The author's conclusion that calling the Fire phones a failure and the Alexas a success is "just pos-hoc narrative" is a bit tautological. If you are evaluating

The author's conclusion that calling the Fire phones a failure and the Alexas a success is "just pos-hoc narrative" is a bit tautological. If you are evaluating the performance of something in the past and labelling it as a success or failure, you are going to be making post-hoc narrative. Why "just" post-hoc narrative? Is there something wrong with analysis of past events? Saying that the Redkins lost too much this season including last weekend because of failure to address personnel needs on defense, thereby missing the playoffs is "just post-hoc narrative." But talking about why you think an organization did well or poorly isn't useless. On the one hand, yes, the author is right. Fire and Alexa were experiments before we knew how they turned out, and a company making experiments can be a good thing. On the other, there were indicators that Fire might not be a good experiment and Alexa might be, and perhaps Amazon wasn't as keyed in as it could have been to these indicators.