Leadership calls its own failed initiative a learning experience and calls comparable failures by its opponents plain incompetence, so accountability applies to one side only and critics learn to stay quiet.
An example
Two pilots failed at a national home-goods retailer in the same year. The chief executive's mobile loyalty app missed its sign-up target by 60 percent and was shut down in October. A regional vice president's small-format store pilot in four suburban markets missed its sales plan by 25 percent and closed in November. The all-hands deck in December gave the app a full slide titled what we learned, praising the team for failing fast. The store pilot appeared in one bullet under execution issues. In the regional vice president's next performance review, the pilot was cited as a judgment concern. A peer who had pointed out the contrast in a leadership meeting was told afterward that it was not the time. When the next round of pilot proposals opened, the regions submitted none.
Why it works
Calling a failure learning makes continued support for its sponsors cheap: allies can honor the investment and skip defending a loss. Holding those sponsors accountable starts to look like arguing against learning, which makes accountability look worse. The harsh label on opponents' failures makes criticism costlier, because anyone who presses the point invites the same standard on themselves and their allies. Whoever runs internal communications and post-mortems can do it, provided opponents have parallel failures to contrast, the audience lacks outside benchmarks, and the culture celebrates failing forward.
The thinking behind it
Lee Ross Lee Ross's The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process (1977) named the fundamental attribution error: observers explain other people's behavior by their traits and underestimate the situation they were in. The double standard applies that error selectively, explaining one failure by circumstance (an experiment, a shifting market) and the other by character (poor judgment). Insisting on the same causal language for both forces the situational question onto each case.
Erving Goffman Erving Goffman's The Presentation of Self in Everyday Life (1956) describes how people performing together act as a team with a shared interest in the show succeeding, keeping contradicting facts in a back region. Leadership's failed app becomes a front-region story about learning, told on the all-hands slide, while the comparable numbers stay out of view. Putting the two failures side by side, with the same measures, brings the contradicting facts on stage.
Warning signs
Two comparable failures receive opposite causal language in the same channel: learning, courage, and experimentation for the favored one, poor execution or judgment for the other. Post-mortems are run by the people whose initiative failed. Performance reviews carry the difference forward into ratings. Fewer proposals arrive from outside leadership's circle in the following cycle.
How to defend yourself
Argue for one standard, and leave aside whether your failure deserves the harsher label. Put the two failures side by side, in writing, with the same measures: target, result, cost, and cause. Ask in a public forum, such as a leadership meeting or a post-mortem, for the same causal language to be applied to both. Propose an independent chair for post-mortems, or a blinded review, so the framing leaves the narrator's hands. If a board, an auditor, or an analyst applies a symmetric standard, the double standard itself becomes the story, so give them the comparison.
How a leader shuts it down
Use the same post-mortem template, language, and chair for every failed initiative, whoever sponsored it. Separate the people who run a post-mortem from the people whose project is being reviewed. Check performance reviews for failures judged differently depending on whose they were.