Rethinking evaluation in ADDIE

Why is the Evaluation phase in ADDIE put at the end?

I know, it seems obvious to evaluate the results at the end, but since the inception of ADDIE, we’ve learned two things:

1. Evaluating learning results only at the end is difficult.
2. What if the evaluation shows that the desired results haven’t been achieved?

In agile software development, we try to evaluate every step of the development process. During the design phase, we build prototypes and test them with end users, test our software on different screens and devices, spec out edge case scenarios and address them, perform load-balance tests, conduct pen tests for security reasons… and then we publish it.

I know many instructional designers follow a similar routine, rigorously testing and evaluating their work during every step of the process—simply because it’s the right thing to do.

But it would be great if we talked about that more. ADDIE is a great model, but instead of using it in an outdated waterfall fashion, we should aim for a more iterative approach.