Note
When a Key Technician Leaves: What Goes With Them
For a contractor running electrical installations at Thai factories, there are two versions of the single-point-of-failure problem. The first is the schedule: one person who knows which technician goes where and when. That problem is worth reading about on its own. The second is harder to see: site-specific knowledge that lives inside the heads of the technicians who did the work.
When a senior technician announces they are leaving, the schedule gap is visible. The site knowledge gap is not. This post is about the second one.
What actually leaves when a senior technician goes?
A technician who has been with you for four years and commissioned 15 factory installations personally carries a mental archive that never made it into any system.
Which sites they know well. Three of those 15 factories have distribution boards in non-obvious locations. One requires 48 hours advance notice to access the gate. Two have facilities engineers who are slow to sign and need a reminder call on the day of the inspection.
What the installed equipment is actually doing. Solar inverters at one site were configured with non-default protection settings because of local grid conditions. The settings are documented, but the technician knows why they were changed: context that matters when the next inspection raises a question.
Which sites are still in warranty. They remember roughly when each job was commissioned and which components came from which supplier. If a fault appears, they know who to call without looking anything up.
What outstanding items were left at handover. Two or three sites were handed over with minor items pending: a cable tray section flagged for replacement, an inverter display that was unresponsive. That information lives in the technician’s notebook or memory.
None of this is unusual. Knowledge accumulates through doing the work. The problem is that it accumulates in a person, not a place.
How many sites are inside one person’s head right now?
A simple count shows the exposure.
List every technician who has been with you for more than two years. For each, count the factory installations they led. That number is a rough measure of how many sites lose an informed contact if that person leaves this week.
Example: a contractor with 18 technicians has three with more than two years’ service. The most senior led 14 factory installs. The second led nine. If either left, between nine and 14 sites would lose the person who knows them best. Those sites are also the most likely to have an inspection due in the next 12 months and the most likely to generate a warranty callout.
The question is not whether a technician will eventually leave. Some do. The question is whether the knowledge goes with them.
Why does knowledge transfer fail at the point of departure?
When a technician announces they are leaving, the standard response is to ask them to brief whoever takes over their sites. This rarely works well.
Verbal briefings cover what the person thinks to mention: the recent, the dramatic, whatever prompted a question last month. The quiet, stable knowledge (where the panel is, what the site contact’s name is, why the earthing configuration looks the way it does) tends not to come up because it has not come up recently.
Briefings also happen under time pressure. Someone leaving in two weeks is finishing handover files, running remaining jobs, and mentally transitioning out. A knowledge extraction that depends on that window rarely captures the full picture.
What does captured site knowledge actually look like?
Useful site knowledge is not a long document. For each site, it is a few specific facts that let a technician who was not there the first time show up and do the job without calling anyone.
From the installation:
- Equipment type and location (inverter position, distribution board access, panel layout)
- Configuration notes where non-default settings were applied, and why
- Customer contact for site access and document signing
- Access procedure if the site has specific entry requirements
From the handover:
- Date of practical completion
- Warranty expiry for major components
- Outstanding items at signing, if any
From subsequent visits:
- What was observed at the last inspection
- Any fault history and how it was resolved
A site register built from closed jobs holds most of this automatically when the closing process requires it. Configuration notes and access procedures are short additions that a technician fills in when closing the job, while the details are fresh.
What can you do once a departure is announced?
If someone key has announced they are leaving and this information has not been captured, the best option is a structured extraction session rather than an open briefing.
Before the session: pull a list of every site this technician has led and have the names in front of both of you.
During the session, ask four questions per site:
- Access: is there anything non-obvious about getting in or getting the right person?
- Equipment: was anything configured differently from the default, and why?
- Customer: who to call, and is there anything worth knowing about how they work?
- Condition: anything outstanding or worth watching from the last visit?
For a technician with 14 sites, a structured session covering all of them takes about two hours if the sites are still fresh in mind. The output is notes in the site records, not a document filed and forgotten.
One limitation to acknowledge: this still depends on what the technician can recall. It is significantly better than an open briefing, but does not recover information that was never captured at the time.
FAQ
Does this problem get worse as the business grows?
In most cases, yes. As you add sites and technicians, the density of information known only by one person grows unless the job-closing process actively captures it. A contractor closing ten factory installations a year for four years has 40 sites worth of knowledge to track. If most of that is concentrated in two or three individuals, the departure risk is material.
How do you identify which sites have the highest knowledge concentration?
A rough signal: sites where the same technician attended every visit (installation, inspection, callouts) with no other technician present. Those sites have one person’s mental model and nothing more. A site where two or three technicians each worked at some point has at least informal distribution of that knowledge.
What is the minimum useful capture per site?
Access procedure, main customer contact, and warranty expiry. Those three pieces let any technician get through the gate, reach the right person, and know what coverage still applies. Everything else improves on that minimum but is not required for a basic visit without calling anyone.
A system that captures this at each job close does not require a separate knowledge transfer step later. The extraction happens as part of closing the job: photos, configuration notes, and customer contact go into the site record before the truck leaves. To see how TRACE 30 puts the site record in place as jobs close, visit /program/ or book a short call at /schedule-a-call/.