Q1 2026 valuation reset: why DASH is down 42% and TOST is the new comparable

It’s a Friday morning and I am reading the tape with the second espresso in front of me and the half-finished bagel I keep telling myself I’ll eat. The screen has the usual rotation — DASH, TOST, OLO (delisted, but I keep it for the muscle memory), WIX, SHOP — and the line that keeps catching my eye is the one I drew in red across my notebook two weeks ago and underlined twice. DoorDash from $285.50 to $165-and-change. Toast guides up. The multiple does not move. I stared at that line for a long time before I wrote the next one underneath it, the one that is the whole article: the case against the AI premium has finally arrived in the tape, and Toast is the cleaner way to play it.

That’s the thesis. Let me show you the math.

The DASH drawdown is the loudest data point in restaurant tech right now and it is being widely misread. The TIKR piece that landed last week — DoorDash stock is down 42% from its highs, here’s why the 2026 selloff could set up 2030 returns — frames it as a multiple compression on a name that is still growing. The framing is right but the implication most readers will take from it is the wrong one. The market is not selling DASH because growth has broken. The market is selling DASH because management has chosen to invest through the margin line, and the investing-through-margins decision has reset what investors thought they were buying. Toast, on the other side of the ledger, has done the opposite — guided to a clean 21 to 23 percent recurring gross profit print for FY26 in the Q4 release back in February, held the line on the $790 to $810 million EBITDA range, and is about to ship a drive-thru AI platform on April 14 that the sell-side has barely started to model. The two stories are mirrors of each other and the relative value trade is the easier read of the two.

I want to walk through the DASH drawdown thesis, the TOST mirror, what is actually priced into each name on a sum-of-parts and EBITDA basis, and end on where I would put the entry today.

The DASH drawdown: investing through margins is the whole story

Start with the 42 percent number, because it is the one everyone is anchored on. DASH closed yesterday somewhere in the $165 handle against a 52-week high of $285.50 that printed in the fall. That is a $120-per-share drawdown on a name that, by every operating metric I can pull from the last two quarterly prints, is still growing top-line in the high teens and still gaining share against Uber Eats in the categories that matter. The drawdown is not a growth story. The drawdown is a guidance story.

The single quote I keep coming back to is the one TIKR pulled from the Q3 2025 call — Tony Xu telling the Street that DoorDash would be spending “several hundred million dollars more” through 2026 on a combination of tech replatform, autonomous-delivery R&D, and the Deliveroo integration cost-stack. Several hundred million dollars more. That is a phrase that, in a sell-side model, lands as a 150 to 250 basis-point haircut to the next-twelve-months EBITDA margin. On a $11 billion revenue base that is $200 million-ish of EBITDA that comes off the page. On a name that was trading at roughly 30 times forward EBITDA at the high, that is a $6 billion enterprise-value adjustment before anyone touches the multiple itself. Then the multiple compresses, because investors look at a company that is investing through margins and decide that the right comp set is no longer the asset-light marketplace cohort but the platform-build cohort, and the platform-build cohort trades at a discount.

Flag the interpretation. The 30-times EBITDA figure I am using is the rough mark I have been carrying since the fall and is not from a fresh sell-side note; readers should treat the multiple math as directional. The point is the direction of the rerating, not the precise terminal multiple.

The companion observation — and this is the one I want to draw a line under — is that the Deliveroo integration cost-stack is a one-time number that the market is treating as recurring. That treatment is what creates the dislocation. Once Deliveroo is integrated and the synergies start to print, the cost line steps down. The autonomy R&D is harder to mark, but the DoorDash/SevenRooms transaction and the ongoing tech replatform sit on top of a delivery-platform business that has already proved it can flex margins when the spend ends. The market is pricing the spend as permanent. I do not think it is.

And then there is the ALSO Pass story, which my Wednesday note walked through in detail — DoorDash leaning into the loyalty/membership flywheel as a take-rate stabilizer at the exact moment the marketplace category is being asked to defend its take. Membership economics push the contribution-margin line up over a multi-quarter horizon. They do not show up in the next four prints. So they will not save the stock from the Q2 burn, but they are part of why the 2027 number is going to look very different from the 2026 number.

The TOST mirror: guidance discipline that is not yet rerating

Now flip the page and look at Toast. The Q4 2025 print landed in February and the company guided FY26 to recurring gross profit growth of 21 to 23 percent and EBITDA in the $790 to $810 million band. ARR is set to come up roughly 26 percent off the Q4 exit. Those are not investing-through-margin numbers. Those are the numbers of a company that has decided it is going to print the operating leverage and let the AI optionality compound on top.

The Simply Wall St narrative on the AI drive-thru platform marks a fair value of $36.75 against a stock that is trading materially below it. The drive-thru launch on April 14 is the moment the market gets its first concrete look at what Toast’s AI investment cycle is actually producing. The platform is being rolled out into a quick-service segment Toast historically did not address — the drive-thru cohort is roughly 70 percent of QSR transaction volume and has been a Square-and-NCR fortress for a decade. If Toast can take even a low-single-digit share of new drive-thru wins in the back half of 2026, the recurring gross profit guide is conservative.

And yet the multiple does not move. The stock is, on my back-of-envelope, trading at roughly 18 to 20 times forward EBITDA against a peer group that is happy to pay 25 to 30 for SaaS-plus-payments names that are growing at half the rate. Flag the multiple math as interpretation — I am pulling forward EBITDA from the company guide and the share count from the Q4 release; the precise multiple shifts depending on whether you mark the diluted or basic count, and I am using diluted.

The asymmetry is the trade. DASH is being marked down on a guidance choice (invest through margins) that is reversible in 18 months. TOST is being held flat on a guidance choice (print the leverage) that the market has not yet rewarded because the AI optionality has not yet shown up in a product launch the sell-side can model. The drive-thru launch is the catalyst that closes that gap.

What’s actually priced in: sum-of-parts vs. EBITDA multiples

I want to lay out the relative-value framework explicitly. Every number in this section should be read with an interpretation flag; these are my marks, not company disclosures.

DASH at $165. Equity value somewhere in the $70 billion neighborhood depending on the share count you use. Net cash on the balance sheet still meaningful, call it $5 billion. So an EV in the $65 billion zone. Against FY26 EBITDA that the company has guided into a band the sell-side has marked down to roughly $2.0 to $2.2 billion after the “several hundred million more” comment — that is a 30 times multiple at the midpoint. Direction: that is a premium multiple for a company that is admitting it will spend through its margin line. The bear case is that the right multiple is 22 to 25 times, which is another $7 to $10 billion of EV that comes off. The bull case is that the FY27 EBITDA reverts to a $3.0 to $3.2 billion print once the investment cycle ends, at which point the same 30 times multiple gets you back to $90+ billion of EV. The market is pricing the bear case for 2026 and putting an option value on the bull case for 2027.

TOST at $32-ish. Equity value in the $18 to $19 billion neighborhood. Net cash positive, call it $1.5 billion. EV in the $16 to $17 billion zone. Against the $790 to $810 million EBITDA guide that is roughly 20 times forward. Direction: that is a discount multiple for a payments-plus-SaaS name growing recurring gross profit at 22 percent. The Simply Wall St narrative fair value of $36.75 implies roughly 22 to 23 times forward EBITDA, which is still below the peer group. If the drive-thru launch produces even moderate sell-side upgrades to the FY27 number, the multiple should re-rate toward 25 times, and the equity value compounds against a guidance band that the company has been disciplined enough to hold.

The trade I keep drawing on the back of the bagel napkin: short the rerating risk on DASH for the next two quarters, long the catalyst path on TOST through the drive-thru launch and the Q2 print. Flag this as commentary, not a recommendation. I am not your advisor. I am a guy with espresso.

The harder question — and the one the buy-side AI-premium note gets at from a different angle — is whether the entire restaurant-tech cohort is being repriced or whether the cohort is bifurcating. My read is that it is bifurcating. The names that have leaned hardest into the AI capex cycle (DASH, Olo before it went private — see the Olo take-private note from March) are being marked down to a build-platform multiple. The names that have leaned into operating discipline plus targeted AI shipping (TOST) are being held in a holding pattern that breaks one direction on the catalyst.

The bet: if you can stomach Q2 burn, the entry is now

The case I want to make on the way out the door is the simple one. The DASH drawdown is not the end of the story; it is the middle of one. The “several hundred million dollars more” guidance is going to print in the Q1 and Q2 numbers and the stock is going to feel it. The catalyst for the rerating back to a marketplace multiple is the Q3 print where the spend starts to fade and the synergy line from Deliveroo starts to show. That is two quarters away. If you cannot hold through two quarters of margin compression you should not be in DASH.

Toast is the cleaner expression of the same thesis. The company has already absorbed the AI-investment quarter; the Q4 2025 print was the one where the guidance band came in tighter than the bulls wanted, and the stock has been digesting that ever since. The drive-thru launch on April 14 is the moment the optionality stops being optional. The 26 percent ARR exit-rate guide is the number to mark against the Q2 print. If Toast prints in that band and the drive-thru platform produces even a single named QSR logo by the Q2 call, the multiple should start to walk. If it doesn’t, the multiple stays at 20 times and the stock does not move and you have a $32 entry on a name growing recurring gross profit at 22 percent.

The asymmetry is real. The bear case on TOST is the multiple stays compressed for another year. The bull case is the multiple walks to 25 times and the equity value gets a 25 percent move. Against a DASH where the bull case requires you to wait two quarters and the bear case is another 15 percent drawdown, the TOST trade is the one with the cleaner skew.

I have been writing this column long enough to know that the cleanest trades are the ones that are obvious in retrospect and contested in real time. The DASH drawdown is contested right now. The TOST flatness is contested right now. The mirror is the trade. The April 14 launch is the first read. The Q2 print in early August is the confirmation.

Two espressos in, the bagel is finally cold, and the conviction is high enough that I am going to close the tab and go for a walk. The model will be here when I get back. Sometimes the right thing to do with a clean asymmetry is to mark it, sit on it, and let the catalyst do the work.

— Marcus edits The Bottom Line for TableTransfers. Tips: [email protected].