Build a Skills Matrix: Who Can Do What
Why this matters
In a small shop, who can do what lives in the owner's head. That works until you are sick, until you grow past the point you can track it, or until the one tech who can handle a certain job quits and takes the knowledge with them. A skills matrix puts that knowledge on paper. It tells you who to dispatch, where your training gaps are, and where you have a dangerous single point of failure. It takes an afternoon to build and pays off every week. Here is how.
What a skills matrix is
A skills matrix is a simple grid. Down the left side, list your people. Across the top, list the skills and tasks your shop does. In each cell, mark how capable that person is at that task. That is the whole thing. The power is in what the completed grid reveals.
Step 1: list the skills that matter
Brain-dump every meaningful task and competency your shop performs. Be specific enough to be useful but not so granular it becomes a novel.
- Group by job type or category so it stays readable.
- Include the soft and safety items, not just the technical ones: customer communication, safe work practices, accurate paperwork. These matter as much as wrench skills.
- Include the rare, high-stakes tasks especially. Those are where single-points-of-failure hide.
Aim for a manageable list. You can always add rows later.
Step 2: pick a simple rating scale
Do not overcomplicate the levels. A four-step scale covers almost every shop.
| Level | Meaning |
|---|---|
| 0 | No exposure, cannot do it |
| 1 | Learning, needs supervision |
| 2 | Competent, works solo reliably |
| 3 | Master, can train others |
The level 3 distinction matters most. A skill where only one person is at level 3 (or worse, only one is even at level 2) is a flashing warning light.
Step 3: rate honestly, with the techs involved
Sit with each tech, or with your leads, and fill in the grid honestly. Two rules keep it real.
- Rate observed reality, not titles or seniority. A 20-year tech might be a level 1 on a newer technique. That is fine and it is the point.
- Let people self-assess, then sanity-check it. Most techs are honest about what they cannot do when it is framed as planning, not judgment. Make clear this is for training and dispatch, not for punishment.
Resist grade inflation. A matrix that says everyone can do everything is useless and false.
Step 4: read the grid for what it tells you
Now the payoff. Scan the completed matrix three ways.
- Down each skill column: how many people can do this solo? A column with only one "2 or 3" is a single point of failure. If that person is out or quits, you cannot serve that work. These are your top training priorities.
- Across each person's row: where is each tech thin? That is their personal development plan, ready-made.
- For coverage: does every critical skill have at least two competent people? Aim for no critical task depending on one human.
Step 5: turn gaps into a training plan
The matrix is only worth the afternoon if it drives action.
- Attack single points of failure first. Pair a learner with the lone master and get a second person to competent. This is risk reduction, not just development.
- Fill each tech's biggest gap that the schedule actually needs. Do not train skills you never use.
- Set a target on the grid. "By next quarter, two more people at level 2 on this task." Concrete and checkable.
Step 6: use it to dispatch and keep it alive
A current matrix makes scheduling smarter. You can send the right capability to the right job and deliberately pair a learner with a master to grow coverage on the clock.
Keep it updated. Review it every quarter, mark new competencies as people level up, add rows as your services grow. A matrix that goes stale is just a snapshot of last year. Five minutes a quarter keeps it true.
What good looks like
A healthy shop's matrix has no critical skill resting on one person, a clear development target for everyone, and an owner who can dispatch and plan training from the grid instead of from memory. That is resilience you can see at a glance.
References
- See related: The Green Apprentice: First 90 Days Training Plan
- See related: Promoting a Tech to Lead: Readiness Decision Tree
- U.S. Small Business Administration (SBA), workforce planning guidance
- Trade-standard practice for skills assessment and cross-training