investment. So I think the team really holds themself accountable to making sure that we're getting great returns on those dollars. And then we really take the time to make sure we got the right message and the right vehicle at the right time.

I wanted to pivot now to just some food innovation sort of keeping -- sort of in the cadence, you've talked about. At a high level, I think it was a couple of years ago, Brian, you and I talked about maybe there is room for a couple of more menu items or a couple of more permanent menu items on the menu, trying to keep this an efficient system where throughput is paramount. Do you think digital is changing that? Do you think your view of how many items you can actually have on the menu now is expanded because digital and those channels are so much bigger and maybe you're allowed to make more product that doesn't really confuse or slowdown the production?

Yes. Look, I think one of the things, John, that's happened is with the scale that we now have in our digital business, we have the flexibility to do digital-only menu innovation. And the experiment we've been working on has been quesadilla for 2 reasons. One, today, we make quesadillas in the frontline, but they're really slow. They take like 1.5 to 2 minutes on a tortilla press to do a full-sized quesadilla. The kids' quesadilla is not as much of a challenge. It goes pretty quick on that tortilla press, but the full-sized quesadilla slows us down. So now that we're doing well over $1 million average unit volumes, you're starting to say, "geez, that digital business, most restaurant companies would kill to just have that sized digital business as their whole concept, right?" So we do think that presents the opportunity to do some unique menu innovation just in digital. And we kind of started this when I first got here with the idea of Lifestyle Bowls, which we're using all the same ingredients, the same cooking platforms. Quesadilla is going to be our first test at using a unique cooking platform to provide a unique menu experience. So the answer is, I think, we're going to be able to do menu innovation now in a digital world. And we'll be able to do menu innovation in, call it, an in-store world. And that's why we have to use the stage-gate process, though, to make sure that we've got the discipline around understanding the consequences of whatever we should decide to do before we actually go do it across all 3,000 restaurants.

Yes. And you've been talking about quesadillas, I think, for a while now. Just remind us where are you in that process? Is it taking longer because it's such an important initiative? Or maybe where are we -- when should we see it at scale, if we're not there yet?

Yes. Look, I think this is a perfect example of our commitment to making sure that we learn and then take advantage of the learnings before we go roll something out nationally. We started this out thinking maybe we could just pull it off on the front line. We got some quick feedback that -- even though it was faster, it really didn't make our experience that much better. And frankly, the product wasn't terrific. So insert, we've got a new cooking platform. And now we feel like we're pretty close on the operational execution. And obviously, we've got to put equipment in all our restaurants in order to pull it off, but we're feeling pretty good about what we're seeing in -- I think it's Indianapolis and Cleveland, where it's a digital-only quesadilla proposition. We think we got the right conceptual proposition. We know the food is terrific. It's -- whenever I can, I get a barbacoa quesadilla. It's just off-the-charts good. So -- and frankly, it's the perfect product for our digital business. If you really stop and think about it, it really is the perfect product for our digital business because it stays hot, it travels really well. And frankly, it's pretty good at on-the-go eating if you need that. It's a pretty portable proposition as well.

Yes. So I want to pivot and talk a bit about digital, right, in its -- all of its manifestation. You brought up loyalty earlier, but maybe just at a high level, I think you said that this year, you may have about $2.5 billion digital business, which is phenomenal. It's almost half of your sales, I think, as of last quarter. Can you just break that down for us in terms of -- when you say digital, there's mobile order and pay or order-ahead. There's the delivery channel. Maybe just sort of size up for us kind of how big each of those pieces of business are today? What are the relative growth rates as we think about that digital business in the future?

Yes, sure. So digital continues to be roughly 50% of our business, a little shy of that. And the thing that's terrific is that order-ahead business makes up greater than 50% of that. And then we've got delivery that happens in the marketplace and delivery that happens in our app, right? And that breakdown is probably like 65-35 marketplace to in-app. So that kind of gives you an idea. And then, look, when we get past COVID, catering will come back. But right now, that's pretty much gone to 0. Not a lot of big catering events going on these days. But the thing that we get really excited about is things like Chipotlane where we give additional access for that order-ahead business, the order-ahead business grows even more. And you see people still utilizing delivery, but the convenience of not having to get out of your car and being able to order ahead and have a specified pickup time, people pivot to that convenience as well as the additional value relative to delivery.

And maybe just -- before we get into Chipotlanes and the discussion around that access, what are the relative growth rates of the mobile -- the order-ahead versus delivery? I think a lot of brands had assumed delivery would grow rapidly, but they were surprised at how many consumers actually just wanted to order and pick it up themselves. And when they first experienced brand through delivery, then they sort of pivoted. How do you see the balance of that business in terms of relative growth rates?

Yes. The order-ahead business growth rate has been one of those things that's constant. Obviously, delivery had a huge spike back in, what was that, March, April, May, and has continued to grow. But our order-ahead business has been the thing that, I would say, over the last couple of months has probably surprised us as well as what you mentioned to how willing people are to order-ahead and pickup. And I think it's also just a function of how convenient it is. And -- look, there's another value proposition associated with picking it up on your own. You can control the situation more. And you don't -- you can avoid those delivery costs. So there's more control and there's less cost. And I think customers think about that, especially for the occasion that they're dealing with. And I think that's why we're continuing to see people move into that order-ahead business.

Yes. And can you talk a little bit about -- I want to talk about this as it relates to margins later, but just as you think about the cost for delivery to the consumer, right, you've experimented with a range of price points to essentially cover some of your costs. What's the [indiscernible] have you been surprised? Has this been growingly absorbed by the consumer base? Has it created favorable reactions from the consumer? Are you pushing more toward pickup, which is your preferred margin channel? How have you -- what have you learned from this process of upcharging for delivery?

Yes. What I've seen, and Jack, feel free to chime in on this, is they are -- for the occasion of delivery, the price sensitivity as it relates to the increase in menu prices that we've taken is pretty inelastic. Where there does seem to be a greater sensitivity as to those actual delivery fee that you see out there, $1, $2, $3, so that's why we've been very, I think, smart about understanding each of those levers associated with that delivery proposition and figuring out where the right place is for the customer to have to pay for the additional cost of that service of delivery. And so we've not seen really any resistance on the menu prices. We have seen some variability on the delivery fee on how that plays out. But for the most part, if you want that occasion, people understand you have to pay for the additional convenience. And then to your point, the good news is we've got a lot of other alternatives, whether it is order-ahead and pickup or order-ahead and grab it through the Chipotlane window.

Got it. And I want to go to now to loyalty. I think you maybe just provided a number. I think last quarter, you said you had 17 million members. Is that a 90-day active number?

Yes.

No. That's not a total cumulative [indiscernible]. So I mean that rivals the size of Starbucks loyalty program, right?

Yes. I know. It's pretty amazing.

It's pretty amazing, right? So I mean, one is, what do you know about -- you must know obviously about those customers. What -- are they different from your base customer? Or are they just -- that is the Chipotle nation, if you will? And how are you using that information? The 17 million, as I said, it's probably among the top few largest loyalty programs in the country now despite only having 3,000 outlets.

Yes, I know. It's been -- John, it's been amazing. The -- what I would tell you is that's one of the things probably that is a tailwind as a result of the COVID crisis is we went from 8 million to 17 million in a matter of months. And that has been driven by, I think, the function that people had to adopt our digital system faster. Like I think this was something that would have happened anyway. It just sped it way up. And the delivery channel was a huge acquisition tool for us to get people into our digital system and then, ultimately, to sign up for rewards. And then what we've discovered is, we got a lot of light and new users into the business. We've got deeper understanding of our medium user, and we've got even deeper understanding on our heaviest user. The thing that I'm really excited about is, we're definitely seeing the ability to influence purchase frequency with all those groups. And we're also seeing a higher level of engagement. We're also seeing a more commitment to the idea of trusting the brand, believing in the brand's purpose. A lot of people would talk about, I love Chipotle's food and I feel good about eating Chipotle. You ask them what do you feel good about, they have a hard time articulating it. And what we're able to do now is, I think, go deeper with folks, so that they understand our commitment to sustainability. They understand our employee value proposition. They understand how we're sourcing ingredients totally different than the industry. You might have seen our Real Foodprint that we just launched recently. So it's got a higher level of engagement. We can do, I think, the CRM classic things that you would think we would do, which is add a drink for extra points, come on the weekends if we only see coming during the week, add [ guac ] in this scenario. And we understand your history of what you've ordered. So if we realize you're always ordering veggie balls, we're not going to serve you up a carne asada ad, right? So we've got that level of understanding, and obviously, the guys are doing it. And then we've got this cohort understanding where we call it journeys that we put all these different cohorts on journeys that drives at their behavior. And what we're seeing is a really nice impact on their frequency specifically associated with Chipotle.

And just to be clear, where are you on that journey? I mean, are you already marketing to those cohorts on their journey and sort of stimulating frequency? Or is that something that comes in '21? Same thing with kind of those bespoke offers, right, among the weekday come over at weekend. Where are you in implementing some of those tools?

Yes. So we started a lot of the experiments on these journeys, and we're doing at greater scale than we have in the past. But I would say you're not going to see the full effect of this until we get into '21, and probably the back half of '21 is probably the way I would think about it because we still -- the good news is, we went from 8 million to 17 million. So in a lot of these cohorts, we have much more significant scale than we had, and we're really learning a lot to understand what influences their behavior accordingly and also making sure we've got people in the right cohorts. So I would say it's experimentation. It's definitely a tool we're using to drive same-store sales growth right now. But I just think it's going to continue to build on itself and become even much more effective as a comp tool as you get into '21, '22 and beyond.

That's great. Let's talk about unit development. It's an exciting part of Chipotle's story. I think sometimes it gets overshadowed by discussion of comps, but you're still very much a growth company. First, just talk about Chipotlanes. Some of us may have been listening to you talk about Chipotlanes, but there's others probably on this webcast that don't. So what is -- just from a layman person's perspective, what is Chipotlane do? What does it look like? What role does it play in your development in 2021 and beyond? And then maybe you can provide further some details.

Yes. So at the simplest form, I think this is the digital drive-through of the future, okay? And I think the way this works is how consumers actually want to experience this convenience of not getting out of their car. And so the way it works is, you use our app or our website. You order whatever customized Chipotle order you want. You select your pickup time and then you pull up to our window. And you basically look the guy in the eye, payment has already been done, the order has already been done. If you ordered a drink, they need to fill a drink. And otherwise, they hand you your food. So it takes all of like a minute or less when you get to the window. So it's really fast. It's really convenient. It's highly customized. And people love the idea of kind of having a reservation for their food. So that's the power in this, and it is our most profitable transaction because it's no different than an order-ahead transaction where you would park, walk in, grab your food and go. The difference is you don't have to get out of your car. So we've got about, I think, what, 150 of them now, Jack?

In that ballpark. We were 126 last quarter, and we keep building them. So yes, right around there.

Okay. And I think you've heard us talk about, we're definitely pushing our pipeline towards Chipotlanes. We are in that 60% to 70% range of Chipotlanes going forward for our new unit openings. And the reason is because the economics are phenomenal. It's minimal additional cost for a meaningful incremental volume, that is our best margin transaction. So some of the things we are still testing with it are remodeling existing stores versus just relocating a store. And we're feeling really good about landlords' willingness, which I would say, about a year ago, they weren't so excited about the idea of converting an NCAP into a Chipotlane. They've now become much more flexible to the idea of converting a Chipotle to a Chipotlane. So we've probably got more opportunity on remodels than we had a year ago, but the bulk of the way this is going to work is going to be through new restaurant openings. And to your point, John, we only have 2,700 restaurants. There's a lot of restaurants in our future, and the bulk of them are going to have Chipotlanes.

And can you talk -- Brian, just put a couple of numbers to that. I think in the past, you said that Chipotlane prior to COVID was maybe did open at 10% higher average unit volume. I'm assuming the same cohort of a non-Chipotlane, during COVID, that's risen substantially. So like what is the first year average unit volume of now at Chipotlane in 2020 as an example, just to give us sort of a sense of what these volumes are capable of. And maybe you can talk to what the more mature Chipotlanes are doing now to the extent you have ones that are over a year old.

Yes, sure. So -- look, I'll start and I'll let Jack chime in on this. The first thing you should know is the order-ahead business grows by -- I think it's like 10 or 15 points now. It seems like as every month goes by, we get another tranche of growth on that order-ahead business. And so that's pretty meaningful. So then our digital business just becomes a bigger proportion of the Chipotle business. And then at the same token, the average unit volumes are higher in those restaurants. What's the latest, Jack, on the restaurants that are comping? I know we've got a handful that are comping now.

Yes. It's like 18 to 20 that are comping, Brian, and they're performing -- when you compare them to new stores without a Chipotlane that opened up at the same time, they are also just become comp. There's a 10% premium -- higher sales for the Chipotlane. They also are comping at a higher level by 500 to 600 basis points to a higher level. And to your point before about the order-ahead, the total digital is like 60% and 2/3 of that is order-ahead. So order-ahead grows and delivery shrinks. And so every single metric that we look at with Chipotlane is better. It's better in terms of sales, better in terms of margin, better in terms of order-ahead. So it's greater convenience and it skews towards that higher-margin transaction.

And by the way, Brian or Jack, just as you -- the second make line, which has been really the enabler of this, is that actually -- you put them in all the stores. They were designed to fulfill these digital orders. You've had this massive surge in digital orders. Are they actually -- do you have to rethink them? Or are they capacity constrained in any way? Or do they work just as you would have thought and they're perfectly capable of handling this kind of new volume that you're doing it from the digital channel?

Yes. I mean, this has been the thing that we are -- me personally, when I first got to Chipotle, I was super excited about because I was like, "geez, we've got a second make line in every restaurant. We're going to digitize them. And they have capacity to do what the frontline is doing." And frankly, we control the queue on that digital make line. So it's less of this unexpected surges, and it's more us controlling the surge. So the good news is, as we understood our demand and our capability to fulfill that demand, we've not run into any place where we've got a capacity issue. And the good news is, to make a burrito, it takes seconds moving down that digital make line, not minutes. And it's a very powerful proposition because, like I said, we're doing now well over $1 million off of that second make line, that digital kitchen. And when our dining rooms come back, we know what we're doing off those front lines before we had this pandemic. So there's plenty of capacity on that digital make line. That's why we're testing things like quesadillas. And when we get our throughput going again on that frontline because we'll have customers standing in line again, it works really well together.

That's great. I want to come back just to development for a moment more. There's been a lot talked about the vacancy or -- the availability of real estate improving. Having said that, a lot of brands are going after the same real estate, right? I suspect it's good for a Chipotlane. It's going to be good for a lot of other brands, right? Everyone is moving to these lower contact, easier digital access real estate. So as time has gone on, has that changed your view at all about availability of real estate for Chipotle going forward? Or do you still see it as attractive as you did prior when COVID first hit when you thought, "boy, this could be a real windfall for you from a development standpoint?"

Yes. Look, I think there are lots of sites available. We're not having an issue where we have demand that outstrips the supply of sites. I think it's quite the opposite. The amount of inbound desire to have Chipotle take up these sites as well as our desire to go find the right sites, it's really rich and robust. So we feel really great about our ability to move back into that 200-plus building range because we now have the people capability. We know when we open the restaurants, we have the operating capability. And now we've got arguably more business than ever to get the return we want out of that physical asset, whether it's a Chipotlane, a traditional Chipotle or like the restaurant we just are testing out right now, where it's a digital-only restaurant. So I really do believe the building opportunity in front of us is tremendous.

And then 2 other questions on development, 1 domestic, 1 international. So from a digital-only standpoint, so what is the thought process there in terms of where does it apply? Is this an urban market strategy largely? Or are you thinking about it in terms of this could actually just be an infill for existing markets? How do you think about that digital-only asset and where it's going to be applied? Or is it just too early to understand that?

Yes. Look, we're testing both of those ideas. It's like is this a theme restaurant? Is it a unique trade area opportunity? And obviously, as we think about new markets, is it also a way for us to get broader penetration faster just because it's less capital involved, and you can reach more people quickly. So the example we're testing right now is a very specific trade area. It's not urban by any means. Highland Falls, New York happens to be a small town with the West Point Military Academy right there that is a heavy delivery proposition and an order-ahead and pickup proposition. That's a unique trade area that if we didn't have this Chipotle execution, we probably wouldn't be in that trade area, frankly, versus we're getting ready to test, I think, it's in Kansas City, where we're going to do a Chipotlane-only restaurant because that's a market that's an older market for us, and we think we can have the ability to infill with these digital-only restaurants to extract more Chipotle dollars out of a trade area that, otherwise, we would say, we're going to have to wait before we're ready to put in another Chipotle. So we're experimenting on all these fronts. And that's why I get really excited about the flexibility in the assets that we have as well as the flexibility or the resiliency we have in our business to be able to pivot to a digital proposition or an on-site proposition.

Just before leaving development, Brian, I understand you're early and you're still in your development in the U.S., right, whether it's 5,000 or 6,000, whatever the ultimate potential is. International was a piece of business at one time. And I think you and I talked a couple of years about that. Where does that fall in your priority buckets now. Is this still, let's think about it later because we've got so much out of us with digital and development in the U.S.? Or are you starting to think about a little more planting those seeds? Where does international now fall in your priority list?

Yes. Look, I would definitely say the U.S. same-store sales growth and new unit opportunity is still the primary focus area around the strategies we talked earlier. But as you look out over the next 3 to 5 years, I don't see any reason why we can't start having success outside the U.S. And so we're going to start the process of using that stage-gate disciplined approach to looking at how we can build-out the markets where we already have a foot in the door and start evaluating new markets where we don't have our foot into those markets yet. So the health of the U.S. business compounded by the fact that now we have a really robust digital business, I think, just gives us a greater chance of success when we go outside the U.S.

Jack, I wanted to maybe turn a little bit to some of the conversation around margins. It's a conversation, obviously, we've had for a number of years about the relationship with average unit volume and restaurant global margin. Just a high level, right -- you've talked about this many times at a $2.2 million average unit volume gets you 22, and a $2.5 million gets you to 25. The actual math would suggest that your incremental margins keep getting better and better, right? In other words, at $2.2 million, you're doing 40% incremental margin. At $2.5 million, you're doing almost 50%. Is that just the nature of the restaurant business, and we've been always thinking about a constant 40% restaurant incremental margin, but in fact, it's not? Or is there something unique about Chipotle that allows your incremental margins to get better and higher as your volumes go up?

Yes. I don't know that I would follow the same math, John. I think our margins at a [ 2 2 ] -- incremental margins at a [ 2 2 ] versus [ 2 5 ] are going to be very, very similar. When we talk about food cost, food cost is 100% variable. Labor variability is very similar at a [ 2 2 ] versus [ 2 5 ]. So I think they're very similar. Now -- you're in the same ballpark. Now what's changing though is digital, especially order-ahead and pick up, it's going to be higher than -- like, let's say, you're talking about 40%. It is going to be higher like in a Chipotlane. Our margins -- if you go back to the same cohort that we talked about before, the comp stores that these 18 or 20 Chipotlanes, they have higher margins than the non-Chipotlanes opened at the same time by quite a bit, partly because it's higher volume. And so you get -- if you're a $200,000 higher volume, you're going to be 200 basis points higher in margin. They're higher than that even by a meaningful amount because the business going through digital and the order-ahead and pickup generates even a higher than 40% incremental margin than you typically see. So I don't feel like it goes up over time, John, but I do think we're able to maintain it at $2.5 million and then $2.8 million and then $3 million. The only wildcard as you get up there is what is your rent going to be and how much of the mix is going to be delivery, which is our lowest margin transaction versus order-ahead. Those are the variables that we see changing as we grow.

Got you. And then, Jack, maybe sticking to the margin conversation, but maybe making a little bit nearer term, right? There was a conversation about margins in the third and fourth quarter versus AUVs, right? So if you think forward about the fourth quarter average unit volumes, maybe close to $2.3 million or above $2.3 million, restaurant margins may not be there. And there are pieces to that bridge that would get you from, let's say, the 19-ish that you talk about to 23. Can you just walk me through what those pieces are, so investors can maybe understand what's temporary with transient, which is not? And how you build the bridge to get to that parity between AUVs?

Yes. John, there's a couple of buckets. One bucket is there's specific COVID direct cost. Like, for example, we still have a lot of people that are -- we exclude them from working, but we still pay them. So for example, if you work in one of our restaurants and you call us up and say, you have a family member or you were possibly exposed to somebody that tested positive, we certainly don't want you to be working. You shouldn't be working. You should take care of your health. But we will pay you for a 14-day period. So we have that cost that's in our P&L right now. You have indirect cost. Because we have so many new people that have come to Chipotle through either loyalty or they're attracted to the touchless experience, whether it's order-ahead or it's delivery, that we're selling actually more burritos than we historically have. Our mix has moved back towards burritos instead of balls, and burritos have a slightly lower margin. The tortilla costs a little bit more than the packaging. We also sell more steak now. And these are things that really, really accelerated with the increase in their digital business. And these things, the burrito and the higher steak incidence, they do skew more towards the digital business. I call those more indirect. Those are having an impact on our margins, but it's -- we think it was triggered by COVID, which triggered more of the digital business. And we think that's likely to level off over time. So we think that's going to be something temporary. And then there have been things that are unrelated to COVID, like higher steak costs. I mean, you could say that the supply-demand out of balance was driven by COVID, but steak cost just got much, much higher in the summer, and we carried that through most of the third quarter. That is also something that we see as temporary. So there's a number of things like that, John, that are in those buckets that when we look at those, we either see that they're temporary and they will go away or they're permanent like the shift to delivery we've talked about. That is something that did impact our margins. We think that, for the most part, we'll be able to recover that through this menu price experimentation that we're talking about. So everything that we see in terms of getting back to our algorithm, we see that we'll be able to get there. It may take us a number of quarters, and we're not going to be in a hurry to get our margin up at the expense of maybe chasing some customers away from coming to Chipotle, but we still feel we'll get there.

Can you maybe just -- a couple of topics within that, Jack. Just -- one is just as you think about that incremental pricing for delivery, is it really designed to fully cover that cost on a margin perspective. So once you fully implement this increased cost, that should really wash out from the P&L. That shouldn't be a [ factor of ] in the fourth quarter or is that not simple?

It doesn't fully cover it, John. It covers a big part of the -- mind our delivery went from -- last year, it was -- call it, it was just under 10%. And now it's better than 20% -- kind of almost in that mid-20% range. So it more than doubled. So the fees related to that more than doubled. And you've seen in our other operating expenses that it has gone up, and it's not coming down. It's now -- it used to be in the low teens, now it's in the high teens, 16%, 17%, 18% range. That's not going down as long as delivery is this higher. So it'll offset most of those costs. It didn't get it all the way there. It's possible we may experiment with some additional levers to pull with fees right now. We're going to leave things as is with the increase that we've taken. But we're close enough on the delivery, John, with all these other things that I mentioned that are either temporary or cyclical that we think we can still get back onto our algorithm.

Yes. And Jack, just on that again, you said there's some -- you're not in a hurry to show people those margins. I can appreciate that. And that there may be some other levers over time that may be weren't originally either not contemplated or you haven't really expressed yet, things that may include supply chain. I'm not sure where labor scheduling is, but there was kind of a laundry list initially about here's all these opportunities from the cost savings standpoint. Where are you on some of those? Is there a bigger opportunity you think in one of those versus buckets versus another in '21 or '22 that we should think about longer term as a margin driver?

Yes. There's no silver bullet, John. I would say we have found some great supply chain efficiencies that are in our P&L right now. You just don't see it because there's so much other noise there. There's additional opportunities that we're seeing as we look to 2021. We also think there are some labor opportunities. I wouldn't say it's huge because I think the teams have done a great job of being very, very efficient. But we think there's a little bit more of efficiency that our digital make line can afford that we're not capturing all of that. But again, during this time, when people are being excluding, so it's putting pressure on staffing, that business has grown dramatically. That's not the time to try to squeeze and try to get every single line item to perfection. The way we think about it is, if we capture all the margin opportunity we have, let's say, by the second half of 2021, we know that we have the ability to build to the 5,000 or 6,000 restaurants at our algorithm. To get there in Q1 or Q2, it's just not that meaningful to us, if it means that we might either put more pressure on our staff or we might risk that we're going to repel some of our customers. With the delivery menu price increase, we felt comfortable with that because we studied very carefully what other restaurant companies were doing, and we were lagging -- dramatically lagging in terms of what fees were being charged, what we're doing with menu prices with delivery. So we felt like that was a safe thing to do. And I think anything we do to go after margin, we're going to be careful about making sure we do what's right for our customers, what's right for our people and that we don't accelerate it too quickly and maybe hurt ourselves in the long run.

And maybe just as a final question, maybe it's both for Jack, you, and Brian. Brian, just thinking about how the organization has evolved over the last several years, I think when you first arrived, there were a number of key positions that needed to be filled from marketing to some development. Where are you in that process now? Is everyone -- do you think that all the essential roles or the important roles have now been filled? Are you adding staffing in certain areas? And then, Jack, as you think about the G&A in the business that carries that -- from a development standpoint, are there places where you still think there's a lot of incremental investment needed or do you think you're sort of in the right size -- obviously, as business grows, G&A goes up, but you're in the right size proportion on the G&A basis? Maybe, Brian, first to you.

Yes. Look, I would tell you, we feel really good about all the key leadership roles. We definitely are going to continue to have to invest in people from a technology standpoint as well as -- we talked about this earlier, we got 20 million, 30 million people in a database. We're going to need more analytics capability in-house. Obviously, we're going to lean on some external partners. And then as we start to think about some expansion outside the U.S., we have to think about how you create the right organization around that as well. But for the most part, I would say we feel really great about the key roles and the key leaders that are in place. My leadership team is in place, stable group of people that, I think, are driving the culture and driving the business effectively.

And Jack, from your perspective, from a spend standpoint, when Brian talks about needing to spend more on from a technology standpoint, I think that's been a theme these last couple of quarters. A lot of restaurants have talked about needing a step-up in spend. Or do you feel like you've already got like proportionately to the size of the business, right, the number -- amount of spend, particularly in that area?

Well -- no. I think in that area, we're going to have to continue to spend, John. We have grown that area quite a bit, both in terms of head count to drive a lot of the innovation that you're seeing, but also with third-party partnerships as well. I see that continuing, probably even increasing. But I think we can find leverage opportunities, and we're going through our planning cycle right now where we're seeing it in some of the more traditional areas, the non-tech areas, the non-data areas, where we can find efficiencies there. We don't have to grow at the same rate that we're growing sales or that we're growing our restaurants so that we can fund the growth in technology and the growth in this data. So I think we'll be able to still drive some modest G&A leverage, while making significant investments in these key areas.

Listen, we're almost at the end of the session. I want to thank you both for your time this afternoon, this morning, depending on where you are. Be well. Happy holidays to you all. And for everyone listening, thank you as well for your -- the audience today, and we'll talk to you soon. Thanks, guys.

Thank you.

Thanks, John.