Pretty poor paraphrasing, but still on point. When outlining a project with your clients one must ensure that any "failure" is, in fact, success. A common mistake that dooms many engineering consulting projects is the failure on the part of the Techni-HO in question from recognizing that sometimes they may successfully arrive at an answer of "NO".
This is different from prior discussions on the concept of "Shooting the Moon" Guarantees earlier mentioned here. But instead a need for you the Techni-HO to be able to recognize a no-win situation long before it ever happens.
For example, a common engineering problem in my other life as a Self-Proclaimed-Die-Expert is Formability and Springback. I'll pick on formability since it is the easier one. Some of the early tricks in my career relied on my forming "expertise" to identify and correct forming issues. These forming issues were usually automotive sheet metal parts (big body panels, exterior and under body) that split. Of course, producing fenders for Cadillacs and hoods for Oldsmobiles that had big splits in them was totally unacceptable.
The idea was to make the splits go away. It was our roll to make them go away, but here is the trap, what if the splits can't go away? What if there is no solution to fully resolve the failure? We always had to be careful not to take too much ownership in the absolute execution of the solution. For if the suggested solution does not resolve the problem then what do you do next?
We tell them that to make the splits go away they need to reduce the strain in that area x%. By definition in sheet metal the split is an indication or symptom of excessive strain. It would be a mistake to point out that they need to increase or decrease some specific variable to fix the split. (unless you are very good at recognizing the "moon-shot" opportunity).
Carefully craft your solution. The good news is that many of your customers won't recognize restatement of the problem, as a failure to solve the problem.
Return to profitability, by bringing revenues and expenditures in balance. (no shit!?!)
The "HOW" is for the process stakeholders (your clients) to decide and implement. Help them use process to identify potential solutions, help them to implement, then be ready to be the one to measure the success of the change. Just be smart enough to say, "that didn't do it, but we are headed in the right direction. Or that was wrong, now we know it is not related to [yada yada yada]."
I hope you never promised that a specific implementation would achieve the required end. After all, a plausible solution might not exist for their given parameters. However, we may eliminate a plan that did not work, and that has value.
