BCSP CIT common mistakes
BCSP CIT · Understand the traps and how to avoid them.
Accepts a stakeholder's request ("we need a two-hour lockout/tagout class") and immediately begins drafting agenda and slides without ever testing whether the gap is a knowledge/skill gap or an environment, equipment, feedback, or consequence problem.
The requester arrives with a pre-packaged solution and a deadline, and the trainer wants to be seen as responsive and cooperative rather than obstructive. Asking "is training even the answer?" feels like pushing back on a client.
What to do instead
Before any design work, complete a short, documented gap analysis: describe the desired performance, the actual performance, and the evidence for the cause. If the cause is not skill or knowledge (e.g., no access to the right tool, no feedback on performance), record a non-training recommendation and get the requester's sign-off before designing anything.
Builds content from what subject-matter experts say learners "should know" instead of from an analysis of the actual job tasks and an assessment of what the target audience already knows, resulting in courses that spend an hour on material learners already do correctly and skip the steps where they actually fail.
SMEs are accessible and confident, while observation of real work and learner pre-testing take travel, scheduling, and time. Experts also have trouble recalling what novice performance looks like because their own knowledge is now automatic.
What to do instead
Conduct a task analysis on the highest-frequency, highest-consequence tasks (interview plus direct observation), then gather a sample of the target audience's actual work products or run a short pre-test to establish the entry level. Build objectives only for the delta between observed current performance and required performance.
Writes objectives as content topics or as states of mind ("cover respiratory protection," "understand the importance of hazard communication") rather than as observable performance with a condition and a criterion.
Objectives are written after the slides exist, as a table of contents copied from the content outline, so they describe what the trainer will say rather than what the learner will be able to do.
What to do instead
Write each objective before developing any content, using an observable action verb, the condition of performance, and a measurable criterion (e.g., "Given a supplied-air respirator and a written scenario, select and don the correct respirator and perform a user seal check with zero critical errors"). Reject any objective containing understand, know, appreciate, or be aware of.
Selects the delivery method by habit - typically slide-driven lecture - for objectives that are clearly psychomotor or decision-making, and then expects learners to perform the skill correctly on the job.
Lecture is the method almost every trainer has seen modeled, requires the least preparation and equipment, and is easy to time. Method selection is treated as a formatting choice rather than an instructional decision.
What to do instead
Build an alignment table that lists each objective, its learning domain, and the method and assessment that match it. Require demonstration with hands-on practice and corrective feedback for any skill objective, scenario or case work for any decision objective, and reserve lecture for declarative content that learners cannot reasonably discover.
Produces slide decks in which every slide is a full page of reference text and delivers them by reading the slides aloud to the class.
The trainer is also the technical expert and fears omitting a detail, so the slides become an archival record for the trainer instead of a visual aid for the learner. Cutting content feels like reducing quality.
What to do instead
Move reference detail out of the slides and into a job aid or participant reference document; enforce one idea per slide with minimal text, and write speaker notes for the narrative. Then rehearse so the deck can be delivered without reading it, and confirm any slide the trainer cannot explain from memory gets rewritten or cut.
Ships a course that exists only as a slide file - no facilitator guide, no participant workbook with practice space and note-taking structure, and no performance-support job aid learners can use back at the worksite.
Materials development is squeezed at the end of the schedule, and slides are the artifact the trainer personally needs, so everything else is dropped. Transfer and consistency across multiple instructors are not tracked as deliverables.
What to do instead
Define the materials package in the design phase as a checklist - facilitator guide with timing, transitions, and answer keys; participant workbook; assessment instruments; and a job aid designed for use on the job - and treat a missing item as an incomplete project, not an optional extra.
Demonstrates a safety-critical procedure once, answers a few questions, and moves on, so learners leave having watched the skill performed but never having performed it with corrective feedback before being assessed.
Practice consumes class time and requires equipment, resets, and cleanup, and demonstration feels efficient because the trainer controls it and nothing goes wrong. Watching an expert also creates a false impression that learning has occurred.
What to do instead
Design guided practice into every skill objective and protect that time in the agenda: demonstration, then supervised learner practice with immediate specific correction, repeated to a criterion, then assessment. Cap the ratio of talking to learner practice and cut content rather than practice when time is short.
Checks for understanding by asking the room "Any questions?" and, when no one speaks, treats silence as evidence that learners have mastered the content before moving to the next topic.
Successive head-nodding and silence feel like validation, and asking named individuals a question risks embarrassment for the learner and discomfort for the trainer.
What to do instead
Plan targeted checks in advance: pose a question, pause, then call on a specific person; use a quick scenario or one-minute pair exercise; or use a show-of-hands poll with distractors that reveal the specific misconception. Treat any check that cannot distinguish a learner who understands from one who does not as not a check at all.
Reports end-of-class smile-sheet ratings ("4.7 out of 5, excellent course") as proof that the training worked and that the performance problem is solved.
Reaction data is cheap, immediate, uniformly positive, and easy to put on a slide for leadership, and positive ratings reward the trainer rather than challenge them.
What to do instead
Tie every evaluation instrument to a stated objective and to the level of evidence it can support: knowledge or skill testing for learning, observation or work-sample audit for behavior, and an operational metric with a stated baseline for results. Report reaction data only as satisfaction, and state explicitly which questions it cannot answer.
Collects evaluation data at the end of the course but never uses it - no post-course observation, no comparison to a pre-training baseline, and no documented revision to objectives, content, or delivery in the next course iteration.
The course is considered finished when delivery ends, evaluation is bolted on after design is complete, and no one owns the follow-up because the trainer has already moved to the next class.
What to do instead
Write an evaluation and revision plan during design that specifies the baseline to capture before delivery, the follow-up measurement method and timing (including an observation or work-sample check after learners return to the job), and a named owner. Schedule a revision review after the first delivery and record each change and its justification in a course revision log.