Earnings Labs

Chipotle Mexican Grill, Inc. (CMG)

Q3 2011 Earnings Call· Thu, Oct 20, 2011

$32.86

-2.35%

Key Takeaways · AI generated
AI summary not yet generated for this transcript. Generation in progress for older transcripts; check back soon, or browse the full transcript below.

Same-Day

+8.28%

1 Week

+11.53%

1 Month

-0.81%

vs S&P

+0.83%

Transcript

Operator

Operator

Good afternoon, and welcome to the Chipotle Mexican Grill Third Quarter 2011 Earnings Conference Call [Operator Instructions] As a reminder, this conference is being recorded. I would now like to introduce Chipotle's Director of Investor Relations, Alex Spong. You may begin your conference.

Alex Spong

Analyst

Hello, everyone, and welcome to our call today. By now you should have accessed our earnings announcement released this afternoon for the third quarter 2011. It may also be found in our website at chipotle.com in the Investor Relations section. Before we begin our presentation, I will remind everyone that parts of our discussion today will include forward-looking statements as defined in the Securities laws. These forward-looking statements will include projections of the number of restaurants we intend to open, comp restaurant sales increases, food cost trends, margins, effective tax rates and shareholder returns, as well as other statements of our expectations and plans. These statements are based upon information available to us today, and we are not assuming any obligation to update them. Forward-looking statements are subject to risks and uncertainties that could cause our actual results to differ materially from these forward-looking statements. We refer you to the risk factors in our annual report on Form 10-K as updated in our subsequent Form 10-Qs for discussion of these risks. I'd like to remind everyone that we have adopted a self-imposed quiet period restricting communications with investors during that period. That quiet period begins on the first day of the last month of each fiscal quarter and continues until the next earnings conference call. For the fourth quarter, it will begin on December 1 and continue through our fourth quarter release in February. On the call with us today are Steve Ells, our Chairman and Co-Chief Executive Officer; Monty Moran, Co-Chief Executive Officer; and Jack Hartung, Chief Financial Officer. With that, I'll now turn the call over to Steve.

M. Steven Ells

Analyst

Thanks, Alex. I'm very pleased with our financial performance during the third quarter, but even more pleased with our continued strengthening of our food culture and our people culture, as well as the stronger bond that we're creating with our customers as they begin to care more about where their food comes from. All of this gives me great confidence that we're on our way to achieving our vision of changing the way people think about many fast food. We talked about food culture at Chipotle, instead of talking about new food products. We do this because of our philosophy of constantly working to serve better-tasting, more wholesome food, and that leads us to improve every ingredient we use. In order to achieve these improvements, we evaluate how our ingredients are raised or grown, improve our recipes and cooking techniques, and improve our kitchen equipment, all with the objective of serving better-tasting food made from sustainably raised ingredients. Though we don't have time to talk about all the food improvements that we're currently pursuing, I do want to highlight 2 changes that happened during the quarter. First, we've made some changes to our green tomatillo salsa that we think may get even better tasting than before. We've increased the percentage of roasted tomatillo we use in the salsa and are roasting the tomatillos longer in order to mellow the tangy flavor of the raw tomatillo and balance the spiciness of the salsa and improve the consistency. We're using larger vegetable cuts so that the new version of the green salsa is thicker and chunkier than it was before. These changes have made this salsa illustrate how we are always looking at how we can change the food we served to make it better. The second food-related improvement I want to…

Montgomery F. Moran

Analyst

Thanks, Steve. We've been delighted to see that our continued focus has led to another strong quarter of growth for Chipotle. Our performance is the result of our relentless focus on building a culture that allows us to develop restaurant management teams that provide an exceptional dining expense. They do this by empowering teams of top-performing crew members to deliver very high standards in terms of delicious food, great customer service and protecting Chipotle's unique economic model. Our restaurateurs continue to lead the way in developing these excellent restaurant level teams. And our entire team of key [ph] leaders is totally focused on developing as many of these elite managers as possible. Every restaurateur has demonstrated ability to select excellent people and to develop them into our future leaders. As we continue to make progress, identifying restaurateurs and bringing them into the program, where each of these great [ph] leaders continues to grow. As of now, we have about 253 restaurateurs, including 23 former restaurateurs who were promoted to apprentice team lead, team leader or team director positions. But our restaurateurs have a much greater reach than these numbers would suggest. More than half of our restaurateurs now oversee more than one restaurant, and nearly 60% of them now directly benefit -- nearly 60% of all of our restaurants now directly benefit from the leadership of our restaurateurs. As we continue to develop restaurateurs, we're seeing the benefits to our entire people culture. You will recall that restaurateurs are paid a $10,000 development bonus for each crew member that they develop into a manager. In the first 9 months of this year, we have paid more in people development bonuses than in any entire year before, more than $1 million so far. While our people development is as strong as…

John R. Hartung

Analyst

Thanks, Monty. We're very pleased with our financial results in the third quarter as our ongoing focus on building a strong food culture, a special people culture and a unique and strong unit economic model continue to lead to better financial results. By creating an extraordinary dining experience and serving delicious food made from the finest ingredients available in an affordable format, we continue to attract new customers and build stronger loyalty with our existing customers. And as people become more curious about where their food comes from, they appreciate Chipotle even more and they're changing the way they think about and eat fast food. Despite some uncertainty among consumers about the macroeconomic environment, we're pleased to report our fifth consecutive quarter of double-digit comps since the economy began to recover last year, with a comp of 11.3% in the quarter. Overall sales for the quarter increased 24.1% to $591.9 million, driven by new restaurant openings and the 11.3% comp. The quarter comp was primarily driven by increased customer visits, while the menu price increase added about 4.6%. As a reminder, about 3.5% of the increase is from the menu price increase, which took effect during the quarter -- during the third quarter, while about 1.1% is due to the increase taken in the first quarter on the West Coast. Comps were higher than the 10% we saw in the second quarter despite a tougher comparison against an 11.4% comp in the third quarter of last year as the menu price increase more than offset the tougher comparison. While we're very pleased with another double-digit comp in the quarter, we continue to see the lowest comp of the day during our peak hours, which is an indication that we have an opportunity to serve even more customers during peak hours…

Operator

Operator

[Operator Instructions] Your first question will come from Joe Buckley with Bank of America Merrill Lynch.

Joseph T. Buckley - BofA Merrill Lynch, Research Division

Analyst

Can I ask just on the low single-digit comp guidance for 2012, kind of how you're thinking about that in terms of transactions and check, given the price increase you'll start the year with on a year-over-year basis?

John R. Hartung

Analyst

Joe, with the current price increase that we've already taken, that will roll into next year. And assuming that we continue to see no resistance, we expect to get probably right around 2% from that. And so the low single-digit comp guidance, we'd expect that we'd get a little bit more maybe 1% or 2% more from transactions next year. And keep in mind every year when we do this, Joe, we take a look at next year. And the low single digits is what we will achieve if we don't change the trend line. Of course, we're always trying to do things like -- with improved throughput and with improving the quality of our ingredients, including the taste of our food. We're always hopeful that we'll be able to change that trend line. But going up against the double-digit comp from this year if we're not able to, we would end up in that low single-digit category. And that's how we've done and always done our comp guidance for the next year.

Joseph T. Buckley - BofA Merrill Lynch, Research Division

Analyst

And then just one more question on the throughput comments, can you talk about some of the bottlenecks you see or some of the opportunities to improve the throughput, what changes you had made? Is it on the production line? Is it something different?

M. Steven Ells

Analyst

Yes, Joe. I mean for starters, obviously, what we want is a team of all top performers. And you want -- especially during your peak hours of lunch and dinner, you want to have every single person on the line be somebody who's experienced in the position that they're working and able to do it very quickly and efficiently while giving great, great customer service, great eye contact, great communication with the customer. In order to make sure that we have those folks on the line, obviously, we wanted many restaurateurs as possible, and that's our top priority. But what we've been working on last year a lot was to make sure that our scheduling and our deployment were everything that they could be. And in other words, we wanted to make sure that we had -- that we did not use too many hours of labor in the morning or before dinner in preparing all of our ingredients. We wanted to use the right amount of labor then, so we have plenty of hours left to have peak deployment during our busiest times of day, at lunch and dinner. So we worked on that quite a lot, and we've improved in those measures, although there's still a lot of improvement yet to go. But when you do get all the people you need ready to work during the peak hours, the question is, a, are they absolutely the right people for the right job at the right time? B, are they all deployed on the line and not distracted by anything else like doing prep work that they shouldn't be doing at that time. And that's what we're working on now. Some of the key things in terms of bottlenecks are we've always found that the process of cashing…

Operator

Operator

And your next question will come from John Glass of Morgan Stanley.

John S. Glass - Morgan Stanley, Research Division

Analyst

Just first on food cost. I just want to clarify if, Jack, we see 5% or mid-single digit inflation next year, combined with the pricing you took this year, the food cost ratios sort of stayed flat with this year? Are you able to leverage food cost in that formula? Can you just maybe be a little more precise about food cost ratios as well?

John R. Hartung

Analyst

Yes, sure. It's still early to know exactly what food cost is going to look like next year, so we're seeing the same things that you guys are. I would think in terms of -- we hope that our food cost will normalize a little bit in the fourth quarter. Avocados were extremely high. They were the highest we've ever paid for avocado in about 5 years or so. But that was not as much sustained inflation-driven as much as it is, it was a light season. And growing avocados is cyclical, and so we would hope that avocados would return to a more normal harvest next year in California. So if you assume that our food cost return to kind of more normalized cost in the fourth quarter, we hope that would be somewhere in that mid-32% range to the high 32% range, and then we think inflation of somewhere in the mid-single digits on top of that. So we don't think that there's anything that we see with what we're buying today that we'll see leverage. We don't see that what we're buying today's going to be less than what we're paying today. So we see, frankly, food cost continue to rise not decline.

John S. Glass - Morgan Stanley, Research Division

Analyst

So just to be clear, mid-single-digit food cost was sort of a mid-single -- sort of less than mid-single digit price increase at least for the first half of the year kind of equals flattish food costs from the fourth quarter into the first half, at least, of next year. Is that the right way to look at it?

John R. Hartung

Analyst

No, I would say, John, that if inflation of 5% creeps basically on top of our, let's call it, somewhere in the 32%, mid-32s, the high 32s, when the full 5% is in, that would be, call it, 150 basis points or so. If that happens throughout next year, we might see 150 added basis points by the fourth quarter of next year. And if it happens evenly throughout the year, we'd have a little bit in the first quarter, a little bit in the second quarter, et cetera, et cetera. Now as you know, inflation never happened that linear -- in a linear way that way. It kind of jumps up and then comes back, jumps up and comes back. But I would think of it in terms of a 5% on top of what we’ve seen in the fourth quarter.

John S. Glass - Morgan Stanley, Research Division

Analyst

And then just to clarify, the tax rate increase, you're assuming this -- if the higher rate -- if the HIRE Act does not get renewed and an R&D credit doesn't, but they normally do get renewed, are you -- I'm confused as to why you would assume that they're not if they normally do.

John R. Hartung

Analyst

Well, It's all political, John. I mean, if they can at the last minute at the end of the year, if they can get these things approved, then our tax rate would be slightly higher, maybe 20 basis points higher, but we would pick up instead of raising by 80 basis points, it would only rise by 20. And we're just throwing it out there because Washington will need to get their act together. They'll need to approve these things. And if they don't, we wanted everyone to know what the impact on our tax rate would be.

Operator

Operator

From Piper Jaffray, we move on to Nicole Miller.

Nicole Miller Regan - Piper Jaffray Companies, Research Division

Analyst

My question is as you add new things to the menu, for example, the brown rice and make that available, are you able to figure out how much that drives frequency of existing guests and how much you're able to get new guests in the door?

M. Steven Ells

Analyst

Well, Nicole, I don't think that adding something like brown rice adds a measurable new stream of customers right away. But I think what it does is I think it starts to develop a connection with our existing guests, letting them realize that now there are more options, especially more options on the health side for someone who's concerned about perhaps eating more whole grain, something like this. And so I think that when they decide what they might want for lunch or dinner, this health component, this new flavor component might make them think about coming more often. I also think it shows to our customers that we are concerned about creating a dining experience that's something that's good for them that they can eat frequently. Fast food used to be a treat in this country a few decades ago, and it's become people's everyday eating. But the quality of the food has not changed. It's not necessarily something that's appropriate to eat every day. And we want to make sure that we're including things like whole grains that are part of everybody's sort of daily healthy diet. And so I think over time, it will help drive more customers in and bring people back more often, but it's not something that's measurable right away.

Operator

Operator

[Operator Instructions] From Deutsche Bank will go to Jason West.

Jason West - Deutsche Bank AG, Research Division

Analyst

Just, Jack, why don't you clarify a little bit the comments on the more recent trends. You've talked about the tough compare. Could you say that you guys are still running in the double digits, just slightly below or is it -- the slowdown been a little more significant than that?

John R. Hartung

Analyst

Yes, it's -- really, we're seeing the -- when we look at trends just from month to month and week to week and adjust for seasonality, they're holding up just as well in October as they did in September and as they did throughout the third quarter. What we are seeing now as we go against these tougher comps from last year, we're seeing a slight downtick, still right around the double digits, but little more than 100 basis points lower. But it's 100% explained by the fact that we're going up against a little bit more than 100 basis point tougher comparison than last year. So no -- I would call it, Jason, no change in dollar trend either up or down, but just a slight adjustment to the comp because of the tougher comparison.

Jason West - Deutsche Bank AG, Research Division

Analyst

Okay. That's helpful. And then just -- could you guys give us an update on the labor investigation while we got you here, just kind of where we stand. Have they moved that to more of a national overview yet or we're still just looking at the 2 markets? And any thoughts on when we may get some resolution there?

M. Steven Ells

Analyst

Yes. I mean, obviously, we continue to cooperate with the government officials in their investigation. The scope of the investigation hasn't particularly widened. And we are producing -- have been producing a whole bunch of corporate documents. But we don't have any reason to believe that they're widening the scope of the investigation. And according to our team of attorneys and so forth, they tell us that it's going very well and that they feel good about the investigation. And in terms of when it will wrap up, it's just very hard to say. We don't have any visibility on when it's going to wrap up, but we're hopeful that it will be before too long, and we can put it behind us.

Operator

Operator

And next we'll go to Jeffrey Bernstein with Barclays Capital.

Jeffrey Andrew Bernstein - Barclays Capital, Research Division

Analyst

Just 2 questions, first as a follow-up to that kind of restaurant margin question as we look to next year, kind of, balancing the food cost component of it. But it sounds like you're talking about 2% price at this point and 5 points of mid-single digit inflation. Just wondering is the goal in your mind to protect the restaurant margin that we see in this year in 2012? Or do you think of it more as a balance based on the traffic trends and, therefore, the margins go up margins go down? But I'm just trying to gauge whether or not if you see inflation higher than that whether you would take pricing with the goal being to actually mitigate the pressure and protect the margin?

M. Steven Ells

Analyst

Yes, Jeff. Long term, we believe that a restaurant margin that we've seen over the last couple of years is certainly sustainable. We think we -- depending on our comp, if we have strong comps, we can expand that margin. We said, in the past, there's nothing about our margin even though it's the highest in the industry right now. As we generate significant comps, we can generate as much leverage, if not more, than other companies out there that are at a much lower margin. And so we think the kind of the record margins we've seen, especially last year before food cost has spiked, we think over time that's sustainable. Now we're not really interested in trying to protect margins quarter by quarter. We know inflation is volatile. We know it's been up and down this year. We've got things like avocados, and those are things that will happen, and they'll be more cyclical and we'd rather wait and see how those play out before we make menu pricing decisions. One clarification to that food cost question, we don't have plans right now to increase prices next year, but we have pricing power. We know the menu price increase that we took in the summer this year. We did not spend all the pricing power. We went out and looked at each of the markets and looked at what competitors were charging. We had room to increase the prices even more. We thought that was a fair price increases, we thought if would get our margins back to within striking distance of what they were the year before, but we wanted to be thoughtful and remain accessible to our customers. If inflation continues, at some point, we'll be in a position to and we will increase our price to make sure that, longer term, our margins are sustainable.

Jeffrey Andrew Bernstein - Barclays Capital, Research Division

Analyst

Got it. And just as a follow-up on the kind of comp trend question. Doesn't seem like you're seeing any change in 2-year run rate, and it seems like we've seen that actually 2-year rate improve of over each of the past few quarters. I know you can always tweak your guidance for next year as you move through the year, but it would seem like you just lapped the double-digit comp with the double-digit comp. I'm not sure why if you were to sustain this 2-year run rate that we're seeing right about now, why it would even fall to that low single-digit range unless there's a different way to think about it. But from a 2-year perspective, it would seem sustainable.

John R. Hartung

Analyst

Yes. Well, the 2 years that you're looking at right now, the first year of that year is comparing against 2009, which we did 2% comp for the year. So you've got -- the first year where we started our comps [indiscernible] was last year. And the comparison got tougher last year, but we -- 2 years ago, we had a 2% comp. Last year, we started our double digits. Now we're double digit not 1/2 a double digit. So I think, Jeff, the challenge is going to be can we do a third year, which is what we're going to be faced with as we get into the middle of the year. The comparisons are a little -- the 2-year comparisons are a little more favorable early in the year when we've got the pricing increase early in the year. But as you get to the back half of the year next year, we're going up against 2 years straight of double-digit comps, so that's where the challenge really is.

Operator

Operator

From Citi, we'll hear from Alvin Concepcion.

Alvin C. Concepcion - Citigroup Inc, Research Division

Analyst

Your store unit guidance calls for an acceleration in openings next year. Just wondering what you're seeing out there in regards to availability of sites and if you're seeing that availability of sites improve?

M. Steven Ells

Analyst

Yes, we're really not seeing a lot more new developments than we had during the past 2 years. So our ratio of new developments and remodels of existing developments is going to remain about the same. That is to say it's going to remain the same as it's been over the last couple of years where it's been more in that sort of 35% new development 65% renovations. As we've told you before, about 4 and 5 and 6 years ago, that number was flip-flopped and about 2/3 of our stores were coming from new development. So we're really not seeing particularly more new developments. What we are able -- we've got a lot of confidence based on our new store openings and based on our A Model strategy that we can be fairly aggressive, certainly, in all of our proven markets and even take some chances in working on some A Models in our developing markets in order to continue to get more new restaurants. And so that's what we've done for next year, and we're very optimistic that we're going to continue to have terrific openings next year even that at this increased pace of growth.

Alvin C. Concepcion - Citigroup Inc, Research Division

Analyst

Great, and I think previously, in the past, you've talked about needing something in the range of a mid-single digit, same-store sales with normal inflation to maintain our growth margins. Is there anything different in that dynamic as you're looking at 2012?

John R. Hartung

Analyst

There's not. I would say we need that mid-single digit just to maintain our margins when you have kind of normal inflation. Now normal inflation, it depends on if food inflation is in the 3% to 4% range, that's kind of normal. If it's in the 6%, 7%, 8%, 9% or 10% range, that's not normal. And just like you're seeing this year, even though we've had very strong comps, even though we've had leverage in every single line item, our leverage, at least at the restaurant margin level, on 11% comp where we can't overcome 250 basis points of higher food cost. And that's effectively a 12% inflation, if you compare it to the third quarter of last year. So it -- largely, in this environment, it largely depends on what food inflation is for the year and in each individual quarter.

Operator

Operator

And your next question will come from Bart Glenn with D.A. Davidson. Bart Glenn - D.A. Davidson & Co., Research Division: I was just curious. My understanding was the West Coast when they received the initial price increase, that was the sort of catch-up with national pricing based on the cost structure. Now would the intention be to still have an additional wave of price increases to catch them up to the rest of the nation?

M. Steven Ells

Analyst

Haven't decided yet, Bart. One thing by getting them off-cycle, they had a price increase in 2010 around the second quarter, like in May, the May-June time frame. And so by increasing again in March, we did accelerate that a little bit. And so we'll continue to monitor prices out there, monitor transactions, and we haven't made any short-term decisions on whether we're going to raise prices again. But our long-term objective would be to continue to take steps to allow California pricing to catch up because it's very expensive to do business out in California. And for their prices to -- while they're closer to a typical Chipotle market across the country, they're still a little bit behind. They're at least within a few percentage points, so they're at least in the ballpark now.

Operator

Operator

We'll hear from Larry Miller. [RBC Capital Markets]

Larry Miller - RBC Capital Markets, LLC, Research Division

Analyst

I wanted to circle back to the throughput commentary, and I was hoping maybe you could share some metrics. You've done that in the past where you've talked about, I guess, labor efficiency per hour. Maybe that's not the right one but the number of people you can serve per hour and kind of where you're at and where you think you can get to. Or any kind of metrics you could put around that. Then I also had another question, if you'd let me.

M. Steven Ells

Analyst

Yes, Larry. I guess to be more specific, I mean, right now at our peak hour on Fridays at lunch, we're doing about 110 transactions in an hour. That's this time of year, which, of course, is a slower time of year than the April, May, June, July time where we have even more transactions coming through the door. Our very best year ever was 2007 where that number was 112. So we're 2 transactions off of where we were then during our peak hour. And so one could argue that, that's pretty darn good throughput. But what we also look at when we analyze this is a few things. Number one, we have more transactions now coming through our doors than we had in 2007, higher volumes but, more importantly, more transactions. And we have better restaurant teams. We have a lot more restaurateurs. Well, we have about 240 more restaurateurs. We have much stronger field leadership, and we have much more impressive people joining us in the crew ranks than ever before. So we believe that we have all the ingredients to have much, much better throughput than we've had ever in the past, and we won't be satisfied until we accomplish that. All of the regional directors -- and we were in a meeting here in Denver just recently where this was a primary point of discussion, and all the regional directors believe solidly as do all of our field leaders that we can get much, much better in throughput because we know how to do it. And it's kind of funny. I mean, we got sort of bogged down a little bit, I think, in late 2008 and 2009. We're just kind of watching the recession and dealing with trying to do everything we could to keep transactions solid and coming in. But now in 2010, we focused more technically on throughput in terms of getting tools in place that will allow us to make sure that we had all the labor we needed during peak hours to do what we know we can do. And now I think it really comes down to just simply doing what we know we can do with those teams. And given the fact that we've had significant turnover since 2007, we're really talking about kind of reteaching that which we know how to do to allow the people who, while terrific, might not know the techniques well enough. So it's -- we don't think it's going to be particularly difficult to really improve throughput, it's just something that now we have to turn our full focus on in the rest of this year and throughout next year.

Larry Miller - RBC Capital Markets, LLC, Research Division

Analyst

Okay, that's helpful. I was just curious if you could expand on that. Maybe you see, you think it could be higher than you all believed because of some of these things you've been working on. Any order of magnitude? Am I right when I just did the quick math of roughly 5 to 6 people per day. So if you get a few more in per hour, that's about a point in traffic comp, is that right?

Montgomery F. Moran

Analyst

Yes, that's right. We have about 500 transactions per day per restaurant. And so if we can get 5 more through, that will be 1%. The question is, is that 1% an incremental addition to our customers? Or would those people have waited or would they have walked away? We believe that our lines are long enough during our peak hour, particularly lunchtime, that people are walking away from the end of the line. If we can shorten that line and create even more of a reputation for being very, very quick, speedy, people will either choose to stay in line or be less intimidated by the lines than they might be today. So we think it's -- that there will be an incremental benefit of people staying in line and us increasing the overall amount of transactions per day in our restaurants. And certainly, something like 5 a day or 1% would be something that's highly achievable in my mind. And hopefully, we'll be able to do better than that over the next few years.

Larry Miller - RBC Capital Markets, LLC, Research Division

Analyst

That was very helpful. I was just curious. You gave us, with the 2012 unit growth, you gave us the mix of existing versus new markets. You also gave us the mix of sort of A Models versus traditionals. Do you have any sense of sort of the any sense of the numbers that might be sort of what you would call large urban markets and smaller urban markets?

M. Steven Ells

Analyst

Larry, most of our -- if you're talking urban, meaning real urban like New York, Chicago, San Francisco, something like that, that's a relatively small number of our openings. Typically, it's in the 15% to 20% range. I would expect our openings next year would not vary from that very much. So most of our opportunity is more in the suburban. If -- and when I say that, we have 100 restaurants in Chicago, but it's -- maybe 20% or so of those are in downtown Chicago. The rest are in what you might refer to as Chicago Metropolitan.

Larry Miller - RBC Capital Markets, LLC, Research Division

Analyst

And then final for me. What are the -- Jack, if you could expand on that, what are the differences, if there are any, in sort of urban level volumes as you describe them and suburban level volumes. Is there anything material difference in volumes.

John R. Hartung

Analyst

Well, it depends, Larry. Sometimes in urban restaurants, like, let's say, it's a downtown Chicago in The Loop. We'll be very, very busy at lunch, to among the highest lunch that we'll see in the whole country, but then do very little on the weekends, and so they might end up being an average or below-average restaurant. If you can envision a store now at Manhattan where you've not only got a lot of offices, but you've got residents nearby as well and then shopping on the weekend, that could be not only very, very busy lunch during the week, but you've got dinnertime, you've got weekend, and that could be much, much higher. I would say generally our urban restaurants are a little higher volume, but our very highest volumes in the country are -- more of them are not urban. Like our top 20, for example, more of the top 20 are I would consider to be not urban restaurants, if that helps.

Operator

Operator

And your next question will come from John Ivankoe with JPMorgan. John W. Ivankoe - JP Morgan Chase & Co, Research Division: Actually, I think somewhat of a follow-up on that. Could you discuss the A Models. I think 30% again, if I heard that right, in 2012. And since the units were in 2010, 2011 we're achieving kind of average unit volumes, I guess, relative to the bigger units, could you discuss what you've learned from those from, I guess, a productivity perspective? And I guess importantly, why there's not even a higher mix of those units in the new unit development considering it's a lower investment cost?

M. Steven Ells

Analyst

Yes, I mean, that's a great question. We've learned a tremendous amount from our A Model strategy. One thing we've learned is that there is a tremendous amount of pent-up demand for what we're doing throughout all the markets where we're -- especially all of the proven markets where we have an established group of restaurants. And so we were delighted to find that in going into these sort of less main on main sort of locations, these sort of between your locations that we were able to get huge volumes out of those locations. And we built these restaurants very carefully from a cost standpoint and knowing that we had a much better operating and occupancy cost knowing that we could sort of get away with volumes not as good as our traditional restaurants. And what we found was the volumes, the sales volumes very nearly [indiscernible] traditional restaurants, which meant that our return on investment was better than our traditional restaurants because of the lower cost structure. So we learned a tremendous amount about the amount of pent-up demand. We've gained more confidence that we can go into places where we had hesitated to go earlier. And I think also very importantly, we learned a lot of lessons that we are able to apply now in terms of the cost structure to the rest of our restaurants, in other words, to our traditional restaurants. In terms of why we wouldn't build more of them, I mean the answer is we would. It's really based upon how many of these we can find. When we look at the traditional locations, which still account for approximately 70% of our new restaurants, those are great, great opportunities for us. The only way to increase the amount of A Models we will do as a percentage will be to think about decreasing the amount of traditional restaurant. And there's no reason to do that given the fact that there's still a wonderful business model. And so I would say we're out there aggressively. In our proven markets, we are aggressively finding all the A Model sites we can and all the traditional sites we can, and the balance just happens to fall out in that sort of 30% range. So it's not that we're sitting here with a clipboard deciding 30% is the right number of A Models. It's just a question of what the opportunities are with the inventory is out in the marketplace and where we decide that we are able to put new restaurants.

Operator

Operator

Everyone, that appears that is all the time we have for questions. At this point, I would like to turn the conference back over to management for any additional or concluding remarks.

Alex Spong

Analyst

Thanks for joining us today, and we look forward to speaking with you next quarter.

Montgomery F. Moran

Analyst

Thanks, everybody.

John R. Hartung

Analyst

Thanks, everyone.

M. Steven Ells

Analyst

Thanks.

Operator

Operator

Ladies and gentlemen, that does conclude today's presentation. We do thank everyone for your participation.