Your Best Tech Knows Too Much — And That's a Problem

The Night Mike's Phone Stopped Getting Answered

March 2020. I'm running dispatch from a card table in the garage. Hank's two, climbing my leg while I'm trying to figure out which of our three trucks went to which address, and my dad is in the hospital — knee replacement that turned into a six-week complication nobody saw coming.

Whittaker Plumbing had been running since 1984. Thirty-six years of west-side Cleveland customers, the kind where the lady on Broadview had been calling the same number since before I was licensed.

Except the number she'd been calling was Mike's cell. And Mike wasn't answering.

The calls went sideways fast. Customers who'd been with us for twenty years didn't know who we were without my dad's voice on the other end. Everything — who got a discount in 2009, who has a dog that bites, which customer cries when her basement floods because that's where her mother's furniture is stored — lived in one man's Rolodex and one man's head. That man was on a hospital bed arguing with a surgeon about a drainage tube.

The whole shop was one bad hip away from going dark. I learned that in about ten days.

---

What's in Their Head Is Not in Your Shop

Here's what nobody tells you about small shops. The stuff your best people just know looks like an asset right up until the moment it walks out the door.

I carry specific knowledge about Cleveland plumbing that took me about fifteen years to build. Which neighborhoods still have clay-to-cast transitions in the lateral. Which streets the city's got on the dig schedule for next summer. Which 1925 house in Edgewater is going to have a collapsed lateral under the side yard before the owner has any idea. You don't get that from a training manual. You get it from fifteen years of showing up and paying attention.

Now imagine Dom pulls up to a slow drain on West 117th without any of that context. He's two years out of his apprenticeship. He's good. He runs the snake, gets nothing, calls me from the driveway. I tell him to scope it. We find sixty feet of collapsed clay. Okay — that job got handled, because I was a phone call away.

But what if I'm coaching Trent's hockey game in Westlake with bad signal? What's in the job notes that tells Dom what he's walking into? Right now? Not enough.

I spent six years at Crestwood Plumbing before I came back. Thirty-truck operation. I watched what happens when what the senior guys know is spread across written processes versus when it lives in two guys who've been there since 1998. One version survives when those guys retire. The other one doesn't.

I came back to my dad's shop and didn't apply a single thing I learned there. Took 2020 to make me admit that.

---

Your Best Tech Leaving Isn't the Worst Case. Your Best Tech Staying Is.

I mean this.

The shop that loses its best guy in year three is better off than the shop that keeps him for fifteen years and then loses him. Because the shop that loses him early gets forced into writing things down. The shop that keeps him forever mistakes his presence for a system.

I am currently the liability I'm warning you about. I still run a truck two days a week. I know I shouldn't. But I'm out there, and that means twelve years of neighborhood knowledge and customer history is riding around in my head in a work van, two bad decisions away from being unavailable to my own shop.

And the way I took over from my dad — no written operating agreement, no documented customer history, nothing on paper — that was the same problem from the other direction. Just Mike saying "you'll figure it out" and me saying okay because I love him and didn't want to make it weird. When everything transfers on a handshake, you don't inherit a business. You inherit a pile of obligations you don't know about until you start missing them.

I've said before the handshake was a mistake. Still true.

---

What Getting It Out of Their Head Actually Looks Like

Dom on West 117th. Slow tub drain. He runs forty feet of cable, gets nothing, calls me. We scope it. Sixty feet out, under the side yard — collapsed clay. Roots, crack, chunk of pipe sitting at the bottom of the line. Classic 1925 house, classic Cleveland afternoon.

Here's what Dom needed before he knocked on her door. The pipe era for that block. That the lateral runs under the side yard, not the front. That there's a mature oak the customer planted with her late husband and she will not let anyone touch it. That this woman handles hard news better if you sit down at the kitchen table and explain it slowly, not from the doorway with your work boots still on.

I knew all of that. None of it was in Jobber.

Dom handled it fine because he called me and I talked him through it. But the whole operation in that moment was one dropped call. That's the problem. If I don't pick up, Dom's guessing. And a conversation that goes badly on an eight-to-ten-thousand-dollar sewer repair is a Google review and a lost customer.

Becca works two days a week. She's the reason the office runs. And she knows things I don't even know she knows — which customers call back angry after every invoice no matter what, which ones need a heads-up before the bill lands, which ones my dad promised a pricing adjustment to back in 2009 on a job I wasn't on that never got documented anywhere.

Exactly none of it is in Jobber. Not her fault. We never made it a habit. That's on me.

---

The Fix Is Not Software (But Here's Where Software Helps)

No field service software fixes this on its own. Jobber can't pull what Dom knows out of Dom's brain. It can only hold what he types before he drives away. And most techs, most of the time, don't type much before they drive away.

We've been on Jobber since 2019. Becca picked it, which is the actual reason it works. Whatever your CSR can stand to open every morning is the software you should be on. The rest is a feature list nobody reads after month two.

Jobber's fine. Not great. The invoicing is good. The customer texts go out reliably. The reporting is not why you buy it. The QuickBooks integration has a credit memo problem — Becca fixes about three a month by hand. She's stopped mentioning it, which means she's accepted it. I want to note it here because it's real and nobody puts it in the sales demo.

Where the software actually helps: making the habit hard to skip. If you build the close-out note field into the workflow so the tech literally cannot mark a job complete without filling something in, the tool does the nagging for you. The knowledge still has to come out of his head first. But the job creates the trigger. That's where it earns what you're paying for it.

---

What You're Doing Monday Morning

Pull the last thirty work orders. Go through them and mark which ones required a phone call to one specific person to complete. Not a Google search. Not the parts manual. A call to you, or your best tech, or your CSR, because nobody else had the answer. Count them.

That number is what you owe right now. Write it down somewhere you'll see it.

Then pick one thing this week.

Add a mandatory close-out note to every job before the tech can mark it complete. Three fields, two minutes — what did you find, what did you do, anything the next person needs before they knock on that door. If Dom had typed "clay lateral, mature oak, customer needs kitchen-table conversation not doorway conversation" into the West 117th job record, the next guy has something. Right now he has nothing.

After that, build a customer flag for anyone with relationship history that predates your current staff. Mike-era customers, old pricing arrangements, anyone with an expectation that came from a handshake fifteen years ago. "Dad knew this family since '94, treat accordingly" is enough. Get it out of Becca's head and into the record.

Then block thirty minutes on Friday — not a debrief, not an agenda, just whoever did the weird jobs that week talks through them out loud while someone else writes it down. That's it.

You're not fixing this in a week. You're starting it this week.

---

FAQ

What counts as tribal knowledge — is it just customer stuff, or does technical know-how count too?

Both. Customer history is the obvious one — who gets called before the invoice goes out, who has a dog, whose basement floods every spring. Technical knowledge is just as real and harder to document because techs assume it's obvious. Knowing the pipe era to expect on a given block, how to read a specific old water heater brand, which city inspector is going to red-tag a certain fitting — that lives in heads and never gets written down. The guy two years out of his apprenticeship doesn't know any of it yet. He doesn't know what he doesn't know.

My best tech refuses to write notes. Says it slows him down. How do I handle that?

He's not wrong that it slows him down. It does. Your job is to make the case that the slowdown is worth it to the shop, not to him personally. Start with a low bar: what did you find, what did you do, anything unusual. Three fields, two minutes. If he still won't do it after you've made it that simple, you've got a different problem. Some guys don't want to be replaceable. That's a real thing and you should name it out loud. That conversation is better to have now than at the moment he leaves.

We're a two-truck shop. Isn't this overkill?

Two-truck shops are who this matters to most. At a twenty-truck shop you lose a key tech and seventeen others absorb the load. At two trucks you lose yours and you're running calls yourself at midnight. I was running a three-truck shop when my dad went down in 2020. I felt every gap. The smaller the shop, the more concentrated the knowledge, the faster you feel it when it's gone.

How do I document customer relationship history without it feeling intrusive?

You're not building a file. You're briefing the next tech the way you'd want to be briefed before walking into someone's home. "Customer prefers morning calls," "dog in the backyard, put it up before you start," "longtime customer, dad knew the family" — that's not surveillance. That's making sure the person showing up treats them right. Think of it as writing the version of the job you wish you'd been handed before you knocked on the door.

Is there a point where some knowledge just can't be written down?

Yes. The judgment call a twenty-year tech makes in the first thirty seconds on a job — the read on the pipe, the feel of a valve, the sense something's wrong before he can name it — that doesn't fully transfer to paper. I'm not arguing it does. What I'm arguing is that the stuff that can be written down mostly isn't, because nobody made it a habit. Fix the fixable part. Don't use the unfixable part as the reason to skip the fixable part. That's what most shops do.

What does a callback actually cost when it happens because the next tech didn't have the information?

At my shop, a callback on a job under $500 runs me two hours of labor and usually costs me the customer relationship. In labor alone, at my rates, that's real money. But the harder cost is the one you don't see right away. The customer who needed a slow kitchen-table conversation getting a rushed doorway explanation instead — that's a Nextdoor post, or a Google review, or somebody who quietly calls another plumber next time. In a neighborhood like mine, Nextdoor moves faster than any platform I know. You don't find out what you lost until the phone stops ringing.

Some guys don't want to be replaceable. What do you actually do about that?

You say it out loud. You tell them you understand why someone would feel that way, and then you tell them that a shop that depends on any one person — including you — is a shop that's one bad week away from failing the people who work there. That's not a threat. That's just true. I'm the liability in my own shop right now. I tell my guys that. It's harder to resist the argument when the owner is making it about himself too.