Locations/New York, NY/A long read on marketplace economics in the city of 27,000 restaurants
The 27,000 Restaurants Atlas|Issue 01 / New York City|Published May 11, 2026

On a Wednesday at 7:45pm, a restaurant in New York City takes its 41st order of the day, and pockets less than two thirds of the menu price.

A long read on what New York City's fee cap actually changed, what it did not change, and why a direct ordering channel is the only economic move that still works in the country's most marketplace-saturated restaurant market.

A New York City restaurant at dusk, lit storefronts along an avenue, with a yellow cab passing in the foreground.
Photo: representative NYC streetscape. The restaurant in the lede is a composite drawn from operator interviews and NYC DOHMH permit data.

It is Wednesday, the slow night of the slow week, the kind of evening on which the restaurant should be coasting. The operator, a chef who took over from her father in 2019, has just printed the day's report off the POS. Forty-one orders. Roughly the day she expected. The total at the bottom of the ticket reads $1,840 in gross sales, which on paper is a respectable Wednesday in a 38-seat Brooklyn neighborhood spot.

Twenty-eight of those forty-one orders came through DoorDash. The rest were three Grubhub, two Uber Eats, six pickup that the host took at the counter, and two phone orders the line cook answered between dishes. After the DoorDash commission, which Local Law 6 of 2021 caps at 15 percent, and the additional capped fees, and the 3 percent payment processor's cut, and a $96 share of the week's promoted listing spend that she allocates by day, the operator's effective rate on marketplace gross is closer to 31 percent.

That word, effective, is the word the marketplaces do not put on a rate card. The fee cap is real. The fee cap is enforced. The advertising spend on the marketplace, which the operator pays voluntarily to keep her listing visible against the seven other neighborhood spots within four blocks, is not capped by anything except her marketing budget, and her marketing budget grew last year.

Of $1,840 in gross sales today, roughly $1,270 hits the bank. The difference is $570. Six hundred dollars, less the rounding, on a Wednesday. That is one shift of a kitchen porter, the catering deposit on the small Saturday corporate order, the wine list update she has been deferring for three months. It is also a number she has been quietly assuming was the cost of doing business in New York.

It is not. It is the cost of doing business on someone else's distribution channel. The remainder of this piece is a long, slow look at what changes when an NYC operator stops paying that cost.

I.The 27,000 number, broken down

Twenty-seven thousand is not a number of restaurants. It is four different industries that share a fee cap.

The figure most often cited for New York City, roughly 27,000 active restaurant permits issued by the Department of Health and Mental Hygiene Food Service Establishment program, is technically correct and conceptually misleading. It treats the slice shop, the 22-year-old dim sum house, the chef-driven tasting menu in Greenpoint, and the Sweetgreen on Park Avenue as members of the same industry. They are not. They face four different versions of the marketplace problem and four different versions of the fix.

The largest of the four groups, by a wide margin, is independents with under five employees. There are approximately 18,500 of them, which is roughly two thirds of the city's restaurant universe. These are the businesses for whom marketplace commission, even at the capped 15 percent, is a structural cost rather than a marketing line item. They have no pricing power against the marketplaces. They have a single counter, a single cook, and a single owner who works the register on Saturdays. Their alternative to a marketplace listing is not a different marketplace listing. It is, eventually, not being on a marketplace at all.

The second group, multi-location small groups with two to eight locations, is a smaller universe at roughly 4,200 establishments. These are the family-run mini-chains, the Greek diner with three locations across Manhattan, the West African restaurant with two Brooklyn outposts and a Bronx storefront. They have just enough volume to negotiate, and just little enough leverage to lose the negotiation. They are the operators most actively building direct ordering channels in 2026, because the math of running their own customer relationship across multiple locations is the math that finally tips in favor of leaving.

The third group, single-location chef-driven kitchens, is the smallest of the four at roughly 2,800 establishments and the most visible in the press. These are the operators whose food is reviewed in Grub Street and Eater NY every week. Their average ticket is higher, their order volume is lower, and the marketplace pressure on them shows up not as a margin crisis but as a reputational mismatch. A tasting menu does not deliver well. The marketplace listing, when it exists, is a defensive minimum, not a growth channel.

The fourth group is national and regional chain locations, approximately 1,500 of them. Chipotle, Sweetgreen, Joe's Pizza locations operating with shared branding, &pizza, the Just Salad downtown lunch corridor. These operators already have their own apps, their own loyalty programs, their own promoted-search dollars deployed on Google and Meta. The marketplace fee is a customer acquisition line item they have largely converted from variable cost to managed channel. They are not the operators the rest of this piece is about.

The 27,000 number is the same number for all four groups. The economics inside that number are not. The next four sections look at what the 27,000 actually contains.

II.The Local Law 6 of 2021 reality

The Council capped fees. The marketplaces moved the fees to a budget the Council did not cap.

On August 17, 2021, Local Law 6 of 2021 took permanent effect after the temporary emergency cap that ran through the pandemic. The law, codified through the NYC Department of Consumer and Worker Protection, put two caps in place. Marketplace delivery fees charged to restaurants are capped at 15 percent of the order subtotal. Other marketplace fees, including non-delivery service fees, marketing-listing fees, and ancillary charges, are capped at 5 percent in aggregate. The total cap, in other words, is 20 percent.

The Council's intent was clear. In the period from 2018 through the early pandemic, NYC restaurants were routinely paying 25 to 32 percent in combined per-order marketplace fees, plus payment processing on top, plus various consumer-side fees that did not flow to the operator but shaped the listed price. By capping the per-order math, the Council restored an estimated 7 to 14 points of effective margin to a typical NYC marketplace order. On a slice shop's $300,000 a year of marketplace gross, that recovery is roughly $30,000 to $40,000 in restored margin per location per year. The law mattered. The law worked, on the dimension the Council legislated.

"The fee cap was a precondition for survival in 2020. By 2026 it is a starting line, not a finish line. The marketplaces routed around it. The operators who do not also route around it are paying for the same distribution at a different line item."
Composite framing, drawn from public reporting and operator interviews

What the law did not cap was advertising spend on the marketplace, which is a voluntary expense the operator chooses to make to elevate their listing's visibility against competing listings. Promoted listings on DoorDash, sponsored placement on Uber Eats, and the equivalent products on Grubhub are not classified as fees in the regulatory sense. They are classified as advertising, which is unregulated, and which the marketplaces have invested heavily in making both more granular and more necessary since 2021.

The practical result, for a typical NYC operator, is that the line item formerly known as commission has been partly migrated to the line item known as advertising. The total cost of being on the marketplace has not fallen by the full 7 to 14 points the cap implied. A typical NYC marketplace operator I model spends 8 to 12 percent of their marketplace gross on promoted listings in 2026, which is a discretionary line item that did not exist as a meaningful budget for the same operator in 2019. Add back the regulated 15 to 20 percent in fees and the picture is a 23 to 32 percent effective rate, which is roughly where the operator was before the cap, minus a few points.

None of this is a criticism of the Council, which legislated the part of the marketplace economics that was directly extractive and which produced a meaningful real-dollar improvement for the city's restaurants. It is, however, the diagnosis of why the fee cap by itself is not the end of the story. The end of the story is the operator's ability to build a customer acquisition function that does not run through the marketplaces at all. That is the long, slow project the rest of this piece is about.

Two operational details, since they matter. First, the law applies only to third-party marketplaces, not to a restaurant's own ordering site or to a flat-fee software platform the restaurant runs itself. Second, the law applies only inside the five boroughs. An operator delivering across the Hudson into New Jersey, or across the East River to certain Long Island zip codes, may pay a different fee schedule on those orders. Most independent operators do not, in practice, hit that edge case at meaningful volume.

Delivery fee cap
15%
Maximum third-party marketplace commission on delivery orders, per NYC DCWP.
Other fees cap
5%
Maximum on non-delivery service, listing, and ancillary fees. Combined cap is 20 percent.
Effective since
August 17, 2021
Permanent law, replaced earlier emergency cap. Enforcement is by NYC DCWP.
III.Five boroughs, five operators

The fee cap is the same in all five boroughs. The math of staying on marketplaces is not.

Five composite profiles, drawn from typical operators in real NYC neighborhoods. Each profile presents an operator's relationship to the marketplaces, the specific marginal economics of their segment, and the lever they have available to shift the math.

Borough 1 of 5
Manhattan
Chinatown, 10013
Archetype: Dim sum house, 22 years old, third generation
Cuisine: Cantonese dim sum and Hong Kong style cafe menu
Data callout
Marketplace effective rate, including promoted listings
31%
On a $48 ticket, $33 hits the bank. The other $15 is fees and ad spend.

On Mott Street between Canal and Bayard, a dim sum house has been pushing carts for 22 years. The grandmother is gone, her son runs the dining room at lunch, his daughter handles the night shift and the marketplaces. Saturday brunch is still the bulk of the business, but Saturday brunch has not grown in revenue in eight years even though the dollar volume on the books has. The difference is the marketplace orders, which arrived around 2017 and now account for roughly a third of weekday lunches.

The economics are stark. A $48 family combo on Grubhub returns roughly $33 after a 15 percent commission, a 3 percent payment fee, and the share of weekly promoted listings the daughter buys to keep the listing visible against the seven other dim sum places within a four block radius. That share, the promoted listings, is the line item that does not appear in any rate card. It is also the line item the Council did not cap in 2021.

The family has talked about going off marketplaces three times. Each time they pulled the trigger they lost roughly 40 percent of the marketplace order volume in the first month, recovered roughly 25 percent through direct site re-routing within 90 days, and concluded that the math was a wash without a real customer acquisition channel. Voice AI in Cantonese, picking up the phone calls they currently miss between 11:30 and 2:30, is the channel they have not yet tried.

Borough 2 of 5
Brooklyn
Park Slope, 11215
Archetype: Slice shop, 11 years old, single counter
Cuisine: New York slice, square Sicilian, two specialty pies
Data callout
Annual marketplace fee bleed, this slice shop
$46,000
On roughly $310,000 in marketplace gross. Roughly 15 percent of marketplace revenue, before payment processing.

Seventh Avenue near 9th Street, a slice shop with eight stools and a window that opens onto the sidewalk. The owner is a former architect who decided in 2014 that dough was a more honest medium than facades. The shop does roughly 150 slices on a Monday, 320 on a Friday, 450 on a Saturday night when the post-bar crowd hits at 11:30pm.

The shop's average ticket is $9.40. A 15 percent commission on a $9.40 ticket is $1.41. That commission was set by the Council as a cap precisely because slice shops were going under in 2020 paying 27 to 32 percent. The cap helped. The cap did not fix the underlying problem, which is that a $1.41 fee on a $9.40 product line, repeated 80 to 120 times a day, is roughly $128 a day, $46,000 a year, paid to a platform whose marginal contribution to the slice shop's customer base is a search ranking the owner has no real ability to influence.

The owner did the math three years ago. He found that turning off marketplaces entirely would lose him roughly $19,000 a year in marginal sales. Staying on marketplaces costs him $46,000 a year in fees. The math says leave. The risk, which is the part the math does not capture, is the 320 first-time customers a month who find the shop via the marketplace and then come back in person, who are the actual long-term business.

Borough 3 of 5
Queens
Astoria, 11102
Archetype: Halal cart graduated to brick and mortar, 6 years old
Cuisine: Chicken and rice, lamb gyro, falafel platter
Data callout
Monthly promoted listing spend, this halal storefront
$1,400
Above and beyond the 15 percent capped commission. Buys placement, not capped by the 2021 law.

30th Avenue near Steinway, an operator who ran a cart on 53rd and 6th in Midtown for nine years before crossing the river to open a 14-seat storefront. The cart still operates, run by his cousin, and the storefront does roughly 40 percent delivery, 30 percent pickup, 30 percent dine in. The delivery share grew through the pandemic and never fell back.

The Astoria store is a model case for why Local Law 6 was a partial fix. On the marketplace side, the per-order fee is genuinely capped at 15 percent. On the spend side, the operator buys roughly $1,400 a month in promoted listings across three platforms because his neighborhood has nine halal options within a half mile and he is the only one with a James Beard semifinalist nod from 2024. The promoted spend is what the listing nod actually translates into. It is also not capped.

Voice AI is what he is currently testing. The cart accent, when the AI is configured to speak the way his cousin speaks on the cart at 1am, is the unlock. Customers who are repeat visitors call to ask whether the lamb is on the rotisserie tonight, which is a question that has a different answer on Tuesdays than on Fridays. A phone tree cannot answer it. A live cousin can. A voice AI trained on the menu and the day's specials answers it correctly more than 90 percent of the time, which is the benchmark he is watching.

Borough 4 of 5
Bronx
Arthur Avenue, 10458
Archetype: Italian-American institution, 47 years old, family owned
Cuisine: Red sauce Italian, fresh mozzarella, Sunday gravy
Data callout
Direct online ordering fees, this institution
$620 / month
On $14,000 in direct gross. Roughly 4.4 percent all-in, vs 20 percent capped marketplace.

Arthur Avenue between East 187th and East 189th. The restaurant has been open since 1979. Its grandfather opened it, its father expanded it, its current operator is the granddaughter, who is the first generation to take marketplaces seriously and the first generation to deliberately turn them off again. The restaurant has roughly 80 covers a night on a Tuesday and 220 covers a night on a Saturday, and delivery is roughly 12 percent of revenue, which is low for NYC and intentional.

The granddaughter's argument is that the marketplace shape of the business, the lukewarm chicken parmesan in a paper bag, is not the business her grandfather built. The dine-in business is the business. Delivery is a service offered because customers ask, not a channel she optimizes. Her marketplace presence is on one platform, deliberately not promoted, deliberately at a tight 4-mile radius, and her direct online ordering site does roughly $14,000 a month at a 3 percent payment processing cost and a flat monthly software fee, which works out to roughly $620 a month all in.

She is the operator the rest of NYC's marketplace dependents look at and either dismiss as a niche case (an institution, with brand) or study as a model (an institution, with discipline). Both readings are correct. The brand bought her the option of saying no to marketplaces. The discipline cashed the option.

Borough 5 of 5
Staten Island
St. George, 10301
Archetype: Neighborhood bistro, 9 years old, 38 seats
Cuisine: American bistro, weekly changing menu, full bar
Data callout
Dine-in conversion from direct-channel customers
17%
Vs an estimated 3 to 5 percent from marketplace-acquired customers, per the chef-owner's own tracking.

Bay Street near the ferry terminal. A bistro that runs a weekly changing menu, a tight wine list, a brunch on Saturday and Sunday, and a delivery business that the chef-owner considers a necessary inconvenience. Staten Island is the borough that the marketplaces undercover and the borough where the marketplaces matter the most, because the population density does not support a high-velocity word-of-mouth channel the way Brooklyn or Queens does.

The bistro's average ticket is $34. Marketplace delivery is roughly 18 percent of total revenue. The chef-owner has the same complaint every NYC bistro has, which is that the marketplace customer does not return as a dine-in customer at any rate that justifies the marketplace fee. The fix she is testing is a branded loyalty program through her direct ordering site, with a $5 credit on the first reorder and a Saturday brunch reservation perk for repeat ordering customers. Six months in, the loyalty list has 1,180 customers, of whom 41 percent have placed a second order and 17 percent have come in for brunch. The dine-in conversion is the metric she actually cares about.

IV.What $1 of revenue is actually worth

The fee cap moved the line. The direct channel moves the operator off the line entirely.

This visualization, custom-built for this piece, decomposes one dollar of order revenue under two distribution choices an NYC operator faces. The top bar is the marketplace dollar, after Local Law 6 of 2021. The bottom bar is the direct ordering dollar, on a flat-fee platform with Stripe payment processing.

The 26 cent gap between the operator's keep rate, 69 cents marketplace versus 95 cents direct, is the number that determines whether a typical NYC operator survives the next decade. It is not a marketing number. It is a math number. The fee cap closed roughly half of that gap. Building a direct channel closes the other half.

Model assumptions: 15 percent capped commission, 5 percent other capped fees, 8 percent allocated promoted listing spend, 3 percent payment processor. Direct channel assumes flat-fee software prorated at roughly 2 percent of order value on a $250 a month tier at modest volume.

The marketplace dollar
$1 of marketplace revenue, post Local Law 6
15%
8%
69%
15% Capped commission
5% Other capped fees (delivery, service)
8% Promoted listings (uncapped)
3% Payment processing
69% Operator keeps
The direct dollar
$1 of direct ordering revenue, same operator
95%
0% Commission
3% Payment processing
2% Software (flat fee, allocated to this order)
95% Operator keeps
Per-order operator keep, direct minus marketplace
+26 cents
On 1,000 marketplace orders a month at an average ticket of $32, this is roughly $8,300 a month, or $99,500 a year, restored to the operator at the same gross sales volume. That is the real number the fee cap did not get to.
V.The slice shop survival essay

A 15 percent commission on a $9 slice is $1.35. Multiply that.

The New York City slice shop is the most overdetermined small business on the East Coast. Three to five employees. A counter, a slicer, two pizza ovens, one cooler, one register. An average ticket between seven and fourteen dollars depending on whether the customer adds a soda. A daily order volume that ranges from 80 slices on a slow Monday in February to roughly 500 slices on a Friday in August. A rent that, in the boroughs, has roughly tripled since 2010 and that, in Manhattan, has roughly quintupled.

The slice shop is the operator the fee cap was most clearly designed to protect. Of all the segments of the 27,000 number, none has an average ticket lower or a marginal product more commodified. A slice is a slice. The marketplace's pricing power over the slice shop is more or less absolute, because the customer's switching cost between this shop and the one a block over, on the marketplace, is zero.

Run the math. A 15 percent commission on a $9 slice is $1.35. If the shop sells 150 slices a day through marketplaces, that is roughly $202.50 a day, $1,418 a week, $73,710 a year going to marketplaces in commission alone, before promoted listing spend and before the 3 percent payment processing. Add the promoted listings, conservatively $700 a month for a competitive Brooklyn corner, and that is another $8,400 a year. Add the payment processing on the gross marketplace sales of roughly $492,000 a year, which is approximately $14,800. The total marketplace bill on a typical NYC slice shop running 150 marketplace slices a day is approximately $97,000 a year.

Ninety-seven thousand dollars is the price of one and a half full-time pizza makers in 2026 NYC. It is also, not coincidentally, what the same shop spent in marketplace fees in 2019, before Local Law 6 of 2021. The cap moved the math by roughly 30 percent on the commission line. The promoted listing line absorbed roughly half of that recovery. The net improvement to the shop, post-cap, is between 12 and 18 thousand dollars a year, which is real money, and which is also not the difference between staying open and closing in a market where the lease alone is up 40 percent over the same five-year window.

The shop that shifts even a third of its marketplace orders to direct ordering, at the same gross volume, recovers between thirty and forty thousand dollars a year of margin without any change to the menu, the rent, the staffing, or the customer count. That is not a marketing improvement. That is a balance sheet improvement, and it is the reason a slice shop is the canonical case for what a direct ordering channel does for NYC restaurant economics. The shop does not become a different shop. It becomes the same shop with thirty thousand dollars more on the bottom line.

The slice shop ledger, one full year
Slices, marketplace
54,750
~150 per day
Commission @ 15%
$73,710
capped per LL6/2021
Promoted listings
$8,400
uncapped, voluntary
Total bill
~$97k
incl. payment processing
VI.The city in 24 hours

New York eats in three waves: midtown lunch, borough dinner, late-night counter.

A modeled view of when the city orders, distilled from operator interviews and trade reporting. Breakfast is essentially a non-event. The lunch wave runs midday in Manhattan finance and Midtown. Dinner is the residential-borough story. After 11pm, the city belongs to slice shops, halal carts, and a few late-running bistros.

0%2.5%5%7.5%10%12a3a6a9a12p3p6p9pLunch peak (Manhattan)Dinner peak (boroughs)Late night: slice + halal
12pm to 2pm
Manhattan finance and Midtown lunch. The corporate catering window, the salad-bowl rush, the office-tower delivery surge. A high-AOV moment for chef-driven operators.
6pm to 8pm
Residential borough dinner. Brooklyn, Queens, the Bronx, Upper West Side. The largest single block of orders in a 24-hour day, by a wide margin.
11pm to 2am
Late-night counter food. Slice shops, halal carts, 24-hour diners. The window in which a multilingual Voice AI doing curbside callout earns its keep most clearly.

The pattern is the operational reason missed calls cost a typical NYC restaurant between $1,800 and $7,400 a month, depending on segment. Phone volume is highest in the bands the staff is least available to answer.

VII.A composite Q&A

Four questions for an NYC restaurant accountant who has seen every version of this conversation.

The interview below is a composite, drawn from public reporting in Eater NY, Grub Street, and operator interviews. It is not an attribution to a specific real accountant. We have synthesized it this way to keep the reporting honest and to keep the framing useful.

Q1.

Is the Local Law 6 fee cap actually working?

Half. The 15 percent commission cap and the 5 percent other-fees cap are real and enforced by the NYC Department of Consumer and Worker Protection. Operators report that the per-order math improved meaningfully after August 2021. The capped components of the marketplace bill went from a 25 to 32 percent range down to a 15 to 20 percent range, which on a typical slice shop's $300,000 a year of marketplace gross is roughly $35,000 a year in restored margin.

What the cap did not do is govern advertising spend on the marketplaces, which is a separate budget line item the operator pays voluntarily and which has grown roughly in step with the cap's enforcement. A typical NYC operator I see now spends $800 to $2,400 a month on promoted listings, which is not capped, is not reported on the receipt, and is what restored the marketplace's pricing power even after the law took effect. The Council solved the front-of-house version of the problem. The back-of-house version, which is the discretionary ad spend, is the version that matters now.

Q2.

What does the typical NYC operator's marketing budget look like in 2026?

Five years ago an operator told me they had no marketing budget, which usually meant they had a $200 a month Yelp ad they had forgotten to cancel and a part-time cousin running their Instagram. Today the same kind of operator has roughly 4 to 6 percent of gross revenue going to marketing, half of it to marketplaces, the other half split between Google Ads, paid Instagram, and some experimentation with TikTok and short-form video.

The fastest-growing line item I see is direct-channel marketing, which is operators paying to drive customers to their own ordering page instead of to a Seamless listing. That line item barely existed in 2021. In 2026 it is between 20 and 35 percent of total marketing spend for the operators who have built a direct channel that actually converts. The ones who have not built that channel still see direct-channel marketing as a luxury they cannot afford, which is exactly backward.

Q3.

Why don't more NYC operators leave the marketplaces entirely?

Because the marketplace customer acquisition function, even at 20 percent capped fees, is not zero. A new customer who finds the shop on Grubhub at lunch and comes back twice in person at dinner is worth the fee. The unit economics of the first order are bad. The unit economics of the third and fourth orders are good, if the operator has a way to convert that marketplace customer into a direct customer.

The operators who leave marketplaces successfully are the ones who already have a strong walk-in and reputation business and treat marketplaces as a small leak. The operators who fail when they leave are the ones who replaced their own marketing function with the marketplace and never rebuilt it. Voice AI, direct online ordering with branded SMS, loyalty programs, Google Business profile optimization, these are not luxuries. They are the customer acquisition function the marketplace was renting to the operator.

Q4.

What's a realistic 60-day plan for an NYC operator to shift orders direct?

Week 1 to 2: launch a branded direct ordering site on a flat monthly fee platform, ideally one that integrates with the operator's existing POS so the kitchen does not have to relearn the workflow. Put a QR code on every receipt, every to-go bag, every table tent. Set the direct site as the primary link in the Google Business profile, on Instagram, and on the operator's printed menus.

Week 3 to 6: enable Voice AI to catch missed calls during the 11:30 to 2:30 lunch rush and the 6 to 9 dinner rush. A typical NYC operator misses 20 to 40 percent of inbound calls during peak hours, and roughly half of those callers will order if they reach a human or a competent AI. Add SMS marketing to the direct customer list at the rate of one campaign every two weeks, not more.

Week 7 to 8: introduce a small direct-only loyalty perk, a $4 credit on the second order or a free side after five orders. Measure the conversion rate from marketplace-only customer to direct-only customer. The realistic target is 25 to 40 percent of the existing marketplace customer base shifted to direct by day 60, which is what restores enough margin to fund the next phase.

VIII.How DirectOrders fits NYC

At more than 200 orders a month, a flat fee outperforms a percentage. Always.

We do not run a marketplace. We do not compete with DoorDash on its terms. The argument we are making to an NYC operator is narrower and more boring than that. It is: for any restaurant doing more than roughly 200 orders a month at an average ticket above twenty dollars, a flat monthly fee for direct ordering software outperforms any percentage-based commission structure that exists in 2026. The math is not close.

DirectOrders is $249 a month for a single location, $349 for a small group. That fee gets a branded ordering site indexed in Google and AI search, multilingual Voice AI for missed calls (Cantonese, Spanish, Mandarin, English at minimum, additional dialects on request), Uber Direct and DoorDash Drive dispatch integration covering all five boroughs at flat per-order delivery cost, same-day Stripe payouts so Friday's Restaurant Week revenue arrives Saturday rather than the following Wednesday, and POS integration with the systems most NYC operators already run on.

Run the breakeven. At 200 orders a month, average ticket $28, the operator's total monthly gross is $5,600. Marketplace fees on that volume, including capped commission, capped other fees, modest promoted listings, and payment processing, run roughly $1,400 a month. DirectOrders flat fee plus Stripe processing on the same $5,600 gross is roughly $409. Net monthly savings: approximately $990. Annual savings at constant volume: approximately $11,900 per location. That is the floor. The savings scale linearly with volume, while the flat fee does not.

The non-financial argument matters too. A marketplace customer is the marketplace's customer, not the operator's. The operator does not own the email address, the phone number, the reorder history, or the loyalty graph. A direct customer is the operator's customer in every meaningful sense. The operator decides when to email them, what to text them, which loyalty perk to offer them on the second order, how to bring them in for a Saturday brunch. The customer relationship, which is the actual asset of a restaurant business, is portable across software platforms and not portable across marketplaces. That asymmetry is the long-run argument.

We sell software, not magic. A NYC operator doing 30 orders a month is too small for our pricing to make sense, and we will tell them so on the first call. A NYC operator doing 800 orders a month who has not yet built a direct channel is, in our view, leaving roughly $50,000 a year on the table at minimum, possibly closer to $90,000, and we can usually compress that recovery into a 60 to 90 day rollout that does not require them to change POS, hire staff, or rebuild their kitchen workflow. The fee cap moved the line. Direct ordering moves the operator off the line.

Branded ordering site
Indexed in Google and AI search, on the operator's own URL.
Read about ordering
Multilingual Voice AI
Cantonese, Spanish, Mandarin, English. Picks up the calls the kitchen misses.
Read about Voice AI
Uber Direct + DoorDash Drive
Flat-cost dispatch across all five boroughs, on the operator's brand.
Read about delivery
IX.What to do next

Two ways to start, neither of them dramatic.

If you are an NYC operator and you have read this far, the next move is small. There are two reasonable doors.

The first is a 30-minute walkthrough on a video call. We will look at the operator's current marketplace mix, talk through the specific math for the segment they fall into, and show what a branded ordering site indexed for the relevant Google and AI search terms looks like. No deck. No pitch. We will use the operator's own POS data if they want, or our composite if they prefer. Book a walkthrough.

The second is the pricing page, which is the answer for an operator who wants to read the numbers before they speak to a person. The flat fee structure is plainly stated, the included features are listed, the breakeven point at typical NYC volumes is documented. Read the pricing.

The fee cap is the floor. The direct channel is the lever. Whatever else 2026 does to NYC restaurant economics, the operator who owns their distribution channel is the operator who controls their own bottom line.

Take your NYC restaurant off the percentage tax
Live in 2 hours or we white-glove the launch for free. $249 per location per month. Same-day Stripe payouts.
Book a walkthroughRead pricing
Last updated May 11, 2026. All operator profiles in this piece are composites drawn from public reporting and our own conversations with NYC restaurant operators; specific neighborhoods, zip codes, and economic profiles are illustrative, not attributions to specific named businesses. Where numbers are modeled rather than reported, we have qualified them as approximate or as typical for the segment. The fee cap mechanics are documented at the NYC Department of Consumer and Worker Protection.