State of Restaurant Tech 2026.

1. Executive summary

Three findings survive a source-by-source re-check on October 12, 2026, each of them something a company publishes on its own page and read by us on that date. The fourth item is the correction itself.

1. The three US marketplaces publish three different ladders, and the floor is lower than the usual shorthand. DoorDash publishes 15% (Basic), 25% (Plus), 30% (Premier) on delivery and 6% on pickup. Uber Eats publishes 20% (Lite), 25% (Plus), 30% (Premium). Grubhub publishes a marketing commission of 5% (Basic), 15% (Plus), 20% (All-access), with delivery by its own fleet charged on top. “15 to 30%” is DoorDash’s delivery ladder. It is not Grubhub’s, whose Basic plan is 5%, and not Uber Eats’s, whose marketplace plans start at 20% nationally and lower in select cities. The three also treat card processing differently, so this report quotes each ladder as published and does not blend them into one number.^[1][2][3]
2. Two card processors print their in-person rates. Stripe publishes 2.9% + 30¢ for domestic cards online and 2.7% + 5¢ in person. Square publishes 2.6% + 15¢ in person on its Free plan. Those are the only two card-present rates in this report because they are the only two card processors whose pricing pages we read for this edition.^[35][49]
3. iFood publishes a take of 15.2% to 26.2% before any monthly fee. That is 12% (Plano Básico), 23% (Plano Entrega) plus a 3.2% fee on orders paid through iFood, as printed on its plan cards. The FAQ on the same page prints 3.5% for that fee, which would make the span 15.5% to 26.5%. We print both.^[13]
4. Most of the survey statistics in earlier editions could not be found where we said they were. The adoption series for direct online ordering, the series for AI use, the closure rates, the closure-driver ranking and most company-size figures are withdrawn. The correction note lists every one.

2. Marketplace economics in 2026

The three major US delivery marketplaces, DoorDash^[1], Grubhub^[2] and Uber Eats^[3], each publish a three-plan merchant structure. The three ladders are not the same ladder. It is tempting to summarise all three as “15 to 30%”, and that fits only one of them. DoorDash runs 15% (Basic), 25% (Plus), 30% (Premier) on delivery. Uber Eats runs 20% (Lite), 25% (Plus), 30% (Premium) on its marketplace. Grubhub’s published plans are a marketing commission of 5% (Basic), 15% (Plus), 20% (All-access), a different fee with a different base. A restaurant comparing quotes against a made-up 15% floor will read a Grubhub Basic quote, at 5%, as a discount it is not.

What you cannot add up. A single all-in “effective take rate” cannot be built from these pages, and this report does not print one. DoorDash states that its commission covers credit card processing.^[1] Uber Eats lists payment processing among the benefits of its Lite, Plus and Premium plans, and then prints exceptions in a footnote for select cities: in New York City, for example, its Lite plan carries a separate 3% fee for credit card processing. Uber Eats also prints a processing fee for its Webshop product, a different channel: 2.5% plus $0.29 per order.^[3] Grubhub’s page says its rates do not include payment processing fees, and prints no processing rate.^[2] Three platforms, three treatments, and no single number to add.

In the published plans, DoorDash lists^[1] 15% (Basic), 25% (Plus), 30% (Premier) on delivery, with pickup at 6% on every plan for merchants whose pickup menu prices match in-store prices. DoorDash says these plans are for restaurants with 75 or fewer locations in the United States. Orders placed through the restaurant’s own Online Ordering channel are commission-free and carry a payment processing fee (DoorDash merchant pricing page, read October 12, 2026).^[1] Grubhub’s national schedule^[2] is a marketing commission of 5% (Basic), 15% (Plus), 20% (All-access). Delivery by Grubhub’s fleet is a separate fee that the page says starts at 10%, with a note that pricing in select markets may differ. The page says its rates do not include payment processing fees. Uber Eats lists^[3] 20% (Lite), 25% (Plus), 30% (Premium) on its marketplace, 15% when the restaurant delivers with its own staff, and pickup at 7% where in-store pricing is validated (10% where it is not). Uber Eats adds that its Lite plan carries a lower marketplace fee in select cities.

Regulated markets. Grubhub runs a separate New York City fee schedule^[18] next to its national one, and says on that page that the City Council passed a law in 2025 changing the third-party delivery fee cap. Earlier editions also described the New York City and California laws themselves. The city page we linked no longer exists and the California page refused retrieval, so those descriptions are withdrawn. We make no claim here about what either law currently requires.

Our reading, not a statistic. A marketplace order costs the restaurant a published percentage that a direct order does not. That is arithmetic on the rates above. How many operators are moving customers to direct ordering, and how fast, is a survey question, and we no longer print an answer to it (see §4). For the buying checklist we use ourselves, see our RMS buying guide for 2026.

3. POS vendors: what each one publishes about its size

Earlier editions printed a size figure for each of five POS vendors: Toast, Square, Clover, Lightspeed and TouchBistro. On October 12, 2026 we could read exactly one of the five at the page we had linked.

TouchBistro. Its About page^[26] states “16,000+ restaurants powered” and “64,000+ terminals deployed”. Earlier editions printed a larger restaurant count for TouchBistro. The page does not show that number today, so we print what it shows.

Toast, Square, Clover, Lightspeed. The figures we had printed for these four are withdrawn. The Toast and Fiserv investor pages refused retrieval, the Block investor address we linked no longer resolves, and the Lightspeed and Toast pages that did load carry no location count. The figures may well be right. We could not read them at the source on the day we checked, and this report prints only what we could read.

Why this section has no market-share chart. It never had one, and with four of the five figures withdrawn there is nothing to chart. We do not estimate market share.

For vendor-by-vendor pricing with dated sources, see our comparison page, Toast alternatives and Square alternatives.

4. Online ordering adoption: withdrawn

Earlier editions printed a year-by-year series for the share of US restaurants taking direct online orders (Figure 2), and attributed it to the National Restaurant Association and to Toast. We withdraw the figure, the series and every percentage in this section.

The reason is simple. The National Restaurant Association page^[6] we linked is the sales page for its 2026 State of the Restaurant Industry report, which it describes as free to members and $349 for nonmembers. None of the percentages we printed appears on that page. Toast’s report page refused retrieval. We also printed a market-size projection and a contribution-margin claim that linked to a homepage and a resource index, where neither appears.

A statistic we cannot show you at its source is not one we should print, so this section now has no numbers. If you hold one of those reports and can point us to the page, tell us through our contact page and we will restore the figure with its citation.

5. AI ordering: withdrawn

Earlier editions printed a series for the share of US operators using or piloting AI (Figure 3), named specific restaurant chains running voice ordering, and named the model providers behind chat ordering. We withdraw all of it.

The adoption percentages were attributed to the same National Restaurant Association page as §4 and do not appear on it. The customer story we linked for a drive-thru deployment now returns a not-found page. The model-provider links went to product pages that say nothing about restaurants. The trade publications we cited were linked at their homepages, not at an article.

We sell restaurant software ourselves, which is one more reason to hold this section to the same standard as the rest. It will return when it can cite documents, not homepages.

6. Payment processing: what is published

This section prints the card rates we read at the source for this edition, and nothing else. They come from two processors.

Stripe. Stripe’s pricing page^[35] lists 2.9% + 30¢ per successful transaction for domestic cards online and 2.7% + 5¢ for in-person cards and wallets. Stripe Connect^[36] is the product Stripe sells to platforms to onboard and pay out their users. FoodyOS uses it for online card payments at checkout in the United States, which settle to the restaurant’s own Stripe account.

Tap to Pay. Stripe’s Tap to Pay page^[39] says in-person payments can be accepted directly on compatible iPhones and Android devices with no extra hardware. Stripe’s pricing page lists an extra 10¢ per authorization for Tap to Pay on top of the in-person rate.^[35] Earlier editions also printed a launch year and a claim about how many new merchants skip a card reader. Neither appears at the pages we linked, and both are withdrawn.

Square. Square’s fees page^[49] lists in-person rates of 2.6% + 15¢ on its Free plan, 2.5% + 15¢ on Plus and 2.4% + 15¢ on Premium. That page covers Square payments in general and is not specific to restaurants.

Other processors and POS vendors. We print no other processor’s card rate, because we did not read their pricing pages for this edition. An earlier edition made statements about what Toast, Lightspeed and Clover do not publish. We did not re-check those, so they are withdrawn too.

Our reading, not a statistic. When the POS vendor is the processor, changing POS also means changing processor. When the restaurant holds its own merchant account, it does not. Which matters more to you is a business decision. For how that works in practice, see why owning your Stripe account matters for direct ordering.

7. Restaurant closures: withdrawn

Earlier editions said that a popular statistic about first-year restaurant failure was wrong, and printed replacement survival rates attributed to the Bureau of Labor Statistics. They also printed a ranked list of closure drivers attributed to the National Restaurant Association, and a description of openings and closings by city.

We withdraw all of it. The Bureau of Labor Statistics page we linked is the overview of its Business Employment Dynamics program and shows no survival rate for restaurants. The ranking does not appear on the association page we linked. The two pages we cited for city-level trends now redirect to an index and to a data-licensing page.

That includes our own correction of the popular statistic. We asserted it was wrong and did not show you the table that proves it, which is the same mistake in the other direction. A future edition will cite a table you can open, or say nothing.

8. Brazil: what iFood publishes

Scale. On its institutional site^[43], iFood says the platform registers more than 180 million orders a month, and that in March 2026 it had more than 500 thousand partner establishments. That second figure counts restaurants together with markets, pharmacies, pet shops and other stores, so it is not a restaurant count. Earlier editions printed smaller figures for both and called the second one restaurants.

iFood economics. iFood publishes two restaurant plans on its partner pricing page^[13]: 12% (Plano Básico), 23% (Plano Entrega). The cheaper plan is for restaurants that deliver with their own couriers, the dearer one for delivery by an iFood courier, and the page tells restaurants to check whether that plan is available in their region. On top of the commission, the plan cards print a 3.2% fee on orders paid through iFood, and a monthly fee of R$ 110 on Básico (R$ 150 on Entrega) charged only once monthly sales pass R$ 1.800.

That puts the published take between 15.2% and 26.2% of the ticket before the monthly fee and before any optional promotion. One conflict on iFood’s own page belongs next to that figure: the FAQ on the same page prints 3.5% for the payment fee, not 3.2%. On the FAQ figure the span is 15.5% to 26.5%. We use the lower one in our own comparisons because it is the one that favours iFood. Two more things in iFood’s favour belong here, both from the same page:^[13] 0% commission below 30 orders/month on both plans, published as a limited-time offer. No lock-in: 'Cancele o plano quando quiser.' (iFood partner pricing page, read October 12, 2026.) The plan-by-plan detail, in Portuguese, is in Taxa do iFood em 2026.

Withdrawn from this section. Earlier editions printed a messaging-penetration figure for WhatsApp in Brazil, said a regional tracker had logged restaurants moving orders from iFood to WhatsApp, said an industry association had tracked what operators pay, and described iFood’s market share. The report we linked for the first does not mention Brazil on the page, the tracker’s site no longer resolves, and the other two are not at the pages we linked. This report also makes no claim about Rappi’s fees.

For how the two channels compare for one restaurant, see pedidos pelo WhatsApp vs iFood.

9. Methodology

The rule for this edition. A figure is in this report only if we could read it, on October 12, 2026, at the page we link next to it, on the website of the company or agency that publishes it. Nothing is estimated, modelled or carried forward from an earlier edition.

What the re-check found. We listed every factual claim in the previous edition, which was last revised in August 2026, 82 in all, and fetched each source again. 30 were confirmed as printed. 4 had changed at the source and are updated here. The remaining 48 could not be confirmed: the linked page did not contain the claim, the link was dead, the site refused retrieval, or the claim had no link at all. All of those are withdrawn. Of the 49 distinct sources, 35 loaded.

What went wrong before. Earlier editions linked many statistics to a homepage or a hub and not to the document that carried the number. Some of those statistics came from a report that is sold, which also contradicted our own statement that we used no paywalled research. A link that does not show the reader the number is not a citation. That is the standard this edition applies, and the reason it is shorter.

Marketplace rates. Every commission and fee in §2 and §8 is a rate the platform states on its own merchant pricing page, quoted per plan, cheapest plan first. The two iFood scale figures in §8 come from its institutional site. No effective take rate is computed, because the three treat card processing in three different ways (see §2), and there is no single published number to add.

Who wrote this. FoodyOS sells restaurant software, including direct ordering that competes with the marketplaces described here. Read the report with that in mind, and follow the links.

License. This report is published under Creative Commons CC BY 4.0. You may quote, republish or redistribute it with attribution to FoodyOS and a link back to https://foodyos.com/state-of-restaurant-tech-2026. If you spot a number that has moved, tell us through our contact page and we will correct it and say so.

10. Sources

Sources still cited in this edition are linked, and each was read on October 12, 2026. Sources that earlier editions cited and this edition no longer relies on keep their number, so older references to this list still resolve, and are marked withdrawn with the reason.

1. DoorDash, merchant fees and pricing (15% (Basic), 25% (Plus), 30% (Premier) delivery, 6% pickup)
2. Grubhub, pricing and fees (5% (Basic), 15% (Plus), 20% (All-access) marketing commission; delivery fees start at 10%)
3. Uber Eats, merchant pricing (20% (Lite), 25% (Plus), 30% (Premium) marketplace; 15% self-delivery; 7% pickup)
4. Withdrawn October 2026: Toast investor news releases. Reason: the site refused retrieval.
5. Withdrawn October 2026: Toast Investor Relations homepage. Reason: no location count on the page.
6. National Restaurant Association, State of the Restaurant Industry 2026 (sales page). Cited only for the fact that the report is sold. Every statistic earlier editions attributed to it is withdrawn.
7. Withdrawn October 2026: National Restaurant Association research hub. Reason: the claim was not on the page.
8. Withdrawn October 2026: Anthropic product page. Reason: says nothing about restaurants.
9. Withdrawn October 2026: OpenAI API reference. Reason: says nothing about restaurants.
10. Withdrawn October 2026: Google AI developer page. Reason: says nothing about restaurants.
11. Withdrawn October 2026: Abrasel homepage. Reason: the claim was not on the page.
12. Withdrawn October 2026: Abrasel news index. Reason: the claim was not on the page.
13. iFood, planos para restaurantes (12% (Plano Básico), 23% (Plano Entrega); taxa de pagamento 3,2% nos cartões de plano, 3,5% no FAQ)
14. Withdrawn October 2026: Restaurant Business Online homepage. Reason: a homepage, not an article.
15. Withdrawn October 2026: Nation's Restaurant News homepage. Reason: a homepage, not an article.
16. Withdrawn October 2026: New York City consumer agency page on third-party delivery. Reason: the page no longer exists.
17. Withdrawn October 2026: California legislative page for a delivery-fee disclosure bill. Reason: the site refused retrieval.
18. Grubhub, New York City pricing and fees
19. Withdrawn October 2026: Toast Restaurant Success Report. Reason: the site refused retrieval.
20. Withdrawn October 2026: Toast quarterly results page. Reason: no figures on the page.
21. Withdrawn October 2026: Block investor relations. Reason: the address no longer resolves.
22. Withdrawn October 2026: Block quarterly results. Reason: the address no longer resolves.
23. Withdrawn October 2026: Fiserv investor relations. Reason: the site refused retrieval.
24. Withdrawn October 2026: Clover About page. Reason: the claim was not on the page.
25. Withdrawn October 2026: Lightspeed investor relations. Reason: no location count on the page.
26. TouchBistro, About us (16,000+ restaurants powered)
27. Withdrawn October 2026: Datanyze POS market share page. Reason: the site refused retrieval.
28. Withdrawn October 2026: BuiltWith POS trends page. Reason: the site refused retrieval.
29. Withdrawn October 2026: G2 Restaurant POS category. Reason: the site refused retrieval.
30. Withdrawn October 2026: Restaurant365 resources index. Reason: the claim was not on the page.
31. Withdrawn October 2026: Restaurant Dive homepage. Reason: a homepage, not an article.
32. Withdrawn October 2026: Google Cloud customer story. Reason: the page now returns not found.
33. Withdrawn October 2026: QSR Magazine homepage. Reason: the site refused retrieval.
34. Withdrawn October 2026: Modern Restaurant Management homepage. Reason: a homepage, not an article.
35. Stripe, pricing
36. Stripe Connect
37. Withdrawn October 2026: Apple Tap to Pay developer page. Reason: the claim was not on the page.
38. Withdrawn October 2026: Square blog index. Reason: the claim was not on the page.
39. Stripe Terminal, Tap to Pay
40. Withdrawn October 2026: Bureau of Labor Statistics, Business Employment Dynamics overview. Reason: no restaurant survival rate on the page.
41. Withdrawn October 2026: National Restaurant Association economics index. Reason: the claim was not on the page.
42. Withdrawn October 2026: Yelp economic reports address. Reason: now redirects to a data-licensing page.
43. iFood, institutional site (orders per month, partner establishments)
44. Withdrawn October 2026: DataReportal Digital 2026 Global Overview. Reason: the claim was not on the page.
45. Withdrawn October 2026: Regional digital tracker. Reason: the address no longer resolves.
46. Withdrawn October 2026: Bureau of Labor Statistics, Industry at a Glance. Reason: was listed and never cited.
47. Withdrawn October 2026: Stripe customer stories. Reason: was listed and never cited.
48. Withdrawn October 2026: DoorDash payments page. Reason: the page no longer exists.
49. Square, payment fees