We were curious: if we only had a few hours to drive, when should we do it? So, we went digging through the data to figure it out, and we found that the highest-earning delivery window of the entire week is Sunday from 6 to 8 pm. Interesting right?

That number comes from Gridwise, which tracks real earnings across delivery platforms including DoorDash, Uber Eats, and Grubhub. We pulled it from their 2025 earnings breakdown, built on data from more than 115,000 drivers. Everything below is gross pay before gas and expenses.

Here's what we found….

Though it's close, Sunday is the highest-earning night for delivery. Here's the top of the leaderboard:

Our read on why? Demand holds steady because people are still ordering dinner before the week starts, but driver supply drops off. Plenty of drivers grind Friday and Saturday, then take Sunday to rest or handle the rest of their life. Fewer drivers splitting the same orders means better offers land in front of whoever's still online.

But Sunday isn't winning on one window. It's the highest-earning day overall, with more blocks above $17/hr than any other day, including 3 to 5 pm at $17.12/hr. So it's not a fluke in one two-hour slot. The whole day seems to be cost-efficient.

Sunday 6 to 8 am comes in at $17.30/hr. Gridwise flags the pay in that block as surprisingly high. The likely explanation is the same one that makes Sunday night top of the leaderboard. Grocery orders for the week go out, and fewer drivers are online to take them. Low competition plus steady demand raises the rates.

It's early, and it's the one morning you probably wanted to sleep in. But it might be worth testing out if this rings true for your neighborhood and city.

Tuesday, 12 to 2 pm, at $14.17/hr. That's the single lowest window in the dataset. Weekday mid-mornings aren't much better, with Tuesday and Thursday 9 to 11 am both landing in the $14.25 to $14.45 range.

Here's why this one is worth paying attention to, though. Weekday midday feels fine while you're doing it. Orders keep coming, you're not sitting idle, and it reads as reliable.

But steady low volume isn't the same as profitable. Tuesday lunch pays $4.11/hr less than Sunday dinner and $3.13/hr less than Sunday morning. Every two-hour weekday midday shift you swap for a weekend one is $6 to $8 you didn't have to work any harder for. Nothing about the effort changes. Only the timing does.

The 6 to 8 pm block is the highest-earning window on every day of the week. Not most days. Every day. Dinner hours pay between $15.67 and $18.28/hr while midday hours run $14.17 to $16.30/hr, which works out to as much as a 29% gap for the same time behind the wheel.

Two things drive that. Evening orders tend to be bigger, and bigger orders pull bigger tips since most apps suggest tips as a percentage of the total. Peak pay also shows up more often during dinner, stacking on top of base rates that were already higher.

Everything above is delivery data. Rideshare runs on a different clock.

Delivery demand follows meals, which is why the dinner block dominates so cleanly. Rideshare demand follows commutes, nightlife, and airports, so the strong windows land in different places: weekday morning rush, evening rush, Friday and Saturday late night, and airport runs.

The pay structures are different too. The 2026 report shows base pay makes up the vast majority of rideshare earnings, while tips account for roughly half of delivery pay. That's why a tip-friendly window matters enormously on the delivery side and much less on the rideshare side.

If you run both, you’ll make the most. Rideshare during weekday commutes, delivery in the evenings and on Sunday. Mystro runs your gig apps side by side so you can always choose the best option.

These are national averages, so your market has its own version of this. A college town runs differently than a business district, and a driver near a stadium has windows that don't exist in the suburbs. Worth keeping in mind for the fall, too: NFL Sunday reshapes the entire afternoon in football markets.

The move isn't copying this schedule exactly. It's testing your own city’s version of it.