Instructional Design for People Who Will Not Sit Down

Training designed for a classroom and delivered to a workshop fails in a specific way: it is started, interrupted, abandoned, and eventually completed in a rush to clear the notification.

The constraint is not attention span. It is that a technician on flat-rate pay is choosing between training and income, in a workshop that is behind, with a job waiting. Design that ignores that produces completions rather than capability.

The constraints that actually apply

Time is money, literally. Under flat-rate pay, an hour of training is an hour unpaid. See flat rate and clock time.

Interruption is normal, not exceptional.

The environment is loud, dirty and badly lit.

Connectivity is poor where the work happens.

Reading is not the preferred mode for a substantial share of the population, and technical documentation is frequently written as though it were.

And the learner is an expert in most of the domain, which is different from a novice and requires different design. They do not need the fundamentals; they need what is different about this system.

What follows from that

Short units that complete. Something finishable in the time actually available. A module that cannot be completed in one sitting will be abandoned in the middle repeatedly.

Resumable at the exact point, on a different device.

Front-load the thing they need. A technician looking for the change in this model year should not watch eight minutes of introduction. Structure by what is different, not by what is complete.

Show, do not describe. A procedure demonstrated is understood faster than a procedure written, and technical writing is where a great deal of courseware fails.

Assume expertise. Explaining what a multimeter is to a technician insults them and costs the attention you needed.

And design for the second visit. Most technical content is consulted again later, at a vehicle, which is a reference use rather than a learning one. See mobile learning.

Microlearning: what is supported and what is not

The term covers two claims that should be separated.

Spacing works. Distributing practice over time produces better retention than the same amount of study in one block. This is one of the more robustly established findings in learning research, across domains and decades.

So short units delivered repeatedly over weeks beat a single long session — not because attention is short, but because spacing aids retention.

Retrieval works. Being asked to recall something strengthens memory more than re-reading it. A short question days after a module does more than the module repeated.

Neither of those claims is about brevity as such. A short module delivered once is short and it is not spaced or retrieved, and the benefit does not appear.

What is not supported: that complex skills can be built from disconnected fragments, or that attention spans have shortened in a way that requires it. A diagnostic procedure has structure, and delivering it as unconnected pieces removes the structure that made it learnable.

The practical version: break into short units, sequence them so the structure survives, space them over weeks, and include retrieval. That is a design pattern, not a length limit.

Structuring a technical course

Start from the task, not from the system. What must the technician be able to do, and in what situation.

Then the decisions. Technical work is a sequence of decisions with a reason behind each. Teaching the decisions transfers; teaching the steps produces someone who cannot adapt when the vehicle does not match.

Narrate reasoning. A recorded diagnosis where the technician hears why each test was chosen is worth several times a demonstration of the tests. This is the single highest-value content type in technical training and it is under-produced because it is harder to script. See training the trainer.

Include the failure cases. What it looks like when it goes wrong, what the misleading reading is, what the common error is. Experts hold this knowledge and courseware rarely captures it.

End at the task, not at a quiz. See assessments that predict performance.

The organisational half

Design cannot fix these and they determine whether design matters.

Training during paid time, or it is done at home or not at all. This is the single largest determinant of engagement and it is a pay policy rather than an instructional one.

Protected time, actually protected. Training time that is interrupted whenever the workshop is busy is not protected.

A reason it matters to the technician. Certification, progression, pay. See certification pathways.

And relevance to the work they actually do. A technician trained on systems they never see concludes, correctly, that training is a compliance exercise.

Common failures

Length inherited from the classroom. A two-hour module because that is what the face-to-face version was.

Narration of on-screen text, which is slower than reading and prevents both.

Compliance framing. "You must complete this by" is a different message from "this is what changed in the new model".

Assessment that tests attention rather than capability.

No offline capability, which excludes the workshop floor entirely.

And no way back in. Content completed once and then unfindable when the technician needs it at a vehicle three months later. Most technical training should have a reference layer, and building it separately is usually right.

The short version

The constraint is that training competes with income, not that attention is short.

Short units that complete, resumable, front-loaded with what changed.

Spacing and retrieval are supported; brevity by itself is not. Break it up, sequence it so the structure survives, and space it.

Narrate reasoning — a recorded diagnosis explaining why each test was chosen is the most valuable content type in technical training.

And train on paid, protected time, or the design does not matter.

For an established framework for accessible instructional design, see Universal Design for Learning Guidelines.