Tim Brown's piece in HBR ("Permission to Innovate") prompted a good amount of responses and opinions (the holy grail for a blogger!). His thesis surrounds the idea that certain companies create a culture of innovation by first fostering a level of tolerance - or permission - from users, channel partners, etc. to occasionally miss the mark, provided the company hits the mark more often than not. These innovative companies tend deploy multiple strategies to experiment and grow in adjacent categories.
What prompted such a volume of comments on Brown's post was his assertion that "what you can't measure, you can't get better at." (As one comment noted, it was Einstein who said "not everything that counts can be measured, and not everything that can be measured counts.") The community's response is understandable since metrics are usually code for success v. failure.
I'd suggest a different way in which to view metrics - they should be viewed as a source of learning.
When I worked with Toyota as a client, I was taken by the company's culture of Kaizen, or continuous improvement. Kaizen is a culture of test, learn and improve - not pass or fail. Metrics are used as a way of learning, not assigning glory or blame. If a program misses the mark, managers are celebrated if they can identify root cause solutions and learnings for v2 of the program.
The onus here is on leaders to instill this culture within teams. When reviewing metrics in staff or project meetings, ask all presenters to focus more on learnings that can be applied going forward, as opposed to discussing metrics in the past tense.