Evaluate
Once a reasonable number of ideas have been laid forth, the designer should begin a triage process to weed out the ideas that are unworkable or don't have merit. This should be a distilling process where the mix of ideas is reduced down to a select few.
The process for this reduction should rapid and repetitive. At this point all of the ideas should meet the baseline requirements which were determined back in the research phase, so the main purpose of the evaluation phase is to determine whether or not the idea will work, and what amount of time and effort will be needed to bring it to fruition.
A smart man told me once that "We can do anything you want. It's all just a matter of having enough time and enough money." I'm sure this is true, but these days both time and money are in short supply, so it's important to spend time and resources where they can do the most good.
With that said, I'll also note that this DOESN'T mean that you should always choose the idea that's fastest to complete or cheapest to implement. In the evaluation stage it's important to look at the entire life-cycle of a product. I can't count the number of times I've worked on projects where I've come into the fray once implementation had begun only to find that the decision-makers had failed to account for long-term things such as supportability and extensability.
