Your dispatcher picks up the phone. It's a tech, again. Fourth call today, same as yesterday. Field techs calling office isn't one problem wearing a uniform. It's three or four different problems, and most shops try to solve them all with the same instruction: call less. That doesn't work, because the calls were never the problem. Missing information was.
If you want to fix it, start by finding out what's actually getting asked, not by buying new field communication tools before you know what's broken.
Log one week of calls before you fix anything
Have whoever answers the phone keep a running list for five working days. Doesn't need software. A notepad by the phone or a shared spreadsheet column works fine, this is a $0 fix and the only one that comes before any other decision. Every time a tech calls, write down two things: the job, and what he needed.
At the end of the week, sort the list into buckets. Most logs land on something close to this:
- Address or access: wrong gate code, can't find the unit, customer not answering at the door.
- Pricing: what do we charge for this, can I quote X.
- Parts: do we have this in stock, is this covered under warranty.
- Approval: customer wants a discount, job's running long, needs someone to say yes.
The categories are the roadmap. Each one points to a different fix, and none of them is "communicate better."
Four categories, four different fixes
Address and access problems are usually the cheapest to kill. The information already exists somewhere, it's just not in front of the tech when he needs it. The fix is to put the full job record, gate codes, parcel notes, callback number, in the record the tech already has open on his phone. If your dispatch software has a notes field nobody fills in at intake, that's the gap. If intake and dispatch don't talk to each other at all, a low-cost automation rule, something like a $10 to $20 a month Zapier connection, can push those fields into the job record automatically. No custom build needed for this one, most of the time.
Pricing calls get solved with a price book the tech can open himself, a shared PDF, a page in the field service app, even a laminated card in the truck for the common jobs. The goal is job information in the field that doesn't require a live human to retrieve, because looking up a number for someone standing on a customer's porch is dead time for both people on the call.
Parts questions are harder, because they usually require checking real-time inventory, not a static document. This is where a spreadsheet or PDF stops being enough, and where a connection between the field app and the parts system starts to earn its cost. Once a shop is fielding enough of these calls that a tech is sitting in a driveway waiting on an answer, the lookup needs to happen on the phone screen, not through a person.
Approval calls are the ones that actually need a human decision, and the fix there isn't information, it's a rule. Give techs a dollar threshold they can approve on their own. Anything above it gets routed, by text or app alert, to whoever is actually free, not to whoever happens to pick up the office line. That routing is the piece worth building custom once a shop has enough techs that "whoever answers" stops being reliable.
Notice the order: cheap fix first on three of the four. The custom build only earns its cost on parts lookups and approval routing, and only once volume justifies it.
The real cost of field techs calling office
Say a shop runs six techs, and each one calls the office four times a day, split roughly one call per bucket above. Figure five minutes a call once you count hold time, the lookup, and the callback. That's 24 calls a day across the crew, 120 minutes, two hours of office time spent answering a phone instead of doing scheduling, billing, or the next intake call.
Two hours a day, 20 working days a month, comes to 40 hours. At a loaded office rate of $25 an hour (adjust for your own market, this is an estimate), that's roughly $1,000 a month going to calls that, in three of four cases, didn't need a live person at all.
That's close to a part-time hire's worth of hours, spent fielding questions the job record, the price book, or a simple approval rule could have answered. It doesn't count the tech's five minutes per call either, which is real labor cost on the field side, and it doesn't count the job that sat half-finished while he waited on hold.
Run your own numbers with your own call count and your own loaded rate. Most owners haven't done this math because the calls feel small in the moment. They aren't small in the aggregate, they're a standing cost that renews every month whether anyone looks at it or not.
None of this requires a program to reduce office calls through policy, telling techs to call less, or hoping people remember better. It requires moving the information to where the question gets asked, and building a rule for the one category that actually needs a person.
