Reasons why field technicians make a second trip, and how to stop it?

The walk-in cooler quit overnight, there is inventory inside that will not survive the afternoon, and the part is not on the van.

Technicians at a billion-dollar distributor of commercial kitchen and HVAC replacement parts could spend close to two hours identifying an unfamiliar component. An InstaLILY part identification agent returns a likely match in a fraction of a second. In the divisions with the company's worst first-time fix rates, requiring technicians to run it before the visit lifted first-time fix by five to six points.

Industry

Foodservice and HVAC parts

Who uses it

Field technicians

What it does

Field service parts identification

Measured

Cumulative to early 2026

A refrigeration technician pulls into a restaurant parking lot at seven in the morning. The walk-in cooler quit overnight and there is inventory inside that will not survive the afternoon. He opens the panel, finds the failed component, and does not have the replacement on the van.

Now it's a second trip. Tomorrow, if the restaurant is lucky. Nobody is happy.

The company is a billion-dollar distributor of commercial kitchen and HVAC replacement parts. Its customers are the people who keep restaurant equipment running, and the hardest part of their job was never the repair. It was working out which part they needed.

A technician standing in front of an unfamiliar fryer could burn close to two hours identifying the right component, cross-referencing model numbers, schematics and superseded part codes across legacy systems that did not talk to each other. Two hours of a skilled tradesperson's day spent not fixing anything.

The company and InstaLILY built an agent to collapse that. A technician describes the problem the way he would describe it to a colleague, by voice or text: what the machine is, what it is doing wrong, and for whom. The agent returns the likely part in a fraction of a second. It has been trained on millions of parts drawn from previous work orders, which means it is not guessing from a manual. It is drawing on what actually broke, and what actually fixed it, across years of real service calls.

By early 2026 it was in the hands of roughly sixteen hundred technicians, had run more than half a million searches, and had returned more than eighty thousand hours.

Then someone asked a sharper question: what if the technician did not have to look anything up at the job at all?

The company ran a pilot in the service divisions with the worst first-time fix rates, the jobs most likely to need a second visit. The change was not technical. It was procedural. Technicians in those divisions were required to run the agent before the visit and load the most likely part onto the van.

Which means five or six calls in every hundred that used to require a technician to drive back the following day now get finished on the first visit. For the restaurant, that is the difference between a cooler running by lunch and a cooler running tomorrow.

And it landed hardest exactly where the company was weakest. Not in the divisions already performing well, but in the ones that had been struggling.

The team's own read was that the work had barely started.

"The job is not done," she said. "We'll never be done. This is something we continue to evolve. It's a major part of our business, it's a big differentiator, and it's something that's going to be living and breathing for a long time."