A Pending Restaurant IPO and AI Buzz

In this podcast, Motley Fool senior analyst Jason Moser discusses:

• The growth of Cava and the company's plans for the capital raised during its proposed IPO.
• How Cava could end up as the next Chipotle (or not).
• What another big week of retail earnings portends. 

Plus, a lot of companies are talking about artificial intelligence (AI) this earnings season. Motley Fool producer Ricky Mulvey chats with Motley Fool analyst Meilin Quinn about a business that may be getting a genuine tailwind from AI.

To catch full episodes of all The Motley Fool's free podcasts, check out our podcast center. To get started investing, check out our quick-start guide to investing in stocks. A full transcript follows the video.

This video was recorded on May 22, 2023.

Chris Hill: One company is walking the walk on AI, and another company is getting ready for its debut in the public markets. Motley Fool Money starts now. I'm Chris Hill, joining me in studio Motley Fool senior analyst Jason Moser. Happy Monday.

Jason Moser: Howdy.

Chris Hill: I'm excited about a potential IPO. Potential, I mean barring something unforeseen. Cava, the Mediterranean restaurant chain, will be going public later this year. They have filed the requisite paperwork. I think we talked about this on MarketFoolery five years ago because Cava is a private company. But in 2018, they bought Zoes Kitchen, which was a publicly traded chain that was struggling, and $300 million is the price tag I remember. I think we said at the time, it seems like they're basically buying the real estate. They're going to turn all of these Zoe's Kitchen into Cavas, which they have methodically done to the extent that you've had the chance to look over the S-1 filing. What stands out to you about this? I'm excited about it because it's a restaurant. I think that's how my animal brain works.

Jason Moser: What would you say you frequent more? Cava or Chipotle?

Chris Hill: Chipotle. But it's a location thing for me. If there was Cava closer to me than it currently exists, I'd be going there more often.

Jason Moser: I live there right next door to each other almost. I would say I personally favor Chipotle a little bit more, but Cava's right there.

Chris Hill: It's the same restaurant concept. It's just thinking for those who are what's Cava? I've never been there. It's Chipotle with Mediterranean food.

Jason Moser: Exactly, and much like I've learned how to make my own Chipotle burrito bowls at home, I've done very much the same thing with Cava. From that perspective, they're onto something. They make good food. You're right. The acquisition of Zoe's back in 2018, it get gave them ultimately, that intention the portfolio of real estate. It wasn't to roll in another restaurant concept and become a Zoe. This is ultimately been a multiyearlong transition to basically rid the market of Zoe's and make them all Cavas, and so far it's working OK. Cava just opened up their 263rd restaurant. That's along with making all of these Zoe's transitions along the way. They should have 300 Cavas by the end of the year. 

When you look at it, compared to something like Chipotle it's still a much smaller concept today, which for investors I think that's encouraging. One of the big problems you see with so many of these IPOs. But these companies like Uber and Airbnbs are going public and they're already so big. You have to question how much growth is actually there. In the case of Cava, it's going public. It will still be relatively small. Good to know that they intend to use the proceeds from the offering for new restaurant openings and general corporate purposes. This doesn't sound like it's to pay off someone else. But the company itself I wouldn't say I'm all in, but I definitely, I'm interested to learn more. Reading through this S-1, the business itself has grown very quickly, the Cava revenue, and just excluding the Zoe's for a second here, but Cava revenue has grown from 41.2 million in fiscal 2016 to 448.6 million in fiscal 2022. That's just under 50% CAGR there. Digital revenue, around 35% of sales, which is really strong. You look at the stores themselves, very productive, $2.4 million in average unit volume in 2022. You compare that to something like Chipotle, which their goal is try to push beyond 3 million this year or next. A lot of similarities there, but clearly Cava is much smaller. It's still just getting the growth engine going.

Chris Hill: In the interest of full disclosure, I think that this is a local business, Brett Schulman, who is the co-founder and CEO, is here in the greater D.C. area. He has the experience of running this business for the past 13 years. He does not have the experience of running a public company. But I believe Ron Shaich's group made an investment in Cava Group to help fund that Zoe's acquisition. I think I have that right and Ron Shaich who ran Panera all those years. I'm assuming he is advising and helping Schulman as he prepares to be a public company CEO. You're right. It seems a little bit like a Goldilocks S-1. Do you want to look for the good? Do you want to look for the bad? They've been operating at a loss since 2016. But as you said, they appear to planning to use the money in the right way. There is this opportunity, and the question on my mind is, is this the next Chipotle or is this the next Sweetgreen? Because one of those is a much better investment than the other. 

Jason Moser: Yes, you are right. So far that is the case. I'm going to have to defer my answer here. I don't know. I am hopeful that this is more Chipotle than not. You're looking at some comparables. Again, you look at these two businesses side-by-side to see where one is and where one could go one day. Looking at loyalty members, for example, Cava as of April of this year they had approximately 3.7 million. Chipotle has got 33 million. Again, I had mentioned the sales per store year-in and year-out. But looking at the market opportunity itself you see Chipotle has something better than 3,000 stores today versus Cava hoping to have 300 by the end of the year. In the longer-term, they see that there's a market for 1,000 or more stores by 2032. And so I think if you're looking at it from the investing perspective today, that's a fair timeline there. Let's look at this thing out over the next decade and see where it could go. If they get to 1,000 stores by then, it's certainly possible. So far more competitive landscape today than it was back when Chipotle came to market. But you have 1,000 stores and maybe they can bring in two-and-a-half million dollars in revenue per store. You got a business that's bringing in two-and-a-half billion dollars in revenue, and you compare that to the margins of Chipotle. Maybe it's bringing in 250,000 or $250 million in net income by that point. Nothing to sneeze at there, but it would definitely take some time to get there.

Chris Hill: One more interesting wrinkle about this business that I think investors will want to watch is they have a consumer product.

Jason Moser: I am glad you brought that up.

Chris Hill: Unlike Chipotle, unlike Sweetgreen, there are hundreds of locations across America that sell Cava dips, hummus, all that thing. It'll be interesting to watch that part of the business to the extent that they can grow, not just the footprint in terms of locations, but maybe the product suite as well.

Jason Moser: I'm glad you brought that up because that really does play into their growth strategy. The growth strategy is to open up stores in existing markets, to open up stores in new markets, and then to try to grow the CPG business. Now, is it elite? I don't know. We'll find out in time. But we've seen other companies do a pretty good job with this. You look at something like a Panera or Starbucks that have done pretty well getting that CPG business out there. With Cava today, there are dips, spreads, dressings. I think they sell in more than 650 grocery stores nationwide, including Whole Foods Markets. Now, Whole Foods as we know, owned by Amazon. You just never know, there can be interesting tie-ups there in regard to distribution that can help take this thing to another level.

But primarily the growth is Cava in that CPG segment. It doesn't mean that management can pursue other opportunities later. But remember what happened with Chipotle, whether it was ShopHouse or whether it's Pizzeria Locale. All of these different ideas of being able to take throughput format into different concepts. We saw how that worked out with them. If we saw Cava's management indicating they wanted to go that direction, I feel like it's fool me once, shame on you, but fool me twice. I think I'll look at that one a little bit more closely and maybe not give them as much credit up front for that if they decided to go that way.

Chris Hill: It is another big week for retail earnings. Last week we had Walmart, Target, Home Depot. This week we've got Costco, Lowes, Best Buy, Gap. I was saying to you this morning, last week, not awesome [laughs]. Not necessarily terrible, but not awesome. But I'm curious if you think there's any reason to think what we're going to see this week is dramatically different, either to the upside or the downside. I'll just throw out there that the commentary that we got last week around consumer discretionary spending and big-ticket items, particularly from Home Depot really makes me worried for Best Buy.

Jason Moser: That's what they do. I don't think many people are going to Best Buy for DVDs anymore, are they? Not for a while now, not now. I don't expect really any surprises. I think we're probably going to get information that more or less reinforces what we just learned last week. Ultimately my takeaway from last week, while it wasn't anything operations out there really lit the world on fire. I mean, we're certainly seeing this move from products to services, but what we're also seeing, and I think this is a light at the end of the tunnel here is we're starting to see inventory levels get back to normal. We're starting to see retail inventory levels come back down. Some examples, Target's inventory down 16%, TJX down 8%, Walmart down 9.4%. Ross Stores down 16%, Home Depot was even flat. But ultimately what that means is, we've been talking about this excess of inventory of the past year plus these companies having to really resort to deals in order to just move this stuff. The longer it sits on the shelves and absolutely it becomes a real challenge. It's encouraging to see these inventory levels coming back down. Maybe things are starting to normalize and we're working our way through this COVID hangover.

Chris Hill: I said this to Bill Mann the other day and I'm saying to you now this is not my last episode of Motley Fool Money. It's not your last, but it is our last together. I just wanted to thank you again for being so game to do the show so many times. I think we've been in the studio together more than anyone else and I really appreciate that so thank you for that.

Jason Moser: Well, I could go on and I'm not gonna because they'd probably get a little bit teary. But you're right. I think we probably have been at this together more than anyone here and that was just because of the good fortune when I got here to just fall us backwards into this concept that you and Mark had come up with back when it was just MarketFoolery and it was not in the job description when I was hired. I can't tell you how grateful I am to have these past 13-plus years with you because there's been a lot of fun. You have made me a better investor, but really, at the end of the day, you're just a great friend, and I'm going to miss you tremendously. 

Chris Hill: Jason Moser, thanks for being here.

Jason Moser: Thank you.

Chris Hill: If there's one theme that has emerged on conference calls this earnings season, it's the number of times CEOs have used some version of the phrase "artificial intelligence." But one company isn't just talking the talk and they may have a real tailwind from generative AI. Ricky Mulvey has more. 

Ricky Mulvey: Generative AI has some exciting possibilities, but some companies are going to take a hit as this change shakes out. Joining us now to talk about some of the companies who may be feeling that risk from those chatbots is Motley Fool analyst Meilin Quinn, welcome.

Meilin Quinn: Hi Ricky, great to be here.

Ricky Mulvey: Great to see you.

Meilin Quinn: Great to see you.

Ricky Mulvey: It's really tough, I think to guess what the ripple effects of generative AI are going to be, reminds me of those old newspaper cartoons when they would envision the future, you would see helicopter spinners on top of cars and elevators up to Mars. But as we think about the future of generative AI, what are the signs to you that a company could face a real threat from those chatbots?

Meilin Quinn: A company could face real threats from ChatGPT or other language models if their business model is exposed to essentially AI's language modeling skills. Some of these professions include copywriters, tutors, copy editors, consultants, or if a business is exposed to AI's image and video generation skills, such as graphic designers and artists and even voice narration. There are now, AI tools that let you upload an audio sample of anyone you want. It could be Ariana Grande and then ask it to narrate text or sing and it will sound just like Ariana Grande. You can have AI read anything in someone's voice. You'll notice that many of the tasks that I had mentioned, are tasks that people would typically use websites like Fiverr and Upwork for. These are two companies that definitely are at risk from the proliferation and the advancements in AI. But Fiverr and Upwork, they have a pretty diverse set of categories and tasks that they facilitate on their platform. I don't think AI is an existential risk to these two companies just yet. There are other companies, however, like Chegg, the homework help platform, whose business model relies solely on a pretty straightforward task, which is explaining the answers to homework problems. This is something that AI can already do well and can do for free. I do not have high hopes for Chegg's ability to keep up with this trend without it completely reinventing its business model.

Ricky Mulvey: Yeah, Chegg, the education tech company, basically you can plug it in homework problems and they'll give you detailed solutions. In its latest quarter, CEO Dan Rosensweig said that ChatGPT was impacting customer acquisition but not necessarily retention so the retention numbers were still strong. They can still keep their customers and it's going to introduce something called CheggMate, which they are describing as homework help with ChatGPT. Apparently, it's better to have their proprietary dataset with ChatGPT to help people with homework. But I don't know if that's enough of a moat or if that dataset can be easily replicated by OpenAI and what ChatGPT already has access to.

Meilin Quinn: Yeah, that's a very fair concern. I generally agree and I actually used Chegg in college. I'll explain how it works quickly for members who might be unfamiliar. Basically, it's used by students and they pay a monthly membership. Right now the highest here cost 20 bucks a month and that includes a Q&A area where you can ask questions to tutors, which on the platform they call experts and this membership also includes explanations for homework problems and popular textbooks. To put it bluntly, almost everything you can use Chegg for or you can use ChatGPT for. In fact, GPT is even nicer for homework because like I mentioned, it's free, but it's also unlimited. With Chegg, you're limited to 20 questions a month to ask these experts and 20 questions might sound like a lot, but I remember encountering this limit a lot in college and I remember students would try to work around it by asking multiple questions in one post and then the experts would get really annoyed and explain that you can only ask one question at a time, I think because they only get paid per post that they respond to you. But with ChatGPT is you can really ask it an unlimited amount of questions. You can ask it to dumb things down for you. 

You can ask for follow-ups. You can ask it to go into as much detail as you need to really understand the problem that you're trying to work through. If I were a student, it really wouldn't even be a question that I'd be using GPT instead of Chegg for homework help. There's something really funny that I want to share. I had actually asked a series of questions on Chegg about three years ago and in the past month, I've been getting notifications that these questions have been finally answered. They had been unanswered all these years, and they've just now been answered. I wonder if that means that perhaps there is not enough of an inflow of questions on the platform because students are going to straight to GPT. So maybe in order to get paid, experts are scrubbing the platform for old questions that have gone on answered, or maybe it's just that it's so easy, it's just a matter of copying and pasting into GPT a question that experts are going back and answering questions that might have been too complicated. 

Yeah, just a funny observation. I actually did go back to one of those questions that I had asked an a and put it into ChatGPT because I was curious and it gave me a pretty thorough explanation. You touched on this. I don't see their new customer growth improving much at all even with the addition of this new GPT solution. They have had pretty solid retention. I don't see them being able to sustain this as the current generation graduates from college and there's no new cohort to replace them. I don't think it will be able to pull this off and I'm not exactly sure what this mysterious proprietary data entails. Chegg hasn't exactly spelled it out and it hasn't explained why exactly this data will enable it to come out with something more powerful than ChatGPT. The way that I see it is that homework is homework. Unless it's a written prompt, there's usually only one answer to a homework question. It could be that Chegg is implying that it can come out with even more accurate answers than GPT with its data because GPT is known to sometimes spit out false information. ChatGPT is only getting better. If I were a student, I would still be inclined to start with ChatGPT.

Ricky Mulvey: It's got a churn problem because in most cases max life for a customer on Chegg is about four years. I'm always hesitant about full paradigm shifts. No one's ever going to go to the gym again would be an example or hotels shutting down during the pandemic. No one is ever going to get out of their home again. Chegg is down more than 90% from pandemic highs and it's facing slower revenue growth. It does have a high retention and it does solve a problem. It's also free cash flow positive and is trading for less than 10 times earnings. Is this a value story? Do you think this is a buying opportunity for long-term investors?

Meilin Quinn: I don't. The Warren Buffett quote comes to mind. That goes like, it's far better to buy a wonderful company at a fair price than a fair company at a wonderful price. Chegg is free-cash-flow-positive, but it has a high level of debt. It has $1.2 billion in debt, which is actually more than its market cap right now. I think it's going to have to make a lot of investments as it works out how it's going to get ahead of these advancements in AI. Plus it's going to have to pay OpenAI for API access for its CheggMate feature. I know that this won't be cheap. I think this will bring down its margins for a while. So I would rather hold off and I'd rather wait to see whether anything actually materializes in Chegg's plans for CheggMate.

Ricky Mulvey: When you think of the signs of a company that could experience more of a long-term tailwind from generative AI. Maybe the CEO is not just filling in the AI square on every analyst's bingo card right now. What are the signs of those companies that could be seeing those long-term tailwinds?

Meilin Quinn: If they're working to adapt and incorporate AI into their business in a way that makes sense and in a way that still gives it an unrivaled value proposition compared to ChatGPT. I actually think of Duolingo in this way, the language learning app. They're rolling out something called Duolingo Max, which is powered by GPT. This will couple Duolingo's unique proprietary data. This is data that Duolingo has actually explained. This data includes info on the best ways different categories of users learn and stay engaged on the platform. Which order to ask questions to different users in order to optimize their learning and keep them on the platform. This isn't data that GPT has. But even more exciting, Duolingo Max will allow users to have role-play conversations in either texts or speech with a tutor in the language that they're learning. The speech part I think is what's critical because this gives people a low-stakes way to practice speaking a language. I used to actually pay a tutor on Fiverr 10 bucks for 30 minutes. Super cheap, by the way, but I stopped doing that because I got so embarrassed. We would essentially have conversations in French, but I had to keep saying, can you repeat that? Or I would mingle in English words in my sentences because I didn't know how to say them in French. I just called so embarrassed and I felt that it wasn't exactly productive. I think a lot of beginners, even introverts, and even advanced language learners just want a speaking partner in their pockets. I can see a lot of people opting for Duolingo Max, especially as it is combined with Duolingo's data on how people optimally learn languages. Especially when Duolingo makes this feature available for more languages.

Ricky Mulvey: It's also a completely different service, right? You're looking for a singular text-based solution with Chegg in most cases.

Meilin Quinn: Yes.

Ricky Mulvey: With Duolingo it's more of that service where -- Google Translate's significantly better. You can even use Google Lens to translate signs around the world into your preferred language. Maybe there will be a decrease, but I think there always will be a demand for people to learn different languages to connect with others in a more human-to-human way.

Meilin Quinn: Absolutely. You know what, Duolingo does it so well. It's really got that gamified system down to a T. It's almost made it addictive to learn a language. There's a competitive aspect to it with their scoreboard and they're really good about those notifications that they send you. There are memes about it. They'll send a literal crying Duo bird, [laughs] something really sad like you're not leaving, are you? Come back and learn Spanish. But it works. Their marketing has worked really well for them. Duolingo's data to me makes a lot of sense. It's always been one of their moats. I'm excited to see them couple that with ChatGPT.

Ricky Mulvey: Meilin Quinn, I appreciate your time and your insights.

Meilin Quinn: Absolutely. Thank you so much, Ricky.

Chris Hill: As always, people on the program may have interest in the stocks they talk about and The Motley Fool may have formal recommendations for or against, so don't buy or sell stocks based solely on what you hear. I'm Chris Hill. Thanks for listening. We'll see you tomorrow.