Cheesecake Factory Inc (CAKE) 2020 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to The Cheesecake Factory's Second Quarter Fiscal 2020 Earnings Conference Call. (Operator Instructions) Please be advised that today's conference is being recorded. (Operator Instructions)

I would now like to hand the conference over to your speaker today, Stacy Feit, Vice President, Investor Relations. Thank you, please go ahead.

Thanks, Rob. Good afternoon, and welcome to our second quarter fiscal 2020 earnings call. On the call today are David Overton, our Chairman and Chief Executive Officer; David Gordon, our President; and Matt Clark, our Executive Vice President and Chief Financial Officer.

Before we begin, let me quickly remind you that during this call, items will be discussed that are not based on historical facts and are considered forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Actual results could be materially different from those stated or implied in forward-looking statements as a result of the factors detailed in today's press release, which is available on our website at investors.thecheesecakefactory.com and in our filings with the Securities and Exchange Commission. All forward-looking statements made on this call speak only as of today's date, and the company undertakes no duty to update any forward-looking statement.

In addition, throughout this conference call, we will be presenting results on an adjusted if-converted basis, which reflects the potential impact of the conversion of the company's convertible preferred stock into common stock. An explanation of our use of non-GAAP financial measures and reconciliations to the most directly comparable GAAP measures appear in our press release on our website as previously described.

Finally, please also note that our press release includes an additional table, calculating basic and diluted net loss per common share, which shows the line item impact of the company's convertible preferred stock. Holders of our preferred stock participate in dividends on an as-converted basis when declared on common stock. As a result, our preferred stock meets the definition of a participating security, which requires us to apply the 2-class method to compute both basic and diluted net income per share. The 2-class method is an earnings allocation formula that treats participating securities as having rights to earnings that would otherwise have been available to common stockholders.

In addition, as our preferred stock is a participating security, we are required to calculate diluted net income per share under the if-converted method, in addition to the 2-class method, and utilize the most dilutive results. In periods where there is a net loss, no allocation of undistributed net loss to preferred shareholders is performed as the holders of our preferred stock are not contractually obligated to participate in our losses.

Now for the agenda. David Overton will begin today's call with some opening remarks, and David Gordon will provide an operational update. Matt will then briefly review our second quarter results and provide a current financial update.

With that, I'll turn the call over to David Overton.

Thank you, Stacy. After pivoting quickly to an off-premise-only operating model in March with the onset of COVID-19, we saw a solid increase in Cheesecake Factory's off-premise sales in the second quarter as we geared up to reopen our dining rooms in mid-May. We have taken a deliberate approach in our reopening strategy as the health and safety of our teams and guests remain our top priority. In addition to securing adequate PPE for our staff members and implementing additional safety protocols, we also made a number of operational changes and technology upgrades, including contactless menu and payment technology, text paging, in order to ensure the best and safest possible experiences for our guests and staff.

Our strategic decision to maintain our restaurant management teams has enabled us to reopen our dining rooms very effectively. We have seen strong pent-up demand with many Cheesecake Factory locations running a wait on weekends. Our large restaurant footprint and flexible seating layouts are enabling us to recapture meaningful sales levels despite capacity restrictions. During the second quarter, we managed the business well with strength across the Cheesecake Factory restaurant key performance indicators, including food efficiency, labor productivity and controllable profit.

With some states reversing course on their reopenings, our operators have done an extraordinary job navigating evolving regulations. As COVID cases continue to rise in many parts of the country, we know the situation will remain fluid. We believe our operational excellence and strong financial position will enable us to navigate this dynamic environment as effectively as possible.

Tomorrow, we're celebrating National Cheesecake Day. To commemorate the occasion this year, we're launching 2 new cheesecakes, Chocolate Caramelicious Cheesecake Made With Snickers, and our Low-Licious Cheesecake, which is low-carb, no-sugar added and gluten-free. Now more than ever, there is so much need in our communities, so we have decided to use our platform this year to support Feeding America's hunger relief efforts with a donation of $1 for every slice of cheesecake sold tomorrow. Our marketing team has generated great coverage in the press with over 1 billion media impressions. While our current reality is quite different this year, we look forward to continuing to provide our guests delicious, memorable experiences, whether in our dining rooms, in our patios or the comfort of your own homes.

With that, I'll turn the call over to David Gordon.

Thank you, David. While COVID has presented extraordinary challenges and the second quarter was anything but normal, the power of The Cheesecake Factory brand, our strong guest affinity and our talented teams remain a constant. As David mentioned, our guests have been very excited to return to our restaurants when indoor dining rooms reopened, and our off-premise business remains well in excess of pre-COVID levels.

We continue to see strength in the off-premise channel at both reopened restaurants and those continuing to operate an off-premise-only model, with off-premise comprising approximately half of our total sales quarter-to-date. In fact, since reopenings began, we have maintained, on average, 90% of elevated COVID off-premise sales and Cheesecake Factory restaurants with reopened dining rooms.

This has contributed to our recapture, on average quarter-to-date, of nearly 80% of prior year sales levels in these reopened locations. This would equate to over $8 million per unit, on average, on an annualized basis despite 50% capacity restrictions. And we've seen sales at The Cheesecake Factory restaurants that are operating an off-premise-only model continue to accelerate, with current weekly off-premise sales equating to $4.2 million, on average per unit, on an annualized basis.

In states that have rolled back their reopenings, including California and Florida, we're fortunate to have sizable patios in many locations to continue to serve guests in accordance with outdoor-only dining restrictions. In some cases, given the strength of our relationship with many of our landlords, we've also been able to access additional space to augment our outdoor seating capacity. Concurrently, we have seen off-premise sales accelerate in most of these locations as our guests want to enjoy the Cheesecake Factory despite the additional restrictions.

In total, our restaurants open for outdoor dining only are currently doing volumes of nearly 90%, on average, of Cheesecake Factory locations with indoor dining rooms open. Currently, 146 Cheesecake Factory locations have indoor dining rooms open. 36 are open for outdoor dining only, and 22 locations are operating an off-premise-only model in accordance with local mandates. And we have one Cheesecake Factory location that remains closed in San Francisco.

In aggregate, across restaurant operating models, fiscal third quarter-to-date through July 26 comparable sales at The Cheesecake Factory restaurants are down approximately 32%. The breadth of our menu and our dessert offerings continue to differentiate the Cheesecake Factory for all occasions. Our sales performance has been supported by on-brand marketing campaigns, mostly focused on the off-premise channel leveraging the Cheesecake Factory brand identity as a dessert leader and the breadth of our menu to meet demand across all dayparts. Our 4th of July burger offer resonated with new and existing guests as well as the media, with coverage in over 120 outlets, generating over 4 billion media impressions as well as significant social media engagement. The guest response was so tremendous that we had to reduce the duration of the offer given that demand far exceeded our forecasted supply levels.

Notably, a majority of sales for this promotion came through our online ordering channel. As we reported on our last call, our own online ordering platform eclipsed delivery for the first time since its launch in 2018, and that strength sustained in the second quarter, with the channel comprising over 40% of our off-premise sales and delivery contributing approximately 35%. The Cheesecake Factory brand also continues to perform well in the delivery channel. We saw both our average customer order frequency and average customer order value increase during the second quarter.

Guest affinity for The Cheesecake Factory brand also continues to support performance in the retail channel. We achieved our highest level of consumer-packaged goods royalty revenues during the second quarter driven by continued strength in sales of our famous Brown Bread and building awareness of our new ice cream offering.

In regard to North Italia, we began reopening indoor dining rooms towards the end of May and soft sales built throughout the second quarter. Currently, 17 North locations have indoor dining rooms open and 6 are open for outdoor dining only. Quarter-to-date, through July 26, comp store sales are down approximately 32%. During COVID, we've continued to build out North's off-premise business, which was in its infancy prior to the acquisition. We've leveraged our existing relationship with DoorDash and have also launched an online ordering platform for takeout, utilizing the same technology that we use at The Cheesecake Factory. With building awareness through marketing campaigns and solid operational execution, off-premise sales currently comprise over 30% of North's total sales.

The FRC concepts began reopening dining rooms in mid-May and also saw sales built throughout the second quarter, with continued strong contribution from off-premise. Currently, 39 FRC locations have indoor dining rooms open, 4 are open for outdoor dining only and 7 locations remain closed.

As we move into the back half of the year, we know volatility in the restaurant operating environment will remain as COVID continues to impact consumer behavior and our ability to fully operate our dining rooms. I am incredibly proud of how nimble our teams have been, constantly adapting to whatever changes come their way. While there is little doubt that 2020 has been the toughest period the restaurant industry has ever faced, I'm confident in our ability to continue to navigate through this dynamic landscape, and we expect to emerge stronger on the other side of the pandemic.

And with that, I'll now turn the call over to Matt for our financial review.

Thank you, David. Second quarter comparable sales at The Cheesecake Factory restaurants declined 56.9%. For context, we saw sales ramp throughout the quarter with the reopening of our dining rooms beginning in mid-May from a low of down approximately 66% in April to a 42% decline in June.

Revenue contribution from North Italia and FRC totaled $37.2 million. North Italia comparable sales declined 59%. Sales per operating week at FRC, including Flower Child, were approximately $46,200. And including $14.7 million in external bakery sales, total revenues were $295.9 million during the second quarter of fiscal 2020. Note, we had 29 locations across our concepts, including 3 Cheesecake Factories closed for much of the quarter and exited Q2 with 19, including one Cheesecake Factory still closed.

As usual, I'm going to provide year-over-year detail on expenses, but of course, note that the significant disparity in revenues, given the impact from COVID in the second quarter this year, drove abnormal year-over-year variances. Cost of sales increased 210 basis points, reflecting a shift in sales mix as well as inflation in meat and dairy. Labor increased 530 basis points, which was primarily attributable to the cost of maintaining our full restaurant management team in the reduced sales environment as well as higher group medical insurance costs associated with company paid health care benefits for our furloughed staff members, which concluded June 30 and higher large claims activity, partially offset by a benefit from the employee retention credit and The CARES Act.

Other operating expenses increased to 41.1% of sales due primarily to sales deleverage, increased marketing and costs for additional cleaning, PPE and other expenses associated with COVID. And G&A increased 590 basis points, also reflecting sales deleverage, partially offset by a reduction in the corporate variable compensation accrual. Preopening costs were approximately $2.1 million in the quarter, associated with our new unit development department and restaurant management bench. We recorded a $2.4 million impairment charge associated with the Stamford, Connecticut Cheesecake Factory location, which discontinued operations last week as its performance was not meeting our expectations.

Finally, during the second quarter, we recorded $11.7 million in COVID-19-related expenses, including the health care benefits for our furloughed staff members and the cost for additional cleaning, PPE and other expenses that I mentioned a moment ago as well as some other smaller items. These expenses were primarily captured in the labor, other operating expense and cost of sales lines on the income statement.

GAAP diluted net loss per share was $1.61, reflecting the potential impact of the conversion of the company's convertible preferred stock into common stock. And excluding the COVID-related costs, impairment charge as well as other special items, including $1.1 million in acquisition-related costs and a $1 million credit to the acquisition-related contingent consideration, compensation and amortization driven by an increase in our interest rate during the quarter, adjusted if-converted net loss per share for the second quarter of 2020 was $0.87.

Now turning to our balance sheet and cash flow. We ended the second quarter with $250 million in cash and $376 million in debt. This reflects cash used in operating activities of approximately $2.7 million, CapEx of $13.7 million, a $4 million repayment on our credit facility to free up room for letters of credit and the $200 million convertible preferred stock investment from Roark that we closed in April. A $3.7 million dividend for the second quarter of fiscal 2020 was paid in kind to holders of the company's convertible preferred stock.

Our second quarter cash burn came in below what we had projected at the time of our first quarter earnings call given our sales recovery and good working capital management. At present, our Cheesecake Factory restaurants are cash flow positive, on average, at a $6 million annualized average unit volume equivalent with our current cost structure, including continuing to retain all of our restaurant management teams as well as an accrual for full base rent.

Importantly, our 33 Cheesecake Factory locations opened with 50% indoor dining capacity for the full month of June achieved 87% of prior year sales volume with an almost flat year-over-year four-wall margin of approximately 17.5% for the month of June. From a cash perspective, the company has paid a substantial majority of its rent payments through July after giving effect to various abatement and deferral structures in place for certain lease agreements.

While we will not be providing guidance given the level of continued uncertainty associated with the virus, we want to provide some color around our cash flow and capital allocation expectations. At the company level, with the current sales trends and our reduced G&A structure, we are roughly breakeven at the operating cash flow level. We continue to expect approximately $5 million per quarter of maintenance CapEx, and we are currently estimating $5 million to $10 million per quarter of growth CapEx for the back half of the year to finish construction of nearly complete new units under development. We are monitoring operating conditions in their respective markets to determine when to move forward with these new unit openings.

In closing, while the industry operating environment has been challenging, we have managed the business well across brands both operationally and financially. Our sales ramp has produced results in line with our modified expectations for COVID, and we continue to prudently manage costs and our cash position. We believe COVID will continue to bring volatility to the restaurant industry and our business given trends in the virus and the constantly changing patchwork quilt of regulations. Financial strength and flexibility will continue to be a key determinant of the restaurant concepts that can endure COVID. With the strength of our brands, teams, culture and balance sheet, we believe we are positioned to not only continue to manage through the pandemic but also thrive on the other side.

With that said, we'll take your questions. (Operator Instructions) Operator?

(Operator Instructions) And your first question comes from the line of Sharon Zackfia from William Blair.

I guess the -- there are 2 questions, really. Thank you for the context on breakeven cash flow. I'm wondering, though, what kind of level of comp you need at this point to get to breakeven corporate EBITDA. I think that would be helpful.

And then secondarily, given the $4 million-plus AUVs you're doing in off prem only, what does that do to kind of inform future restaurant development thoughts? I mean you've clearly had a seismic change in your business, and I know you're not developing new units per se right now, but there's got to be some learnings here that kind of inform how you think about the business going forward.

Sharon, hope you're doing well. This is Matt. I appreciate the questions. So on the first one, I think we may have answered it, but let me just clarify for you. We did say at current sales levels or breakeven at the company level cash -- operating cash flow. And so I think that's what you're referring to. And our quarter-to-date Cheesecake Factory comps are down 32%.

With respect to the off premise, we do believe it will be sticky. I think consumer trends have definitely changed over time. I don't know that it will greatly influence how we develop going forward given that our restaurants are really set up well to do this already. So I don't know that we would credit ourselves of having the hindsight to see all this, but the foresight -- but we're certainly happy with the productivity that we can get. And we want to continue to have, as David Overton mentioned, all of the options available, whether that's in restaurant, on the patio or at home. So I think we would be happy continuing with these trends, and our restaurants would look and feel similar as we continue to grow.

I guess maybe a follow-up on that. I mean does it make you question the size of restaurants going forward, though, if you can do these kinds of volumes at capacity restrictions with the kind of off premises that you're doing?

Well, Sharon, we do have a number of size restaurants. So they are as small as 5,500 square feet. And so we choose the restaurant based on the demographic and what we think we can do in our estimates. So we have from 10,000 down to 5,500, and it's a possibility that we would want a little bit smaller dining rooms, but that remains to be seen.

Your next question comes from the line of Nicole Miller from Piper Sandler.

I thought it was very interesting in the prepared commentary to say you were on a wait. And clearly, that says a lot about the brand equity of the Cheesecake Factory. But what does that also tell us about consumer behavior? What do you think are permanent shifts? And when and how do you think the consumer really wants to go out to eat and that's what they're going to do?

Well, Sharon -- I'm sorry, Nicole, this is David Gordon. Good to hear your voice. I think that we know that people want to get out. And what's been interesting is that even the restaurants that opened their dining rooms early on, so at the beginning of the quarter have seen waits throughout the entire quarter. So it's not just that they want to get out because they were pent up and had to get out for a couple of weeks. We've seen that consistent behavior happen all throughout the quarter. And we're on waits anywhere from 20 to 60 minutes on the weekend. And that's in every geography, in every size restaurant. So it's not even like -- it's just happening in a geography perhaps where COVID isn't as impacted. So it's a good sign. And to your point, I think it shows the affinity for the brand.

And it does show that a lot of the surveys that were out there in the beginning that one of the first things people would want to do once they got back together was go out and dine. So we think we're well positioned to capture those dining occasions, whether that continues to be in the restaurant on a wait or continuing to build the off-premise business and sustain the off-premise business that we've had.

I don't want to break the rules. So I won't ask another question, but that's -- it's helpful. And again, fascinating, and I'm sure there's a way that you interact with them to get all the data you want as they wait.

Your next question comes from the line of David Tarantino from Baird.

Question about the sales trends you're seeing in the locations that have dining rooms open. I think you said for the Cheesecake Factory, you're recapturing about 80% of the sales. And I was wondering if you could maybe think about throughput constraints that you might be seeing. And I guess the nature of the question is do you think that number would be higher if you were operating at full capacity. In other words, do you think the consumer demand is there to get you back to 100% and it's just a matter of getting the capacity back in the restaurant? Or do you think there's maybe less demand now than what you had before, if that question makes sense?

Thanks, David. It's David Gordon again. I think that any capacity that we have, number one, we can handle. So I think that restriction of 80% is just based on the 50% dining room capacity. We see that people are waiting just like they were waiting previously. And we don't see any increase in people walking away or not being willing to wait.

So I think as Matt said also, the off-premise business, we think, is going to be sticky. So as restrictions continue to lift, not just in those 50% capacity but in the 25% moving to 50% and, who knows when we could even get to 75% for dining rooms reopened again, we think we'll be able to keep a decent percentage of the off-premise sales, be able to handle that capacity and get back to where we were before in the dining rooms, with the waits on the weekends and even the waits during the week.

And David, this is Matt. I think, just mathematically, you can clearly see that in the fact that the restaurants with the dining rooms at 50% are doing higher volumes than the patio only, right? So as we have more capacity, we are doing measurably more volume. So I definitely think that's the case.

So I guess a follow-up and related question would be, do you think -- on the other side of this, when you have 100% capacity back, is it your view that the volumes could go well above where you were previously because of the off-premise sales?

I think that, that was true anyway. I think we believe that anyway. We have been down a little bit off of our peak throughput. So there was capacity that we could have tapped into. But I think also, really, what we've seen with the increase in the off-premise further validates that the total sales output from our restaurants could definitely be above where we were right before COVID.

Your next question comes from the line of John Glass from Morgan Stanley.

My question has to do with what your experience has been with how you're rethinking the business model through this period. A lot of your peers have talked about structurally reducing primarily hourly labor by whatever amount, 150, 300 basis points, that they think can stay post COVID. How do you think about that?

And on a related topic, some of that's come from menu simplification, and obviously, the calling card of Cheesecake Factory has been menu diversity. Have you taken a look -- if you reduce the menu, have you taken a look at that and said maybe there is an opportunity to reduce items if that were to improve labor productivity? Or is that not an opportunity you think is worth pursuing?

John, this is Matt. I'll just start sort of on the financial aspect of that. I think it's very telling that in the 33 locations where we had the 50% capacity for the full month of June, we were able to achieve flat margins year-over-year. And so the power of the Cheesecake Factory -- and that's with the full menu, right? The power is driving those volumes. And if you're getting $8 million, $9 million, $10 million, we're able to flex in a way that many of our competitors just aren't able to do. I think that they're in a position where they have to look at reducing guest options, which is what that amounts to. And so we feel really good about being able to manage our business through this and still provide 100% of what makes the Cheesecake Factory brand so great.

That doesn't mean that we won't look for ways to be more efficient, of course, we will. But certainly, it doesn't need to be in the menu simplification or taking anything away from the guest. I think it just has to do with really sort of fine tune in every line item where there's a little bit of opportunity and not making any sort of game changer type of changes.

Your next question comes from the line of Dennis Geiger from UBS.

Wondering if you could talk a bit more about how you're thinking about further gains across the portfolio but also within the reopened restaurants from here. Is it really more about increased consumer mobility and consumer willingness to dine out? Or do you feel like a lot of this is kind of things within your control? And what's most significant there? Are there ways to increase capacity within the restaurants when states allow, be it dividers, et cetera, ways to drive the customer to other days and times that aren't at or over capacity, opportunity for outdoor dining in a greater way to increase that at all? Just some thoughts around what's in your control from here.

Thanks, Dennis. Certainly, the social distancing requirements and the dining room capacity requirements are a bit of a restraint. As they lift, however -- as an example, if Texas -- before they had an outbreak, they were looking to go from 50% to 75% capacity. We have plans and we're ready if we need to, to be able to put dividers, nicely, highly designed dividers in between booths, so we could easily get to that 75% capacity as fast as possible. And we're planning wherever we can to be able to jump on those opportunities as soon as that capacity is expanded.

I think I may have stated in the opening remarks that on our patios today, since they are large, we have a good seating capacity. And in conjunction with our landlords, in many, many restaurants, we've been able to expand that just a little bit even outside of our traditional patio and add another 1 to 4 stations that can be meaningful, certainly in some of those long waits that we had talked about.

So our operators are exceptionally nimble and able to be creative in ways that still align with the guest experience that we want. And we'll look for every opportunity to expand capacity but still with the main focus being ensuring that it's safe for every single guest and for the staff because we know in the long run, although you might get some short-term sales out of something, it has to be safe so guests feel like they can return and they know it's a safe environment.

And Dennis, this is Matt. I would just add onto that. We know that what's in our control to continue to drive some of the off-premise business. I mean the fact that we, with those open dining rooms, are keeping 90% of our elevated COVID off-premise sales, I think, is a testament to how -- what we've always talked about, the quality and the value of Cheesecake Factory off-premise. And I think guests are seeing that. And we have really been developing innovative new ways to market that, and I think we can continue to do it. So I think that also, that channel, will be something that we can continue to grow. And whether that's through our own online ordering or delivery or even guests still calling in the old-fashioned way, we're happy to do that.

Your next question comes from the line of John Ivankoe from JPMorgan.

I wanted to revisit the margin, the 17.5% margin comment that you made. I mean if that's flat year-over-year in that subset of restaurants, exactly how did you achieve that level of margins on a lower level of sales? In other words, if the menu wasn't changed, the customer hasn't seen any changes, there's not any change in the experience. I mean can you make some comments on some of the operational changes that you made within the store that you could potentially carry forward and even build on as we go forward as sales volumes rebuild across the entire chain?

Yes, John, this is Matt. I think the main driver that I would call out is that we just do much bigger volumes than our competitors. And so the flex or the flow-through, if you will, above the breakeven point for us is mostly variable, and we're able to adjust, right, because we're on percentage rent. At those levels of sales -- we're on percentage of rent, and we're able to move each of the line items in a way that's corresponding to the revenues. So there's not necessarily the magic other than the brand's volumes.

So I would say, to be fair, we do have the full management in those restaurants. There was a little bit of margin pressure there, but because we were bringing back all of the staff that had already worked for Cheesecake Factory, there was less training costs than year-over-year. But those are small pieces. For the most part, everything lined up pretty well. So that's why we talk about not wanting to take away from the brand because the sales driving piece is what really enables that. I think we got a little bit smarter at scheduling and some of the smaller pieces of the P&L that helped to fill in the gaps, but it's really flexing appropriately across every line item.

And were things like R&M and marketing and maybe even some of the fixed rent -- I mean -- or I guess the percentage rent in this case. I mean was there anything that was worth like a couple of hundred basis points here and there, I guess, specifically focusing on the R&M piece, on the marketing piece?

Nothing was that big. They were all in the 50 basis points, 25 basis points that we're trading off.

Your next question comes from the line of Gregory Francfort from Bank of America.

My understanding is part of what's given Cheesecake the ability in the past to have the volumes that it has is a bigger business in the shoulder periods. Can you maybe talk about how that's recovered as dining rooms have opened up and if you're comping sort of at similar level of the shoulder periods as you are at the rest of the year maybe dinner or the lunch dayparts?

Sure. Greg, this is Matt. It remarkably was similar, particularly in lunch and dinner. But obviously, the part that I think made up the difference really for us was that the midafternoon was stronger, and a little bit of that came at the expenses of -- because it's 100%, right? If you think about it, it's got to add up to 100%. Late night was a little bit less, and that was to be expected. We did have some shorter hours in there and things like that.

But the midafternoon business definitely picked up, and some of that was by design. And I think that, that goes to one of the things we've been talking about for a while, and that was through the off-premise channel and our ability to specifically drive day times and days of the week. And so we were really optimizing capacity in some of those situations, right, whether maybe we ran a promotion for lunchtime only, and that ran through 5 o'clock, and we picked up sales at 4 and 5 o'clock that we might not have otherwise. So we felt really good, particularly about the midafternoon in addition to keeping lunch and dinner where it was.

Your next question comes from the line of Jeff Farmer from Gordon Haskett.

As Cheesecake has reopened restaurants in malls that have only recently reopened themselves, I'm just curious what the relationship is between that mall traffic you're seeing in these recently reopened malls and your own customer traffic.

Actually, our traffic has remained pretty consistent whether or not a mall is fully reopened or not reopened. There were some situations throughout the quarter where we were able to reopen and maybe the mall hadn't opened yet or vice versa. But the sales throughout all of the restaurants, mall, nonmall, open, not open, are consistent. So -- and some of the malls that have fewer stores that have opened, our sales have remained very, very, very strong and very similar to the malls where a few more stores may be open and may have higher traffic.

That's helpful. And just as a quick follow-up. When the stimulus checks hit in April and May, what impact were you able to discern on demand for your own business?

I think it was hard to say. I mean to be honest, there was so many factors that were going on right then. And I do think that there seemed like there was like a week or 2 with a little bit of a blip up, but pretty much, the trajectory was relatively consistent. I wouldn't say it was immaterial, but it was not a big driver.

Your next question comes from the line of Jon Tower from Wells Fargo.

Just a couple of quick ones for me. So historically, the company has not really done much by way of promotions. And it sounds like this 4th of July burger offer resonated pretty well and drove some decent sales. So first, I was curious, one, if you'd quantify how much demand that drove during the period. But two, does it also have you rethinking your promotional activity for the company going forward? And then I have one more follow-up.

Jon, this is David. I don't think it has us rethinking our promotional activity. I think that, all along, we thought about doing marketing in ways that align with what's right for the brand. And this particular promotion happened to really, really resonate and was really successful, drove some of the midafternoon that particular week that Matt was talking about. But it doesn't change, I think, our overall outlook on how we'll use marketing moving forward. I think it shows us that, especially in the off-premise channel, there's an opportunity to drive the value and everything the Cheesecake Factory experience can bring in off premise. So we'll continue to do that, but it doesn't meaningfully change our overall outlook on marketing.

Jon, this is Matt. If you didn't infer, we're not going to quantify that for competitive reasons.

Yes, I understand that. In terms -- I was just thinking more around the frequency of it versus what we've seen in the past, which has really been very much focused on Cheesecake offering, not necessarily food items.

Well, I think the off-premise channel is different, right? And I think that -- with the elevated levels of it, I think there may be opportunities to do more there than we would have ever thought with the on-premise side of it. So that could be true.

And I'll say, just even during this time, a good amount of what we've been promoting has still been cheesecake. So this happened to be a one-off promotion, but most of what we've been doing has been around cheesecake and/or gift cards.

Okay. And then the follow-up I have is just in the stores that have reopened for indoor in-restaurant dining, I'm curious to hear what the company is seeing with respect to restaurants that are colocated in, say, the malls or the lifestyle centers that you're in. Are they fully reopening, meaning doing the same in-store dining that you're doing? Can you quantify perhaps how many have closed that you've noted or anything? Any qualifications around that would be great.

I don't think that we could quantify any more that have closed than what you're going to read elsewhere. I think there was recently -- was it OpenTable or Yelp that came out with a particular report around reservations. That's probably where you're going to be able to get a little bit more data than anecdotally from us.

I think that as far as restaurants reopening, when it comes to the larger chains, I think that they're opening more or less in concert with us, if not faster in the beginning. And obviously, I think we have talked about a lot of people were opening as fast as they possibly could. And we were really being very prudent about how we opened and making sure that, number one, it was safe for our guests and staff. But overall, I think that most of our neighbors, especially if they're larger chains, are opening and generally hitting the same capacity restrictions that we are.

Your next question comes from the line of Jeffrey Bernstein from Barclays.

One clarification and then a question. Just on the stats you gave with the stores that, I guess, were at 50% indoor seating, the 33 units. You're saying if they're doing a down 13% comp and they're hitting a 17.5% restaurant margin, obviously, that's quite impressive. I'm just wondering if you can talk about maybe the relationship going forward between comp and margin as we model from here. Or maybe said a different way, what would be the margin, for example, in 3Q if the comps for the system stayed at the down 32%? It just seems like you're hitting margins with significantly reduced comps, so it seems like a meaningful opportunity. But just trying to figure out the relationship between those 2 kind of as a rule of thumb. And then I had one follow-up.

Jeff, it's Matt. I think, again, just to reiterate what -- where we're at from a cash flow perspective, which you can -- kind of backing the margins on, from current sales level at a company-wide perspective, we're breaking even on a cash flow basis. So I think as you model through that, you'll kind of figure out where the margins are. And certainly, the ability of us to go from there to -- which is a negative of 32%, to a negative 13%, you can kind of correlate the flow-through. I mean we typically target between 20% and 40% flow-through depending on the situation. And as we add back sales, we will recapture at those levels.

Okay. And then separately, just on the -- below the restaurant level on the G&A line, specifically, I'm just wondering, as you look into the back half or what you've experienced over the past few months, how much of that is core spend? Or maybe what has been reduced most aggressively? How should we think about the G&A spend as we look forward based on kind of current sales levels?

Yes. I mean I think our objective is to align the business as best as possible but also realize that there's a lot of uncertainty and hopeful, a little bit of optimism that things stabilize in the first half of next year. And so we'll try to balance being cost conscious and keeping the resources that we need. I think that the total G&A that we saw in Q2 has some noise in it just because of timing of different expenses that we weren't able to cut as fast, et cetera. But I think it's a good proxy for the absolute dollars for the back half of the year.

Your next question comes from the line of Matthew DiFrisco from Guggenheim.

I just have a 2-part question here. With respect to sort of looking ahead and the brands you acquired, North Italia, obviously, was sort of called out and has been bifurcated, broken out from the other Fox brands, seeming to be the one that you wanted to grow. COVID's hit. The North Italia brand obviously has a little bit more or leans a little bit more up against retail traffic or dependent on retail traffic, it appears. How does COVID and a post-COVID world maybe change your view of the portfolio and the brands that grow?

I ask because I see Flower Child, I think, if I'm not mistaken, you moved it into a different category in with Grand Lux and away from the Fox Restaurant Concepts that you had it previously in with. Does that sort of represent maybe the Flower Child brand in a post-COVID world could be a higher priority for growth?

So Matt, this is Matt Clark. First of all, I don't think it's changed our long-term outlook on any of them. I think we're still very bullish across the portfolio. North's performance has been really pretty similar to Cheesecake Factory, and comps are running very similarly, and margins are running similarly. So we feel like, look, that's a good indicator of the strength of the brand, and that's with limited capacity. And so I think that the returns there are going to still be strong, and it will still continue to grow.

The segment reporting is just really an accounting piece. Flower Child has probably been outperforming a little bit relatively because it's a fast casual and obviously had a bigger piece of off-premise going forward, and we want to continue to grow that brand as well. So I wouldn't read anything into where it was placed, but we really are still supportive of growing all of those brands. I think, in fact, we have a couple of Flower Child openings coming up potentially depending on the market conditions here in the back half of the third quarter, and we continue to move forward with some of the North locations. So it's all going forward.

And Matt, just to clarify, the timing of the segment change, that happened in concert with our 10-K filing, so it wasn't this quarter. We were just trying to provide a little bit more information in the press release to provide clarity around the segment reporting because sometimes it's missed in the SEC filings.

Okay. That's very helpful. And then I guess, just did you say Flower Child was positive or is positive right now?

No. I said it was running a little bit better than the other brands. It's still down, but obviously, it had more off-premise business to begin with. And so it's been a little bit higher than the other brands.

Your next question comes from the line of Lauren Silberman from Crédit Suisse.

Assuming I heard correctly, you're maintaining an impressive 90% of elevated off-premise volumes in the restaurants that are open for dine-ins. So can you share the demographics of the off-premise versus on-premise customer, the overlap between the occasion? And any commentary regarding new customers you brought into the system over the last several months for both to-go and delivery?

So Lauren, this is Matt. It's an interesting question. We don't typically do that type of data research, but we have captured some of it over time. And really, the guest demographic for dine-in and off-premise is remarkably similar. It's just a different occasion.

So we tend to skew slightly more educated, slightly more affluent, slightly more tech savvy. And those attributes are all similar across each of the channels that our guests are using. And I think one of the things that we've tried to do is move more to the online ordering piece of it, certainly, and that's been working. And so I think there's affinity for all of the different types of ways that our guests can reach the Cheesecake Factory.

Great. And then any thoughts on how you're thinking about that run rate mix of on premise versus off premise in a post-COVID environment?

Yes. It's tricky to know. I mean I think that's the, for us, $4.2 million question, right? And so we believe that it will be sticky. I think that we were pleasantly surprised at the levels that we're maintaining right now. But it's -- I think it's very difficult to know where exactly that will land. I think that you will see that guest behavior does get permanently modified after periods of 3 to 6 months. So I think that, that research has played that out. So it's going to be elevated. It's just a matter of where it lands.

Your next question comes from the line of James Rutherford from Stephens Inc.

My question is really a follow-up on the last couple of questions, and it's around the off-premise business. I think last quarter, you gave a stat, correct me if I'm wrong, but a stat that over 1/3 of delivery orders were from new customers. Is there any way you can quantify or, at least, sort of directionally understand your ability to retain those customers? And I ask this -- that speaks to the brand's ability to come out of the pandemic with higher unit volumes than before. So just, is there a way to quantify that ability to retain those new customers?

I don't think we stated that 1/3 of the customers were new customers. Oh, and last quarter -- we may have said that last quarter. So we get that data through DoorDash. All the new customer acquisition data comes through DoorDash. And they share with us whether or not we have higher repeat rates. I think last quarter, we did state that we saw some really good repeat order rates from first-time DoorDash guests. Some of the promotions we have run throughout this time in off premise have been free delivery if you've never ordered from Cheesecake Factory before. So we're able to track those first-time guests and track their behavior, and we have seen positive return rates from those specific guests.

I would also say that we have seen that customer order frequency and the average customer order value both increased during Q2. So those guests that were coming were coming back more frequently and spending more money. So I think those are both good leading indicators.

Your next question comes from the line of Brian Vaccaro from Raymond James.

Just a few for me. On the international side of the business, could you provide some color on how many units were temporarily closed during the quarter, where does that stand currently and how sales were performing when you reopened? And also, any kind of form of reliefs that you're providing to your international partners?

Sure, Brian. So all of our partners are sort of up against -- or were up against when it comes to social distancing regulations and opening and closing throughout the quarter.

Right now -- as of right now, I believe, all the restaurants are open. Some of them may have restricted capacity. Some may be on a delivery-only model. The majority have dining rooms open and have seen relatively consistent sales. I'd say that the Asia locations are a little bit ahead of the other markets. We've seen in China some volumes recently recovering with comps trending negative 10%. So that's a really good positive sign. The Middle East are a little bit more up and down depending on what the capacity restrictions are. And probably in Mexico, the recovery is a little bit behind the U.S., but all those restaurants are open in Mexico as well.

And for Q2 -- Brian, this is Matt. The preponderance of the sites and partners and through each of the 3 periods were provided relief given the circumstances.

Okay. And then also, the 19 units that remained closed -- I think you said it was 19 earlier in the call. How are you thinking about the time line on reopening those units? And how many might be permanently closed?

It's actually 16 that are currently closed. And a good slate of them are set to open here in the next couple of weeks. And then we'll assess the performance of the others and determine what we would do moving forward.

Great. And then one bookkeeping for you, Matt. The $11.7 million of COVID cost, could you help us allocate that between the cost lines, COGS, labor and other OpEx?

Yes, it's about 60-20-20, with the 60% being in labor.

(Operator Instructions) And there are no further questions at this time.

Ladies and gentlemen, this does conclude today's conference call. Thank you for participating. You may now disconnect.