Except that they discovered they could about 2 months later and...

Except that they discovered they could about 2 months later and then proceeded to do so. That is the problem. While they were panicking they weren't getting together to assess all the options. They just panicked and went to market saying "oh crap, she's broke" with no real plan about what to do next other start again. The interface didn't work but they could bypass that and prove the chip actually did work which they did much later after everyone had lost interest and the damage to the SP was well and truly done. It was the apparent total lack of a plan and the prospect of nothing good happening for maybe a year that really slammed the SP. What they could have done differently is what they eventually did, but they should have realised that was an option and advised that they were proceeding to do that. That eventually changed everything, but nobody cared by then. If they'd realised that at the start it would have been "heh the tech actually works as intended, but the interface needs a bit more work. Easy peasy." I.e. Management had to have a proper trouble shooting and options assessment with the engineering teams before opening their mouths just because they were embarrassed they had screwed up and basically didn't trust their diversity hire, who they should not have put in that role in the first place. The only certainty in life is that stuff will go pear shaped, and a good management team expects that, allows for it and keeps cool under fire while they find a strategy that doesn't involve setting fire to the house because someone broke the TV.

It's like in Med research trials, you have primary end points and then secondary end points. You may miss your primary but the secondary might save the day.