"I championed that project and it flopped. My credibility took a hit."

You remember the moment it turned.

Not a disaster, nothing dramatic. Just a sponsor who stopped showing up to the check-ins. A follow-up session that got moved twice and then quietly dropped. The gradual understanding that the thing you'd argued for, defended in a budget meeting and put your name behind was going to end without anyone saying it had ended.

And then months later somebody references it in passing, in a tone, and you feel it.

The worst part isn't the money. It's that the next time you bring something forward, you'll do it a little more quietly. You'll hedge. You'll frame it smaller than you believe it should be. That's what a flop actually costs, and nobody puts it in a post-mortem.

Ten years in this work, and the programs that die share the same four conditions. Every one of them is set before the vendor is chosen.

• Success was never defined in a way anyone could check. If nobody wrote down what different looks like in six months, the program cannot succeed. It can only end.
• The sponsor had nothing at stake. If the line leader's only commitment was budget and a kickoff appearance, the program was always going to be optional to everyone reporting to them.
• Nothing structural was allowed to change. Roles, responsibilities and standing meetings stayed exactly as they were, so people went back to the environment that produced the problem.
• Nobody owned it after the outsiders left. Insight has a half-life of about ten days without an internal owner. It doesn't matter how good the insight was.

Notice what isn't on that list. The quality of the facilitator. The framework. The content. The price.

Which means the decision you're beating yourself up about, the choice of firm, was almost certainly not the decision that killed it. The one that killed it was made earlier, usually by someone else, usually without anyone noticing it was a decision at all.

That's not an excuse. It's better than an excuse. It's a checklist, and checklists are things you can hold somebody to next time.