How a 70-Seat Bistro Lifted Table Turns to 2.6 an Hour With Service Robots

Servers were covering miles that never reached a guest

A 70-seat independent bistro running multi-level service faced a familiar squeeze. Plates, drinks and bus tubs moved along the same aisles and stairs as the guests, and the servers carrying them were also the people expected to greet tables, read the room and reset seats fast enough for the evening rush.

RobotLAB, a robot vendor, makes the general point that servers often walk miles per shift carrying dishes, drinks and bus tubs, tasks that add little value to the guest experience and lead to fatigue. That is a vendor view and not a measurement from this bistro, but it describes the floor most independent restaurants will recognize.

The bistro's starting point was 1.7 table turns per hour and 3.5 miles of server walking per shift. Those two baseline figures are the yardstick for everything that follows, and they come from the same secondhand account as the outcome.

Food running moved to robots, and servers went back to tables

The bistro moved food running to service robots. That change freed servers to spend more of their shift on tables instead of on the route between kitchen and dining room. The account describes the change in those terms and does not break the deployment into phases.

The public record does not say how the robots were selected, how the floor was mapped, how long go-live took or how staff were trained at this bistro. The vendor blog that carries the figures does say that many serving robots are operational within one to three days, covering site mapping and staff training, but that is a general vendor claim and not a finding about this site.

Because the robot supplier is not named here, and the rollout details are not on the record, this study treats the robots as a single change to food running and measures only the two outcomes reported.

Two measured shifts, both secondhand

Table turnover rose from 1.7 to 2.6 per hour after the change. Server walking distance fell from 3.5 miles to 1.8 miles per shift. Both figures appear in the same vendor blog post, which does not name the operator, the measurement period or the method used to collect them.

A second vendor, OpenDroids, repeats the same pair of numbers without a source of its own. That repetition shows the figures are circulating, not that they were independently confirmed. Anyone citing this case should verify both numbers with the operator before publishing.

Context matters for the turnover figure. An aggregator citing the National Restaurant Association's 2025 State of the Industry report says full-service restaurants average 1.8 table turns during peak dinner service, while the same page elsewhere gives 2.0 to 2.5 turns for full-service dinner. The two statements conflict, and neither was checked against the NRA report itself. On either range, 2.6 turns per hour would sit above the typical level, which is why the bistro's own baseline counts more than any published benchmark.

What a lease-based robot program looks like at a bistro this size

Service Robot Co. is an OEM-neutral commercial robot integrator for US businesses. We select robots across manufacturers, then lease them to customers through financing on a 3-year lease with one fixed payment quoted after a site assessment. From there we deploy and integrate each unit, train the staff who work with it and service it through a nationwide US engineer network, so the customer deals with one partner for the whole lifecycle.

For a restaurant weighing a food running robot, the useful first step is the same one this bistro's figures point to: measure the walking and the turns before any unit arrives. A site assessment is where that baseline, the floor layout and the right robot type are worked out together, so the lease payment reflects a fit rather than a guess.