The first 30 days after leaving DoorDash.

Start from the channel math. On the standard Marketplace Plus tier (30% commission), the kept rate on a third-party ticket lands near 18% after all fees and food/labor cost; the same ticket ordered direct keeps closer to 29%. Those are the two numbers the entire delisting decision rests on, and the side-by-side walkthrough is in the DoorDash margin walk. The case for delisting is not that the platform is bad — it’s that the operator is usually carrying the channel on a belief about discovery that doesn’t survive contact with their own books.

The decision lands on a Sunday night: close the storefront, don’t reopen, don’t announce. The phone calls and walk-ins keep coming. The next four weeks have a shape worth knowing in advance.

Week 1: panic that didn’t arrive

Most operators brace for a 20–30% revenue drop in week one. The actual drop is usually smaller because the platform’s third-party order volume partially redistributes inside the same week — some to walk-in, some to phone orders, some to your own website. The drop is real, but the bigger surprise is how much of it doesn’t leave for good.

Week-one recovery is not passive. The seven-day playbook below is what to run starting Monday morning, day-after-delisting. None of it requires a new tool, a new vendor, or a new line item on the P&L. The order matters: the highest-converting moves go first while the customer attention is still on the change.

Week 2: the “is DoorDash down” calls

Complaint volume usually peaks in the second week. Customers who’d ordered through the platform start calling the restaurant to ask if you’re closed, if the app is broken, or if their order history has been deleted. The platform app is, predictably, no longer showing the restaurant — for a customer used to ordering that way, the absence reads as the restaurant being gone, not as a deliberate choice to leave the platform.

Most of these conversations are recapturable. The script: we left the platform, you can order direct from the website at a lower price, here’s the link. Some convert immediately. Some stop ordering. A handful stay in the “thinking about it” bucket. Budget for a small direct-channel credit ($10–$20) for the loudest complaints — most negative reviews come down once the customer accepts the credit and places a direct order.

Week 3: the cost-line that flexes

The unexpected line is usually packaging. Recaptured orders run larger than third-party orders did — bigger tickets, more items per order, more delivery containers per order if you’re fulfilling through a last-mile service like DoorDash Drive flat-fee. Per-order packaging cost climbs even when total order volume is flat. It’s an operational headache, not a margin problem.

The line that usually doesn’t break is labor. Line-cook schedules absorb the channel shift by moving hours from late-evening third-party prep into mid-evening walk-in prep. No headcount changes, similar labor budget. The kitchen stops running the second wave of packing-and-bagging that platform orders pulled between 8pm and close.

Week 4: where it lands

By the end of the month, revenue is usually back within single-digit-percent of baseline — inside the noise floor most independents already see week-on-week from weather, events, and seasonality. The kept margin in dollars per day, however, is materially higher than baseline, because every recaptured order is now at the ~29% direct-channel kept rate instead of the ~18% kept rate on the third-party channel. The shift is what makes the math work; total revenue is lower, but retained dollars per ticket are higher.

The clearest way to see why is to put the two tickets side by side — the same $42 average order, run through the third-party fee stack and the direct fee stack:

Complaint volume usually tapers around the third week. The loudest review or DM normally comes down once the customer accepts a small direct-channel credit and places an order through the website. By week four the message channels are back to their pre-delisting baseline.

What operators usually get wrong

The most common operator misread is how much of the third-party channel was already “displaced” relationship volume rather than net-new discovery. The operator believes the platform is bringing in ~40% net-new customers and 60% existing customers using a more convenient interface. After delisting, most operators discover the breakdown is closer to 10/90 — the discovery surface was mostly a vanity layer; the relationship was already theirs. The recovery is faster than expected because most of those orders were going to come back anyway.

The other common misread is the conversion rate on a recapture offer. A 10% direct-channel discount works, but it converts at a lower rate than operators usually project. Some customers stop ordering at the previous frequency — not catastrophic, but not the “they’ll all convert” story most operators sell themselves before delisting.

What the math gets right

The basic premise — that the third-party channel is roughly a 30% margin tax on what may be a smaller incremental-revenue surface than the operator believes — holds up across most independent restaurants. The kept-margin lift on direct orders is the cleanest evidence available for the thesis. The decision works for restaurants where the channel is mostly relationship-displacement; it doesn’t work for restaurants in genuinely discovery-driven contexts (urban density, transit-corridor flow, tourist neighborhoods) where the platform is actually bringing in new customers.

Don Goldstein is a restaurant operator and runs Muntin Digital. This playbook describes the typical shape of the first 30 days after a delisting, anchored to the channel-economics walkthrough in An honest DoorDash math for independent restaurants. Numbers in this article are illustrative ranges, not measured outcomes from a specific restaurant.