I'm the author. I agree that the biggest thing that I'm missing at the end of the experiment is customer feedback. But this wasn't a customer development call.

I'm the author. I agree that the biggest thing that I'm missing at the end of the experiment is customer feedback. But this wasn't a customer development call. This was a sales call. Also the contacts didn't want to talk. Not because it was outsourced, but because it was a cold call. The either agreed to a trial quickly (8%), or said they were busy/not interested and hung up (everyone else). Counter-intuitively, user testing data doesn't support the CTA as a button. Thanks for the feedback.