When doing user studies of paper prototypes should you wait till one fails or test in parallel?

Question

So lets say you want to see how confusing a first go at your user interface might be. (new its never been released to the public before). your before the steps at which you might do a heuristic evaluation.

if you have say 3 designs. 1 is the favored design from the design/development team. Do you test this one on the users and if the feedback is poor then test the other designs.

or do you test them all in parallel and see which one has the best results and if that differs from what you think is the best design revisit your prototypes and re evaluate the design.

I think the parallel testing of parallel pro-types seems better but i was wondering if there were any arguments against it. or reasons not to? I was wondering if you multiple designs do you go with the best one from the tests, or do you test the one the design/dev staff though was the best then try to iterate on it if you get poor test results then switch one you cant improve the test results?

the user study i am referring to would be showing them paper prototypes of the interface and asking them to think aloud when exploring it. and based on some sort of interface that they see ask them how they think they would perform a set action. describe or draw it.

Answer 1

This is defined as going "Best shot or battle royale" by the Google Ventures Design Studio in their design sprints. Each approach has its pros and cons, but some factors can help you to decide:

• "Best shot" allows for a more detailed prototype.
• "Battle royale" allows exploring new spaces where there are little conventions, and may provide positive surprises for the non-favorite candidates.
• Hybrid approaches are also possible. You can start with a quick prototype of your favourite candidate (especially if there is consensus of a favourite), and make an exploration of the alternatives focused only on the aspects that didn't work during the test.