Instructional design needs an engineering mindset

I have a computer-science degree and fifteen years in L&D. The most useful thing the first gave the second wasn’t technical. It was a mindset: instrument, measure, iterate.

Most instructional design treats a course as a project. You scope it, build it, ship it, and walk away. Engineering treats what it builds as a product — something that gets better with data after it ships.

That shift changes how you work:

You instrument before you launch. What will tell you a module is failing — drop-off, time-on-task, a question everyone gets wrong? Decide before, not after.

You measure behavior, not applause. A happy post-course survey is the weakest signal there is. The real question is whether anything changed in how people work the following week.

You iterate on the data, not your taste. The module you’re proudest of may be the one quietly losing learners. Let the funnel tell you, then fix that one.

None of this requires an engineering team. It requires deciding that “we ran the training” is the start of the measurement, not the end of the job.