At Isla, a fine-dining restaurant in Columbus, Ohio, chef Andrew Smith spent early September fielding messages from diners asking whether he’d raised prices. He had not. But after an update to his reservation system, Tock, which Resy and American Express bought in June 2024, the figure many diners now saw on the platform included the 20% gratuity Isla adds to every bill. It was a small change that had a big impact, he said.

That change meant the listed price of a meal at Isla was significantly higher than regular diners were used to seeing — even though they’d actually paid the same amount in the past, post-gratuity. Reservations fell dramatically.

When Dawson asked Tock parent American Express about the change, a rep told her, “As price transparency requirements continue to evolve, Tock displays all-in pricing to provide greater clarity for guests. The total price may include the cost of the meal along with all mandatory fees charged by the restaurant, giving guests a clearer view of the total cost when making a reservation.”

In other words, a change to increase clarity for the end user by including end-of-meal fees made diners errantly believe that the restaurant had raised prices.

The drop in business was so pronounced that Smith decided to switch from all-in pricing to a $50 cancellation fee. Isla’s reservation numbers rebounded.

This very specific problem affects a subset of restaurants asking for prepayment at the time of booking, and there’s a (good) argument that a kind of all-in pricing, indicative of how much the diner should expect to pay based on the experiences they’re planning to receive.

That so-called “all-in pricing” came via regulation to the hotel and live events industries, where companies were prohibited by policy to hide required fees until the end of the transaction. Restaurants were a kind of broad exception to such regulations, which also targeted “junk fees,” a term the Biden administration used to garner support for protective legislation and regulation.

Restaurants are not subject to the same laws that now govern things like concert tickets or hotels, but it’s an admitted gray area — a prepaid dinner at a restaurant is an event, too.It’s also an example of the kind of “small” tech tweak that — even if made with good intentions prioritizing customer experience — can cause challenges for restaurants that rely on platforms for discovery, booking, and payments. As of today, Isla is still using Tock.

Full story here.

(Thanks to Gloria Dawson at In the Weeds for sharing her story and reporting with me! Subscribe to her newsletter for more stories about independent restaurants in America.)

ICYMI

I’m late to sharing this news, but my hometown paper, the San Francisco Chronicle, brought back stars to its reviews.

I truly have no opinion on the stars vs. no-stars cycle; I am not a critic and I generally don’t choose a restaurant based on reviews. But many, many people do, and I’m not sure they all recovered from the shock of major outlets dropping star ratings (mostly during the pandemic). Now, the SF Chronicle says, stars are back.

Read more

I spent Monday in New York, moderating the last panel of the day at a small industry conference. It’s a challenging timeslot for audience attention but an ideal time for wrapping a ton of the day’s insights into a convivial conversation. Our position worked to my advantage; the day’s speakers, including leaders from groups like Union Square Hospitality Group and Starr Restaurants and Tao Group and JKS restaurants and a handful of excellent independents, covered more tech topics than I expected, and my group got to focus on some of the biggest challenges — and opportunities! — ahead.

From my notes:

We have to do something about personal AI agents causing chaos… and poor experiences.

Read more

On Wednesday, Square unveiled a new kiosk. Like all of the company’s hardware, it’s thoughtfully designed and technologically sound — remarkably hospitable for a machine, blending touch screens, voice, evocative imagery and, of course, artificial intelligence.

“We wanted it to feel more friendly and more approachable,” Thomas Templeton, head of hardware for Square parent Block, told me in an interview.

It seems like Square designed it to be just as friendly and approachable as — maybe even more friendly and approachable than — a human staffer. It’ll to talk to you! It wants to answer your questions!

A little extra:

D magazine accused Michelin of using AI to write its Dallas guide in an very specific analysis.

Brian Reinhart, the dining critic at Dallas’s D Magazine, accused Michelin’s writers of using AI in its newest Dallas reviews. He reports it with alarming certainty.

“An analysis by D Magazine, including consultation with detection software and academic experts, shows that most or all of Michelin’s new Dallas entries were written by large language model (LLM) software, commonly known as AI,” Reinhart writes. (One of the experts he consulted is his brother, Alex Reinhart, a Carnegie Mellon professor who’s written about differences in human- and AI-generated copy.)

Michelin, with its team of inspectors that can make, break, or otherwise popularize a restaurant with just one review, denies the accusations.

“All Inspector notes appearing on the Michelin Guide website and app are developed collaboratively by the Inspection team based on their firsthand dining experiences. AI is not used in the writing of these reviews,” it said in a statement.

Reinhart defended his reporting and analysis on Reddit, where commenters quickly pointed out that AI-detection software can make mistakes. But Reinhart claims his analysis was not only able to detect the AI, it could also detect the AI model used. “Almost every new sentence was written by ChatGPT-4o or another AI platform built on 4o,” he wrote.

I trust a professional writer’s intuition — I like to think I have good gut instinct, too — but I worry we’re going to get stuck on the slippery slope of these increasingly hard-to-prove accusations. (When I expressed this to a fellow food journalist, they validated my position. “I’m not dying to get sued into the ground!” they messaged. Same, same.)

Also concerning: Will anyone care?

“If we’re featured, I’m honored,” one restaurant owner told Reinhart. “How they write, I have no comment.”

The move comes just a few months after Tucker’s departure from American Express. Tock, the reservations platform that served some of the world’s most iconic restaurants, will soon be absorbed into sister service Resy. Tucker will step away from consumer-facing restaurant tech (and its associated drama) to take on what might be an even messier challenge.

“I’m lucky to stay in the hospitality industry, but wholesale alcohol ordering is completely different. I think there’s a real product-market fit and right to win,” he told me. Plus, he said, “We’re talking about a $170 billion industry. So there’s lots to be done.”

Provi is a b2b marketplace where bars, restaurants, and retailers can order wholesale alcohol from their distributors in one place. The company doesn’t sell alcohol, it instead routes orders to licensed distributors in order to comply with the highly regulated American market. Restaurants and retailers can use it to research products; brands can advertise to a captive audience. Provi’s national network includes 1,400 distributor partners, close to 50,000 active retailers, and over 1.4 million products. The company is backed by investors including including Bessemer Venture Partners, D1 Capital, and Tiger Global.

While Tucker’s tenure at Tock (and his previous role as an executive at online ordering giant Olo) fell squarely into my purview, I’m less acquainted with the particulars of the beverage business, an exceptional and fragmented industry thanks to a tangled web of regulations. At its core: a three-tier system that’s been in place for nearly 100 years since the repeal of prohibition in the 1930s: Producers, suppliers, and importers on the bottom, licensed distributors and wholesalers in the middle, and retailers — wine stores, bars, restaurants, grocers — on top.

I asked my friend, journalist Dave Infante of Fingers — a newsletter I read and lovingly describe to others as the Expedite of the booze business (you should subscribe!) to help me understand this structure. It’s a bigger question than I realized.

“Just know,” Infante said, “that this is like asking you, ‘Can you quickly explain to me whether DoorDash’s model is good or bad for restaurants?’”

Oof. But he did offer an abbreviated and comprehensible explanation of the deeply entrenched system that sets up the scale of Tucker’s latest challenge:

“The ‘three-tier system’ is actually 50 different systems administered by the states to regulate how alcoholic beverages are sold. The most common format has — you guessed it — three ‘tiers’ of nominally independent businesses working with one another: supplier, distributor, and retailer. The idea emerged after Repeal as a way to prevent vertical consolidation of the industry by big suppliers; an unintentional consequence is that it wound up concentrating a lot of power in the middle tier, where distributors work as statutorily mandated go-betweens (or, less charitably, bridge trolls) deciding what actually makes it to market.”

At 10 years old, Provi is already a formidable player in this very particular business. In 2022, it merged with SevenFifty, a New York-based beverage tech company with complementary strengths including tools for sales rep, important partnerships, and a detailed product database (plus a robust editorial site, SevenFiftyDaily, which I’ve quoted in this newsletter). Combined with Provi’s retailer relationships and an online marketplace, it became a leader in a category of one.

Less than two months after the deal closed, Provi sued Southern Glazer’s and RNDC, then the country’s largest distributors, alleging the two had colluded to block orders placed through Provi in favor of their in-house platforms. RNDC settled in July 2025 and agreed to take orders through Provi, the Southern Glazer’s case was settled similarly a year ago. (Both companies denied wrongdoing.) Soon after resolving these lawsuits, Provi reported more $5 billion worth of transactions had flowed through its platform since launch.

(Since then, the business has gotten messier; RNDC essentially collapsed earlier this year, a story that, again, my pal Dave at Fingers has covered with authority.)

Tucker says he understands Provi’s role in the future of alcohol sales. “We’re not trying to disrupt the business,” he said.

Still, “The scale of the mess is real,” he said in a follow-up email, bolstering his case that the American bev-alc industry is ripe for a tech intervention.

“Eight-plus distributor reps call on a single retailer every week, there are over 1 million SKUs in any given market, and SKUs are up about 400 percent this decade. Bringing a $170 billion industry online is a genuinely hard technology problem.”

Tucker wouldn’t share too much about Provi’s finances — it has “low double-digit millions of dollars in [annual] revenue,” he said — but did detail its business model in our interview, which continues below for paid subscribers.

Our conversation has been lightly edited for length and clarity.

Expedite: After we spoke in the spring, I didn’t expect you to take a new job so quickly. What drew you to Provi and the beverage industry?

Read more

Read more

Earlier this week, I watched a technologist terrify a room of restaurant people. Dex Hunter-Torricke is an insider-turned-critic of big tech who once spun tech messaging for Mark Zuckerberg and Elon Musk and Google Deep Mind. Now he’s made a career out of the opposite position, spreading hard-to-hear truths about how artificial intelligence will upend life… and we heard it all before lunch on a Monday.

“Humans are under attack,” he told the audience about a third of the way through his presentation at the Welcome Conference, a 2,000-person event led by restaurateur-turned-author Will Guidara. (“Warm fuzzies are in,” I once wrote in coverage of a past year’s event. This was not that.)

The session also included unsettling images of Chinese “dark factories” — there’s no need for overhead lighting when the factory floor is full of autonomous robots, he told us — and one photo of President Trump’s table at the recent state dinner in honor of China’s president Xi Jinping, reminding us that the collective net worth of the people at the table exceeds $1 trillion. AI will consolidate wealth and political influence, Hunter-Torricke explained. It’ll deepen inequality, weaken democracy. It is, he asserted, to be feared.

These are all provocative statements, louder when declared in front of a group of people steeped in the business of humanity. His take on a probable future was so dark, so doomer, that even Guidara — ever the optimist! — seemed to struggle to offer a positive spin after the talk. The audience can make a difference, Guidara finally said, “if we all leave aligned.”

Except I misheard him. At first, thought he said something like, “The people in this room can make a difference… if we all leave alive.”

Oof.

If Hunter-Torricke’s goal was to sell the room on the gravity of this moment in history, it worked.

His talk inspired whispers and concerned chatter for the rest of the day. But Hunter-Torricke failed to empower the assembled hospitality and business leaders with another fact: This AI-dominated moment presents a remarkable opportunity for the restaurant industry. Human hospitality is the antidote to the unchecked dominance of big tech and its march toward an artificially intelligent future.

Historically, the Welcome Conference hasn’t covered much tech at all, aside from nods to title sponsor Resy — someone has to pay for the revelry! — and the occasional dig at the pervasive creep of tech inside restaurants. It’s a miss, in my opinion, because of how these industries complement and contradict one another. There is no business better positioned to mitigate the effects of an overly optimized, highly technical, faceless new technology. We should be encouraged to act, not scared into acquiescence.

In other words: Dex! Tell us what to do about the scary stuff!

The restaurant industry vastly underestimates its power to challenge the tech industry’s self-assured narrative that they alone control the future. It needs encouragement, not fear.

Here’s how we change this: Instead of featuring the big, scary thing, spotlight the weak points of the big, scary thing. What can’t AI do? How can an empowered group of deeply empathetic professionals capitalize on this? How should they wield their power and influence for good in what we have come to understand is a deeply flawed inequitable system?

I wish the audience had received answers to these questions from the expert who’s shared offices and stages and private jets with some of the world’s most divisive and powerful executives. Instead, we left feeling scared, nervous, and nearly powerless on our way back to work.

Later in the day, some of the fear turned to anger. “What was that AI talk?” more than one person asked me. “Is he right? Are we doomed?”

His facts are correct, I answered, but the message is wrong. In the eternal words of Mister Rogers, I reminded them: When you’re scared, look for the helpers. They’re right here in the room.

I’m headed to the Welcome Conference in New York on Monday. See you there?

Read more

Is it weird to be more excited about a DoorDash product update than its just-announced, purpose-built airborne drone?

Don’t get me wrong, the drone, which will soon hit the skies in Northern California delivering food from restaurant to diner in about five minutes (on average), is impressive. Five minutes!

Still, I showed up to a media event at DoorDash HQ on Tuesday less interested in delivery drones and more into understanding what’s next for restaurants on DoorDash. (Here’s more about the drone if you’re into it.)

It’s been (exactly!) a year since DoorDash announced reservations inside its marketplace app. This was an exciting and potentially behavior-changing move, and a year into the experiment, DoorDash has proven its thesis that diners will, in fact, book reservations inside a delivery app. Now it’s connecting the dots with a “restaurant growth platform” called DashOS that should help operators better understand and engage their customers, building stronger relationships and encouraging repeat business.

At yesterday’s event in San Francisco, Parisa Sadrzadeh, DoorDash vice president of strategy and operations (and the exec I frequently talk to about restaurant things), explained how the company is linking formerly disparate products inside the just-announced DashOS. It connects tools across DoorDash: marketplace, direct ordering, reservations, loyalty, guest management, and marketing. Restaurants can track diners across services and direct them toward new channels and experiences with personalized recommendations and targeted marketing campaigns. In other words, all of DoorDash’s restaurant products talk to each other now — and they’re talking about all of us.

In an example presented from the stage, Sadrzadeh searched for a local Italian restaurant inside the DoorDash app. After adding dishes to her cart, a pop-up prompted her to quickly join the restaurant’s loyalty program (powered by DoorDash), sweetening the deal with an immediate and hard-to-refuse $5 off.

For the restaurant, Sadrzadeh said, “that guest just became a loyalty member with one click. The difference between a marketplace customer and a restaurant’s own customer just disappeared.”

(I did not think I’d ever see DoorDash excitedly direct diners away from its lucrative marketplace app and toward an individual restaurant, but times change, I guess.)

A week later, again in Sadrzadeh’s example, the officially loyal guest receives a marketing email directing them to order from the restaurant’s own website (powered by DoorDash). The diner can see their marketplace order history and personalized dish suggestions — a reminder they know and like the restaurant. Later, the diner might get a push notification for a nearby in-restaurant offer or an invitation to an in-person wine tasting, bookable using DoorDash reservations. (Or maybe both! Lucky them.) This diner behavior — order history, spend, visits, attention — is captured by DoorDash and shared with the restaurant, she explained, which makes a compelling case for a restaurant to adopt a ton of DoorDash products.

Whew. Still with me?

There’s a lot of power in these connections, and DoorDash says they create significant guest engagement. More than 1 in 5 — “I’m not sure why we framed it that way, it’s nearly 1 in 4,” Sadrzadeh later clarified — diners who visit the restaurant’s page in the DoorDash app sign up for a restaurant’s loyalty program when prompted. Once enrolled, customers order almost twice as often, spending 1.75 times as much compared to their non-loyalty peers.

And, again, a diner’s one-click opt-in to a loyalty program is “a direct conversion, moving that person from a marketplace customer to a first-party customer,” Sadrzadeh said.

But is ceding marketplace customers — customers who use the DoorDash app to search, browse, and order— to restaurants a good move for DoorDash? Won’t the company lose money, I asked her, if it’s missing out on potential marketplace commissions from these orders?

“If the merchant grows, that’s a good outcome for DoorDash, and for the consumer,” she said. “Restaurants are going to grow, and I think we just are going to meet consumers in the moments that make the most sense to them.”

It’s a lot easier (and more responsible) for DoorDash to do this now that it owns all the moments — from delivery to dine-in — and can apply data from one function to another. That also means the third-party vs. first-party distinction might not matter as much as it once did.

“I think restaurants are going to slowly become aware that the lines are blurring between the two,” Sadrzadeh said, “and I think that’ll be really powerful.”

Also powerful: an $80 billion commerce platform that has a lot — like a lot — of information about what, where, and how we like to eat to use to its advantage.

DoorDash’s other product announcements, ranked in order of my excitement level, include:

1. “Text DoorDash,” which is exactly what it sounds like: text DoorDash to order, and more, without opening the app. The AI assistant works inside text messages and can easily reorder favorites, like the same bowl of Marufuku ramen I eat at least once per week. (It can do a lot more than simply reorder; you can ask for recommendations, specific dishes, even your “usual” with all its modifications pre-selected.) Per the product release: It’s less like traditional online ordering and more like texting a friend while they’re standing in line at your local coffee shop: they already know your milk preference and can order your usual, or send you the menu so you can tell them what you want.

2. Returns on retail goods: I don’t (yet) use DoorDash to buy jeans from the Gap, but I’m sure some people do! For a flat $7.99 fee, customers can schedule return pickups to have DoorDash’s couriers return previously purchased but unwanted items in-store. They wait in line so you don’t have to.

3. The aforementioned DoorDash Air launch, which on Tuesday included three drones displayed on an outdoor mezzanine. Sleek and apparently quiet (DoorDash says the drone is about as loud as a passing car), the drones deliver orders in an average of five minutes. DoorDash air launches soon in a suburb near me with partners including Chipotle and Popeyes. Here’s another glamour shot:

   

After decades spent outside of the diner spotlight, Olo is stepping directly into it. It’ll (finally) launch a diner-facing mobile app October 20 with 220 (and counting!) restaurant brands including Dave’s Hot Chicken and El Pollo Loco and Waffle house. Last week, Olo released a teaser video that’s part ad, part explainer, part recruitment tool to recruit restaurant participation.

For decades, Olo has worked with large restaurant brands, quietly powering online ordering. The company IPO’d in 2021 (likely riding that pandemic ordering high), but exited the public market last summer in a $2 billion, all-cash deal from private equity firm Thoma Bravo.

While it seems a little… late to jump into the marketplace game, Olo CEO Noah Glass told me the timing is exactly. “The thing that’s been holding us back, truly, being a public company, it would have been hard to do this,” he said.

I agree. A new app and diner-facing ordering experience is risky — a big marketing push, a big and unproven strategy shift — the kind of expense and uncertainty that the public market doesn’t love. I can’t imagine Wall Street loves any company trying to go after the DoorDash machine, either.

Glass calls Olo’s app strategy “second-party ordering,” which is a made-up phrase meant to position the app in a category of one, between first-party (ordering directly with the restaurant) and third-party (placing an order on a commission-based marketplace app).

PRODUCT DETAILS:

Olo has some work to do to get hundreds more restaurant brands onboard — the company works with 800 of them. Glass doesn’t seem worried. Olo’s building ordering infrastructure for the future, he told me, a future laden with bots and agents, digital tech that’ll more for diners. Our interview continues below for paid subscribers.

Our conversation has been lightly edited for length and clarity.

Expedite: You’re marking this app with a heavy “savings” angle, which taps into a big-app weakness that I’m always surprised more people don’t know about. It’s smart, but this is also a brand new app that encourages new behavior patterns.

Read more

Read more

The launch of chef Daniel Humm’s $7 boxed meal kit line at Target may have surprised the legions of fans and followers who associate him with the luxury of Eleven Madison Park, his three-Michelin-starred restaurant in Manhattan. But Riff, developed in partnership with Mark Ramadan, a kind of grocery superstar thanks to his work building condiment company Sir Kensington’s before its reported $140 million sale to Unilever in 2017, has been years in the making.

The premise, that an ultra-high-end chef would dabble in Target pasta, is jarring. Or, as Grubstreet puts it, “weird.” But Humm is hardly the first to try. (To my left, as I type this newsletter, is my stash of dried Momofuku noodles priced at $11 for four packets, also sold at Target.)

Similar to the restaurant industry, there’s no guarantee of product success, no matter who you are. The CPG (consumer packaged goods) cycle, from idea to hundreds of store shelves across the country, is long and complicated. Still, the market is white-hot, with an increasing number of chefs and restaurants hoping to get in on the action. Why?

I asked , an investor and CPG expert who spent almost a decade determining which products showed up on shelves at Whole Foods, to help me understand.

The answer is complex, as Truesdell deftly explained in a recent edition of her newsletter, . But to start, there’s a market for these new products. “People are sort of obsessed with brands and with grocery,” she told me. (I blame Instagram.)

Lots of other factors — social, economic, the siren song of a major exit to a giant food company — might compel a chef or restaurant to seek fame on shelves. Truesdell talked me through everything that goes into packaged-good success (or failure) for a chef or restaurant, including marketing, packaging, audience development, celebrity, and, ultimately, how to judge if a product will thrive.

This conversation, available to Expedite’s paid subscribers, has been lightly edited for length and clarity.

Expedite: As someone who’s seen a lot of consumer products, what’s interesting to you about Riff?

Elly Truesdell, New Fare Partners: “I find this particularly interesting because of Daniel Humm’s profile and background. Being in the industry for a long time — it’s been almost 20 years now — I’ve seen a growing number of chefs entering the grocery world. Ten-plus years ago, it was definitely not like this. There was no cachet to launching a CPG brand, there wasn’t the same sort of luster there is today. So it’s not totally surprising to me that people like Daniel Humm are going after consumer goods.

“On the other hand, it is sort of surprising. There’s some part of me that thinks a lot of the recent culinary ambition that’s going into grocery is largely driven by the exits that they’ve seen — by dollar signs. I’ve talked to dozens and dozens of people and worked with several to help launch their product lines, and a lot of them are very transparent about this. The tricky thing about that is, once people get into CPG they realize very quickly how hard it is.”

Hard how? Where do chefs get into trouble?

Read more

Then, the story was about the cloud-based point of sale company with a cute logo that wanted to do it all. Today, the question is how far Toast plans to go.

Since Toast’s market debut in 2021, its market share has grown from about 1 in 20 small and midsize restaurants in the U.S. to 1 in 5. Restaurants on the platform are paying more, on average: revenue per restaurant is up from about $8,000 to $13,0001 over the last five years, Aman Narang, the company’s CEO, estimated during a September Goldman Sachs conference.

During this time, Toast introduced a host of new products and integrations, including a budding partnership with reservations platform Resy and an AI-powered assistant, Toast IQ, to help restaurants make sense of all of the data it collects. At this point, calling Toast a “point of sale and payments company” understates its importance and position in the market.

The company agrees.

Read more

On Tuesday, DoorDash announced plans to acquire Grubhub’s college campus business for $300 million. Grubhub Campus Dining, f.k.a. Tapingo, serves millions of students at 450 colleges and universities — students place orders for pickup through the Grubhub app, which links to campus meal plans. Grubhub acquired Tapingo from its founders for $150 million in cash in 2018. Then, Tapingo was dominant and specialized, serving students at 150 million colleges and universities. And since then, Grubhub has managed to triple that coverage while taking advantage of the relatively closed networks of college campuses to iterate and experiment — deploying delivery robots, for example. (A self-contained college campus is surely a more bot-friendly environment than most city streets.)

2018 was also the year that Grubhub’s valuation (and probably its popularity!) peaked — the company was worth as much as $13 billion before steadily falling in value and market share to relative newcomers DoorDash and Uber Eats. Early last year, Wonder, the New York-based food hall/delivery kitchen/virtual brand moonshot from entrepreneur Marc Lore, acquired Grubhub for just $650 million — $150 million cash and $500 in debt.

In addition to the acquisition, DoorDash also committed $125 million to Wonder’s Series D fundraising round.

“Wonder was founded to make great food more accessible. To us, that means higher-quality food at lower prices, with more choice and convenience, in more communities,” Lore said in a statement. “The partnership also allows Wonder to focus on our long-term ambition: a fully autonomous food system that can plan, produce and deliver personalized meals at scale, serving all 21 meals people eat each week.”

As Wonder streamlines Grubhub’s business, DoorDash scores big, picking up successful slice of Grubhub’s business that once set it apart from its competitors, reaching brand-new adults as they establish consumption habits.

Lore says DoorDash is well-positioned to grow the campus dining business. I can think of approximately 83 billion reasons why.

It’s clear that professional influencers have changed the restaurant industry for better or worse, but the role is in the spotlight again af…

Read more

Their beef (mostly) stemmed from a single phrase in the notification: system of record. In March, the 28-year-old reservations platform emailed its restaurant customers to flag new language in the TOS requiring them to use OpenTable as the restaurant’s system of record — its primary source for reservations, table and guest management.

The clarification came amid increasing stakes in the bookings business. Page, who, to be clear, leads a direct OpenTable competitor, said the only way for restaurants to comply is to “make OpenTable the center of gravity and manually reconcile everything else into it.” That’s obviously not in her company’s best interest — and Page also argued that it’s not in a restaurant’s best interest, either.

SevenRooms has always marketed itself as a way for restaurants to get closer to their booking data, no matter where any given reservation originated. In June, SevenRooms released a product called Channel Connect, designed to pull reservations and guest info from multiple channels. “Guest data is centralized into a single system of record, with secure, permissioned access across connected platforms,” it promises.

This came months after original TOS kerfuffle, including the system of record language, got enough attention that OpenTable had to clarify its position.

“OpenTable’s client agreement citing usage of OpenTable as System of Record isn’t a new term or a new way of working with many of our partners,” a company rep later said in statement emailed to numerous outlets. “By extending our System of Record standard to all partners, we are ensuring a single source of truth that protects guest and restaurant data, maintains real-time availability, and guarantees that when a guest books a table, it actually exists.”

Ah, yes, welcome to the chat, single source of truth, a close cousin of system of record and a phrase that’s equally as technical. The so-called reservations wars have a new front: the underlying technical architecture that powers them.

Before they showed up in restaurant tech discourse, the terms ‘system of record’ and ‘single source of truth’ were more technical terms.

I asked my favorite technical source, my husband who’s a career software engineer, to help. He explained the terms in context I understand:

Imagine a shared Google Doc that a few people work from, he said. That doc, which tracks changes and allows multiple inputs from varying sources, is a system of record. The same doc, holding many pieces of info, might be the single source of truth, or it could be compiled with information from multiple sources of truth. Regardless, a single source of truth exists when we (collective we!) decide that there is one, true, absolute version of a piece of information, no matter how many copies (like, say, an old Word doc circulating via email) exist.

I love an authoritative source.

My frenemies Gemini, ChatGPT, Claude and their large language model peers, do too. When I — or the AI — produce(s) a piece of written work, an answer to a question, or just an appropriate response, we seek authority. My strength is human judgement and decades of experience… and while the bots care less about these attributes (rude) they still seek authority. A useful, non-hallucinating AI system needs access to one authoritative truth; otherwise the tool pulls from anything it can reach, with no way of knowing which version of a “fact” is correct.

When information on the internet doesn’t match information in real life — for example, the closed restaurant whose AI-powered phone operator insisted was open, the dinner restaurant that AI insisted was open for brunch — customers are underserved. Tech-savvy diners might know to blame the bots; others just blame the restaurant. At the same time, big tech companies say they want to be true partners, maybe even advocates, for their restaurant clients. Becoming their verifiable system of record and source of truth — by technical standards and also practical ones — carries serious weight.

And in a tech era defined by rapidly evolving AI and a growing list of integrations that determine the way a restaurant’s information shows up inside search engines and search agents and chatbots and social media listings and online map pins… it’s good to be the authority.

We all — me, a journalist, Claude, a chatbot, et al. — need access to good data to produce good work. Unfortunately, as we’ve learned, garbage in equals garbage out. A good result is only as good as its source of truth, served by its system of record.

Earning that status is the closest the major players in a fragmented industry can get to winner-take-all. And behind this battle in the reservation wars is a much simpler squabble: the fight over who gets to be right.

I talked to Adam on my podcast, The Simmer, about the book and — most importantly to readers of this newsletter — the one thing he really wants restaurant tech execs to know. Listen on Spotify, Apple Podcasts, or right here:

OpenTable released its ‘largest suite of new and updated product features for restaurants’ last week.

I’ve got beef. So does Wonder.

Six months after buying single-location fried chicken concept Blue Ribbon Fried Chicken, the ghost kitchen / food hall / fast food / delivery / meal kit company last valued at $9 billion bought another NYC restaurant. It’s acquiring New York City’s Salt Hanks, a wildly popular single-location sandwich shop with a viral french dip sandwich. The sandwich will be available at some Wonder locations beginning this fall.

“The sandwich that’s always been worth the wait is about to become a lot easier to get,” a company release declared.

That’s true. But I’m hung up on the rest of the language Wonder used in the announcement, which — unlike the Blue Ribbon announcement or subsequent coverage of Wonder’s acquisition of an eight-unit barbecue chain — doesn’t contain the word “acquired,” “bought,” or even “purchased.” Instead, the company softens the language: “Salt Hank’s joins Wonder,” it says. “The partnership is built on a shared belief that exceptional restaurants shouldn’t have to sacrifice what makes them special in order to reach more people.”

ETA: The word “acquisition” does appear in the official release, but not in a shortened version that was emailed to me yesterday. I don’t remember seeing it then, but it’s also been a real challenge of a week. Regardless, it’s in there now, and if I did miss it, I regret the oversight.

This is true, too! But it understates what’s actually happening. The restaurant, its branding, its menu, its recipes (Wonder calls this IP) have been sold. Chef Daniel Rubenfeld will join Wonder, while co-founder Henry Laporte — Hank — will “partner closely.”

(I asked a Wonder rep to clarify the announcement, and they confirmed it is an acquisition — Eater’s Nadia Chauduhury got it right in her Thursday coverage shortly after the announcement.)

Wonder has a lot of irons in the fire — it owns a meal kit service and a third-party delivery company; it’s expanding locations fast in the Northeast and in Texas, where it’ll soon start delivering by airborne drone. It’s teased a kind of build-your-own restaurant concept for influencers (or anyone that wants to spin up their own AI-generated restaurant concept in a couple minutes) — the robot it bought from Sweetgreen will help assemble the meals. It is, to put it lightly, a moonshot with a lot — a lot — of money behind it.

So what does it want with a single-location sandwich spot? Scale.

“Customers can expect the same iconic French Dip — made with dry-aged prime rib, caramelized onions, melted provolone, horseradish aioli, rich hot au jus and served on a French demi baguette — but without the long lines,” Wonder promises.

In an interview with People (what?) which also avoided words like “acquired” or “bought,” Laporte explained that he trusted Wonder to scale the concept appropriately. “...they don’t wanna f--- up the sandwich,” he said.

I’ve done a lot of reporting on this company, including talking to chefs with notable brands that have worked with Wonder. Some of those deals were partnerships; chefs reported receiving favorable deals that allowed them to walk away at any time. Not all have stayed on, but at the time of our interviews, all of them spoke highly of the company, its leadership, and its commitment to scaling while maintaining food quality.

Still, I’m not convinced that even the most meticulous reconstruction of an iconic, viral sandwich can replicate the original. Will it taste as good if you don’t have to fight as hard to get it? What happens to the magic of an indie business when its essence (and IP!) is taken over by a highly capitalized corporation? We’ll see, I guess. But I hope everyone feels like they got a great deal.

“I’m happy to be able to finally answer those DMs that are like, ‘Why the f--- can’t I get this sandwich!’” Laporte told People. “And now I’ll be like, ‘You can!’ You know, to a certain degree.”

OpenTable is putting influencer videos inside restaurant listings.

OpenTable is adding influencer videos to listings across the country in partnership with a SF-based startup called Seekeasy, an analytics company that pairs restaurants and digital creators. (It calls itself “the demand intelligence layer for the food industry.”)

Influence is hard to prove, but diners do care about social media aesthetics. According to OpenTable’s 2026 dining trends report, well over half of diners age 60 and under (including 79 percent of millennials) said a restaurant’s “Instagram/TikTok worthiness” is important when making a decision about where to dine.

“Creators bring the restaurant experience to life,” Seekeasy CEO Erick Hachenburg told me. “Diners can now quickly use their relatable stories to make decisions and enjoy the experience themselves.”

Diners are more open to AI now, according to a survey from PAR Technology, a point of sale company.

In PAR’s survey of 1,000 customers last year, 41 percent said they would choose a restaurant with no AI over a restaurant that uses AI in any form. This year? Just 27 percent said the same thing, signaling that 3 in 4 diners don’t mind when a restaurant uses the tech. Progress?

ICYMI:

The two San Francisco-founded companies first partnered in 2022 and have already connected reservations and point of sale data for restaurants. Today, they promise “deeper insights sharing… to address the restaurant pain point of disconnected systems,” per a press release.

“We believe in a seamless ecosystem and we believe in giving operators the best products we can so that it’s actually seamless,” James Schonzeit, head of food and beverage at Square, told me during an interview this week.

“The opportunity is bigger than simply connecting two systems,” said Robin Chiang, chief growth officer at OpenTable, over email. “Because OpenTable and Square’s engineering teams built the integration directly, the platforms communicate bi-directionally in real time — giving restaurants one continuous view of the guest from booking and arrival through ordering, payment and future visits.”

This might sound familiar.

That’s because top competitors to OpenTable and Square just rolled out a similar integration. Reservations service Resy and point of sale platform Toast announced a major partnership last year. Last week, they debuted (long awaited) Resy-powered digital chits — notes and info about a diner — to Toast handhelds.

Bookings are back in the discourse after a recent Wall Street Journal piece lamenting the state of the business “Reservation apps have made getting into a restaurant a nightmare,” its headline dramatically declared. “Nightmare” is a strong word for what’s happening to reservations — even in New York, considered the epicenter of reservations mania thanks to its density and probably its status as the onetime walk-in capital of America (as anyone who’s experienced a bit of NYC restaurant serendipity will tell you, this writer included).

But it is true that an increasing number of partnerships, arrangements, exclusive events, and status-driven access allow technology companies and their credit card benefactors to broker the relationship between restaurant and diner. Now additional info — data on customer spend, payment methods, visit history, and more — is creeping into in the (very visible) mix.

“Reservations, payments, and guest data shouldn’t be different systems that restaurant operators have to stitch together themselves,” Square’s Schonzeit said in a release. “No one has time for that, and we’re not in the business of gatekeeping their most valuable insights.”

After a tumultuous few months, Square is once again highlighting its capabilities at full-service restaurants.

A deal with the country’s top reservations platform brings Square serious visibility. It’s now OpenTable’s preferred point of sale partner.

“We expect this collaboration will accelerate our customer acquisition efforts,” execs wrote of the deal in Square’s August letter to shareholders.

Last February, Square parent company Block laid off half of its workforce, some 4,000 workers including its then-head of restaurants and much of his team. Just hours after employees received notice that their roles had been cut, Block’s executives extolled Square’s progress inside restaurants in the company’s fourth-quarter earnings call, flagging a year-over-year increase in gross payment volume — total transaction value — at restaurants and other food-service vendors using Square’s point of sale.

Six months later, gross payment volume from restaurants and other food businesses on Square is up again, this time 20 percent year over year. GPV in the US accelerated to its strongest growth rate since early 2023.

Square got its start with small, independent businesses and still thrives in places like coffee shops and food trucks. But it’s expanding fast. Earlier this year, it added a new drive-thru product, targeting multi-location quick service restaurants. In Block’s early August earnings call, chief financial officer Amrita Ahuja called “mid-markets” — businesses processing over $500,000 through Square annually — the company’s “fastest ramping segment,” growing over 20 percent.

This makes it a great time to highlight the company’s ability to serve full-service restaurants — and their guests. And the companies promise there’s more to come from its OpenTable deal: “Future enhancements,” Square’s press release promises, “will focus on faster onboarding, smarter reservation workflows, richer guest intelligence across every ordering channel, and deeper integration of loyalty, customer engagement, and payment experiences.”

Last week, one of my San Francisco favorites announced an alt funding platform of its own. Pasta Supply Co., helmed by chef Anthony Strong, announced plans to open its third pasta shop-slash-restaurant, selling and serving seriously great pasta — bucatini, egg malfaldine, chicken-prosciutto tortelli — in local neighborhoods.

To raise money to build out the space, an under 1,000-square-foot former Peet’s Coffee shop in Cole Valley near the Haight, Strong instituted a crowdfunding campaign: area “investors” who pay $550 upfront get free pasta — a package that’ll feed two adults, at least — every week for the next year. Those who give $4,000 get pasta for life. (Important fine print: “Non-transferrable, even in the event of mortal termination.”) A few hours in, the campaign was already a quarter of the way to Strong’s initial funding goal.

Pasta Supply Co. previously ran crowdfunding campaigns using Kickstarter. Now, with a following of very satisfied customers (this writer and her butter-noodle obsessed children included), Strong decided to circumvent platform fees and go direct. People sign up for the pasta plans via the store’s Square storefront. (This also makes the funding transactional, clarifying everyone’s role. “We’ve had some ick moments over the years,” Strong told me.)

To manage all the subscriptions, Strong vibe-coded a customer tracker using Claude Code. Patrons are photographed when picking up their first pasta order and get one “skipsie” per month. The tracker pulls data from the point of sale system, tracking trends over time — including a particularly San Francisco phenom:

“We can see how many people skipped Burning Man week,” Strong said.

UPDATE:

A year after announcing an extended partnership, Resy and Toast are finally rolling out a huge part of the deal’s promise: digital chits. That means diner info from Resy’s guest book — guest notes, visit history, special occasions — will show up on Toast screens for servers.

ICYMI:

In the nearly decade-and-a-half between its launch and last year’s acquisition announcement, SevenRooms grew into a reservations and customer relationship management powerhouse. It grew fast inside restaurants but also inside hotels and casinos and in strategic international locations like London and Dubai and Australia.

By the time DoorDash showed up with a $1.2 billion offer, SevenRooms was working with 13,000 venues around the world. “It’s about adding to our platform business,” DoorDash CEO Tony Xu said of the deal at the time, which, according to my recent interview with SevenRooms’ top exec, has proven true.

Joel Montaniel, a SevenRooms co-founder and its founding CEO, still leads the company from within its parent org. Until this conversation, we hadn’t spoken since the acquisition announcement. He couldn’t say much about the future then; the deal wasn’t closed, the companies couldn’t strategize yet. (He was also still trying to return the barrage of text messages one receives after announcing a billion-dollar exit.)

This week, I finally got to host Joel on my podcast, The Simmer. It was an overdue conversation, but I’m glad we waited. Nearly a year since the DoorDash acquisition closed, he’s able to talk about it and the products and initiatives that the once-unlikely partnership has spawned. (I know at least one that has the competition a little rattled, more on that below.)

This episode of The Simmer is available on Spotify and Apple Podcasts.

Paid subscribers can keep reading for highlights from my interview with Montaniel. Quotes have been edited for length and clarity, but I captured the last line — my favorite — exactly as he said it.

On SevenRooms’ decision to go after hotels and casinos as early clients:

“We didn't think that the world needed another reservation system. What we thought was missing was customer data and a system to help the restaurant manage their customers, understand their customers, and then make that data available at their fingertips. So when Kristen walks in, it's not that she has a reservation, it’s that she's a regular of ours. She's friends with the owner, and she happens to have a reservation.

Read more