Toast's Q3 Print, the Uber Partnership, and the Drive-Thru Question No Analyst Will Ask

I read the Toast Q3 print on the Tuesday night it dropped, sitting at my kitchen table with a cold mug of coffee and the print-out marked in red pen. ARR through $2.0B. Net locations up 7,500. Adjusted EBITDA of $176M and a full-year guide nudged to $610–620M. By any reasonable yardstick this is the print of a company executing — and yet the stock has spent the back half of the year doing something between drifting and sulking, depending on the day you check. My contrarian read: the Q3 numbers and the November 3 Uber announcement matter much less than what Toast is about to launch in the first half of 2026. And not a single sell-side analyst is asking the question that actually decides the next $10 of the stock.

Let me walk through why.

H2: The Q3 print, decoded with a pen

Start with the cleanest version of the math from the November 4 release. Toast reported Q3 2025 ARR of $2.0B, up 30% year-over-year, with adjusted EBITDA of $176M and a full-year EBITDA guide raised to $610–620M. Net new locations were +7,500 in the quarter, bringing the installed base to 156,000.

The interpretation analysts will write is “in-line with consensus, modest beat on EBITDA, location adds healthy.” That is technically correct and analytically lazy. Let me do the math they don’t.

Take the ARR run-rate of $2.0B and divide by 156,000 locations: that’s roughly $12,820 of ARR per location, annualized. A year ago, on the Q3 2024 print, the comparable figure was closer to $10,800. That’s an 18% lift in ARR-per-location in twelve months, while the location count itself grew on the order of 25%. Said differently: Toast is monetizing each restaurant harder and adding more of them — and most of that ARR-per-location lift is not coming from price. It’s coming from attach: fintech, payroll, the Toast Capital book, the marketplace.

The 30% ARR growth headline obscures something more interesting in the unit economics. The take is creeping up, location by location, in a way that compounds. If you model another 15% ARR-per-location lift through 2026 on top of, say, 22% location growth, you get to ARR somewhere around $2.8B by Q4 2026 without anything new — no drive-thru, no enterprise wins, no Uber halo. That’s the base case. The bull case is what comes next.

H2: The Uber deal is a distribution story, not a margin story

On November 3, Toast and Uber announced what both companies described as a strategic partnership making Uber the preferred delivery marketplace for Toast restaurants globally. The framing matters. “Preferred” is not “exclusive.” Toast restaurants will still be able to plug into DoorDash, Grubhub, and any number of regional marketplaces. What Toast gains is a default funnel — when a restaurateur opens the marketplace tile inside Toast, Uber Eats is the path of least resistance.

The market reaction was muted. I think the market is wrong, but in a subtle way.

Read this as a margin story and you’ll be disappointed: rev-share on referred delivery orders is real but small relative to a $2.0B ARR base. Read this as a distribution story and it’s much more interesting. Toast just turned its 156,000 locations into a default surface for the largest non-Amazon last-mile network in the U.S., and in exchange got product placement in Uber Eats that competing POS systems do not enjoy. The TAM Toast is targeting in 2026 — full-service enterprise chains, food trucks, ghost kitchens, and the drive-thru segment — all benefit disproportionately from a single, default delivery integration.

The risk on the other side is that Uber is using Toast as a wedge against Square Loyalty and DoorDash’s POS ambitions, and one quarter from now we get a counter-announcement from Square or Block that reframes the field. I don’t dismiss that. But the structure of this deal — preferred, global, multi-year — is not what you sign if you think you’re going to get pre-empted in ninety days.

H2: The drive-thru question no analyst will ask

Here is the question I would ask on the next earnings call if I were on the line.

Drive-thru. When. What’s the SKU look like. How many of the 156,000 existing locations are already drive-thru-capable on the hardware side, and how many are pure greenfield. And — critically — what’s the take rate on a drive-thru location versus a sit-down location, given that ticket volumes are 3–4x higher.

Restaurant Business reported in late October that Toast is preparing a drive-thru product, and the company has been signaling this in conference appearances and on the analyst circuit for most of the second half of 2025. I’d frame the most likely launch window as early 2026, possibly tied to a major industry conference. The reason this matters: the U.S. drive-thru segment is structurally a quick-service-restaurant duopoly between NCR Voyix and Oracle Micros at the top, with a long tail of regional players. Toast has zero share. If Toast can put a credible drive-thru SKU into the market in 2026, the addressable location count for the company roughly doubles in a single product release.

The math gets aggressive fast. Quick-service drive-thru locations in the U.S. number somewhere north of 200,000 — McDonald’s, Wendy’s, Taco Bell, Chick-fil-A, and the long tail of regional chains. Even capturing 5% of that base over three years adds 10,000 net locations, each with materially higher ticket throughput than Toast’s current full-service-skewed installed base. At a generous $20,000 ARR-per-location assumption for drive-thru (versus $12,800 today), that’s an incremental $200M of ARR — and that is the bear case for what’s coming.

The sell-side won’t ask this question because the answer doesn’t fit into a 12-month price target. It fits into a 36-month one. My job — investor, not analyst — is to underwrite the 36-month one.

H2: Toast IQ is a story, not a line item — yet

The other thing I want to flag is Toast IQ, the AI assistant the company introduced at Toast Big Conversations 2025 and has been slowly seeding into the customer base. Restaurant Technology News framed Toast’s positioning around AI-driven operations as the next phase of competition in the POS-and-restaurant-tech market, and that framing is exactly right — but it cuts both ways.

Toast IQ is, as of the Q3 print, not a separately disclosed revenue line. It is bundled into the general SaaS revenue and, more importantly, into the value proposition of the platform. That’s a deliberate choice. Toast is using IQ to defend the existing ARR-per-location uplift (i.e., make payroll and Toast Capital sticky by wrapping them in an AI layer) rather than monetizing it as a standalone SKU. I think that’s the right call for 2026 — you want adoption breadth before you want a per-seat invoice — but it does mean investors looking for a clean “AI revenue” disclosure won’t find one. They’ll find IQ in the qualitative commentary and in retention metrics they have to back into.

I’ll write more about IQ specifically in a forthcoming May piece — see our earlier coverage of the Toast IQ launch — but the short version is: don’t model IQ as a revenue contributor in 2026. Model it as a retention and gross-margin defender. That’s where it earns its keep.

H2: Putting it together — what I’m watching into Q1

Pulling the threads: Q3 ARR of $2.0B and EBITDA of $176M is a clean execution print at scale. The Uber deal is a distribution catalyst whose value the market is under-pricing. Drive-thru is the biggest unmodeled 2026 catalyst in the entire restaurant-tech complex, and Toast is the only horizontal POS player with the gross-margin profile and the balance sheet to attack it credibly. Toast IQ is a quiet defender of the ARR-per-location story, not yet a revenue line.

The stock wobble since August? Some of it is multiple compression that hit the whole growth-software cohort. Some of it is fair caution on whether the 2026 catalysts arrive on schedule. None of it, in my read, reflects the underlying compounding inside the franchise. I’m building toward an upcoming desk review of the full Toast investment case over the next several weeks; the framework I keep coming back to is that ARR-per-location is the single most informative ratio in the model, drive-thru is the call option that resets the TAM, and IQ is the moat-deepener that keeps the ratio compounding.

What I’m watching into Q1: the cadence of Uber-attributable order volume in Toast’s marketplace reporting (qualitative is fine, I don’t need a SKU); any drive-thru product announcement and the chain logos that anchor it; and the trajectory of fintech attach, which is the cleanest read on whether IQ is doing the retention job I think it is.

Pour the second cup of coffee. The print was good. The thesis is the next print.

— Oliver writes The Bottom Line for TableTransfers. Tips: [email protected].