Cava: One Brand, Many Channels

Plus! AI and Incumbents; Buying Reputation; Restaurant Pricing; Debt, Prices, and the Pull-Forward Effect; Assets and Distribution

Byrne Hobart Byrne Hobart

Cava: One Brand, Many Channels

It's been a slow year for IPOs. This list shows a handful of SPACs—lots of offerings whose size seems to indicate that they're targeting retail investors rather than institutions, and the odd biotech or spinoff. But the IPO window is less a function of absolute valuations and more of trailing six-month returns: once it doesn't feel like meaningfully terrible timing to sell stocks, people will start doing it again. Coming up soon we'll have a test case for this: Mediterranean-style fast casual chain Cava is going public soon (S-1 here).

Cava's S-1 opens with a visual acknowledgement that Instagram has replaced television as the main way people discover new food brands, at least for midprice-and-above quick-service restaurants. Commercials can do a lot to jazz up a burger. But restaurants have also recognized that part of what they're doing when they plate a dish is that they're helping with the composition of a photo; brighter-red tomatoes and vividly green spinach leaves are part of the cost of goods sold in an accounting sense, but economically they're all marketing.

Most retailers have come through the pandemic a lot less focused on single channels than they were when they started. App-based ordering has made brick-and-mortar at least partly digital, and also means that pure-play e-commerce companies face competition on convenience and lack-of-friction; waiting for home delivery can be less convenient and more uncertain than just swinging by a store to do a curbside pickup as part of some other errand, for example.

Cava was multi-channel almost from the beginning: they were founded as a full-service restaurant, started selling some of their ingredients through grocery stores in 2008, and hired a new CEO to accelerate that process. In 2011, they expanded into a fast casual format, which is their main style today. In 2018, they bought out Zoe's Kitchen, which had 263 units at the time of the sale; the Cava brand's total restaurant count now is... also 263, as they've been opening up new branded restaurants and slowly converting Zoe's locations into Cava stores.

Like many organizations, they've tried out several models and found the one that works best for them: last year, restaurant sales were 98.7% of revenue, with grocery store sales of their hummus and sauces making up the rest. But that business does have some perks: getting their products at eye level in Whole Foods means running an ad for their restaurant to exactly Cava's target demographic, so the wholesale business is still worth running even if it's not necessarily profitable.[1]

There are many habits a tech-focused investor has to break when looking at other industries, but conveniently for them, restaurant concept IPOs do follow at least one broad rule of tech investing: at IPO time, the company will typically exhibit impressive growth, good unit economics, a compelling story—and persistent losses.

And then you get to the dreary part we’ve all been waiting for: Cava isn’t profitable, and their net margins worsened in 2022 to -10.5% from -7.5% in 2021. In fairness, losses peaked in 2020, and 2019's numbers were also much worse (a -27% margin). And in the last quarter, their loss compressed to a -1% margin. So they're on track to celebrate a profitable quarter soon after IPO, at least as long as they don't insist on stepping up growth too much.

At one level, restaurants are straightforward to analyze; decades ago, Peter Lynch rhapsodized about how obvious it was that if Taco Bell could saturate California, it could go nationwide, and that Dunkin could run a similar expansion campaign from the opposite coast. Taste buds are pretty similar across the country, and once a chain reaches sufficient scale, national advertising, and more recently app penetration, will make the rest of the growth happen faster. On the other hand, Zoe's got acquired at about a third of its IPO valuation. Execution is hard, and a not-uncommon story is that chains expand their store count a bit faster than they learn how to expand their roster of experienced and competent managers; it's entirely possible that the most valuable publicly-traded for-profit education play is whatever share of McDonalds' value comes from Hamburger University. Still, it's hard not to see the Cava IPO as a promising sign: even in centuries-old industries, there are new ideas. And even with the market off its highs and the macro situation uncertain, someone out there is willing to drop on S-1 and raise some funds from public markets to accelerate their growth.

  1. This has interesting implications for consumer packaged goods companies and grocery stores more generally. Grocers have been improving their data collection over time, in part through better collection and in part through M&A ($), and they're increasingly selling white-label products instead of external brands. If there are external brands that can tolerate low margins and that have an incentive to market their brand name in order to benefit the restaurant piece of their business, that's one more source of pressure for the pure consumer packaged goods companies. No wonder they're looking at alternative distribution models—perhaps share-of-counter will be the next share-of-shelf-space. ↩︎

  2. For the detail-oriented: Zoe's had 263 locations, they've converted 145 and have eight more in progress, so they've closed a total of 110. If we make the arbitrary assumption that these Zoe's stores had performance in line with the Zoe's corporate average—$1.2m annualized over the last period where Zoe's was public—and we assume the same same-store-sales cadence at Zoe's as at Cava, we get a ~$170m headwind from closing stores. Add that back and the adjusted annualized growth since 2019 is 15%. Of course, the stores that got shut down were probably underperforming, and given that Zoes' revenue per store was so much lower than Cava's, some of that same-store performance they've talked about probably comes from the converted stores ramping up. So 5% might be a lower bound while 15% is an upper bound on organic revenue growth since 2019. ↩︎

  3. In general the higher-end and concept-focused ones put up the best numbers, and more generic and lower-priced stores do worse. But it's hard to generalize: Chick-fil-a's revenue per store is apparently $6.1m. All of these numbers are via the fast food trade magazine's QSR 50 report. ↩︎

A Word From Our Sponsors

Hire a Superteam to Build Products Leveraging AI

The hard truth is: if your team isn’t already using AI, you’re probably falling behind.

Finding individuals proficient in AI is toughhh. That's why founders choose AE Studio.

Unlike most consultants, they know their stuff when it comes to creating machine learning solutions in business. Their product studio offers:

·    Software developers that build tech-heavy web & mobile apps.

·    PhDs that know how cutting-edge tech will solve your business problems.

·    Product designers that’ll make your product look way better (sorry)!

Here’s what one cofounder said about AE:

“Our app has received internal and external praise for its speed, efficiency, cleanliness of code, QA, and aesthetic. AE’s dedication and efficient project management has led our app to a 46% retention rate with consumers….”

Hire the world’s most effective team today at ae.studio.

Elsewhere

AI and Incumbents

Expense management company Ramp recently launched a sort of copilot for finance teams, automatically categorizing expenses and identifying potential savings. What's notable about this is who's doing it: Ramp has raised $1.4bn, with its most recent fundraise announced in March 2022. It's still a reasonably new company, but had a viable business before AI came along. More to the point, it had a lot of training data; an expense-categorization tool saves time because people had to take the time to categorize expenses in the first place, which turned out to be uniquely useful training data. It's entirely possible that AI will not only be a technology that mostly benefits incumbents, but that it could be one that disproportionately benefits companies that either have more legacy processes or are involved in them. It's Moravec's Paradox as an investment thesis: the most annoyingly low-tech tasks, so long as they involve symbol manipulation, are the ones that LLMs influence first, and years of annoying drudgery can be capitalized into new and useful models.

(Via Random Walk, which Diff readers—especially Diff readers who want more charts!—will enjoy.)

Buying Reputation

After hiring its new CEO, Linda Yaccarino, Twitter has been removed from media buyer GroupM's list of high-risk ad venues ($, FT). This kind of thing has happened before: when Zoom had some security issues in early 2020, they responded by hiring Alex Stamos, who had left two companies (Yahoo and Facebook) because their security policies weren't up to his standards. When a company has a reputational problem, it's partly a problem of uncertainty: even if there's, say, a 10% risk that they're as bad as the worst-case scenario implies, big buyers are not in the business of underwriting that risk. So one thing companies can do is essentially combine credible external due diligence with incentive alignment, by hiring someone who has a lot to gain if the risks are overblown and more to lose if they're not.

Disclosure: Long META.

Restaurant Pricing

The WSJ has a good profile of restaurant franchisee Flynn Restaurant Group ($), which has over 2,300 restaurants across multiple chains (including, but not limited to, Taco Bell, Pizza Hut, and Arby’s). Companies like this are where "inflation" as a broad phenomenon affecting labor and materials turns into a specific quantifiable macro variable, as they trade off between higher prices and lower volumes, and as they measure elasticities across different locations and brands. One thing that stands out is that the company spends a lot of effort tracking what prices competitors are charging. This, combined with the fact that such competitive intelligence pays off more at scale, could make inflation jumpier than it was before. If it takes hours rather than months for one company's pricing decision to be noticed and copied by their competitors, the result is that inflation happens a lot faster, but it’s at least more transient.

Debt, Prices, and the Pull-Forward Effect

US bond issuance so far in May is the highest for any full May since 2020 ($, FT), which itself was high because May 2020 was around when capital markets reopened and non-distressed deals started to go through. (For comparison, May 2020 had 18 IPOs, while January and February 2020 averaged 16.) This time, instead of the volume of offerings being pushed back by pandemic delays, they're actually being pulled forward by concerns about the debt ceiling. The overall market has not been pricing much of anything as if a US default is a material risk, but instead like a remote risk that's just a tad less remote than usual for the moment. But different risk-takers operate in different ways: a bond trader who loses money in a default scenario is losing it alongside many other traders, and probably has opportunities to make the losses back and then some in the subsequent volatility. A CFO who misses an opportunity to issue debt and then has to wait nerve-wracking weeks or months to do the offering again has messed up their One Job of ensuring that the company they work for doesn't get walloped by some market risk that an expert could have seen coming.

Assets and Distribution

Fidelity's S&P 500 index fund now has more assets than State Street's SPY ETF ($, FT), having been a fraction of its size a decade ago. Some of this is from slightly lower fees, but it's also a good example of convergence across industries: Fidelity is best-known for its actively-managed funds, but when investor preferences shift, it's easier for the company to capture that demand itself than to let a competitor have it. And thanks to their vast scale, while this is costly to Fidelity, their size means that they can cope better than a smaller company would. It's a common story across industries: a concept will often get discovered at a big company, implemented by a tiny one with nothing to lose, and, if it works, get adopted by everyone—or, at least, adopted by everyone who survives the new trend's resulting shakeout.

Diff Jobs

Companies in the Diff network are actively looking for talent. A sampling of current open roles:

Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.

Find a Role

If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.

Find Talent with Diff Jobs