Desired failure parable

Doug Samuelson

Once again the OR/MS analyst consulted his very own oracle, a retired executive and analyst who had given him much good advice over the years. “I have a really bad feeling about this new project,” the analyst explained. “My client keeps insisting on goals that seem unrealistic. We’ve discussed this several times, we seemed to have agreement, and then the requirement got raised, not modified as we’d discussed. It’s almost as if they want my team to fail.”

“Maybe they do,” the oracle shrugged. “It does happen.”

“How so?” the younger analyst blurted out, in mild shock. “Why wouldn’t they want to have a success, for my team and for them?”

“There are many possible reasons,” the executive explained. “Sometimes, of course, the sponsor simply doesn’t know what they’re doing, or what they’re asking for. That’s the most typical situation, and maybe you can explain it to them – bit by bit, eventually. Sometimes you’re subcontracting to another company that won by claiming everything would be easy and then subbed out the hard part they realized they didn’t know how to do. Sometimes a boss wants an outspoken, overconfident younger employee to get a little lesson in humility. Sometimes there are personality clashes or conflicts within the sponsoring organization, and you’re being thrust into the hottest part of the battle as cannon fodder – or being sacrificed to show everyone else how hard the task was.

“But sometimes,” the oracle continued, “it really is more sinister and deliberate. Do you remember the TV show ‘La Femme Nikita,’ from back in the 1990s? Where agents who somehow messed up got sent on ‘abeyance missions,’ which were really designed to get them killed? Well, sometimes things like that happen in real life – not getting the people literally killed, but often getting their projects killed. For instance, some years ago, I was a project leader for a company that was a subcontractor to a major consulting and R&D company, one whose name I’m sure you’d recognize, for a government organization that was – sorry, there’s no softer way to put this – basically corrupt.

“The company we were subbing to had expected to win the lead role on this whole set of contracts, and instead, the company I was working for had been chosen for a bunch of them. Many people in the government organization had relationships with the other company and wanted them to win the whole contract in a couple of years, when it would be re-competed. So, it turned out – I think! – that they had structured a task in which they expected we would fail, the other company would be in perfect position to report our failure, and that would help them win the next round of bidding. I don’t know, but I’d guess a couple of the key government people expected they’d end up with jobs at this other company after that. That’s illegal, of course, but it does happen. Nice setup, no?”

“What happened?” the younger analyst asked, wide-eyed.

“As it turned out,” the oracle grinned, “they had underestimated their intended patsies. We delivered a good solution to what had been widely considered an impossible problem, on time, within budget, with strongly favorable assessments from several independent outside reviewers. The other company was caught in the embarrassing position of having to explain to the sponsor why they had reported, prematurely, that we didn’t look as if we’d deliver our part. I believe our CEO quietly informed the other company that future collaboration would have to take place on terms more favorable to our company, or else.

“But the ending of the story wasn’t all happy,” the oracle continued. “The government sponsor showed their true colors by shelving the product and the final report. In fact, they decided to mark the final report ‘For Official Use Only,’ which meant we couldn’t disseminate it widely to show other potential clients what good work we’d done. Fortunately, with the help of a sympathetic executive officer in a different component of the government agency, we were able to get clearance to present a sanitized version of the work at a few conferences, with the key models but without some very interesting policy recommendations. And you know how even conference proceedings can end up getting widely disseminated these days, with all those Internet references. So I’ve had the interesting experience of attending several conferences sponsored by that government agency at which someone got up and talked approvingly about our model, but had no idea that their agency had sponsored the work and owned the model!”

“So what do I do now?” the younger analyst inquired.

“As we’ve discussed before,” the oracle replied gently, “you don’t have a lot of options. If you can do what my team did in that one case – pull out a success from a bad situation, that’s great – but don’t expect a huge reward if your darkest suspicions about the sponsor’s motives were correct. Otherwise, remember Norman Samuelson’s Law, from one of those parables you see in OR/MS Today from time to time: The more someone demands from you, the less they are likely to appreciate whatever you do for them. In short, if discussions with your boss and your lead company if there is one, and your client keeps leaving you feeling more and more frustrated and vulnerable, get out of there!”

Author’s note: The reference to “Norman Samuelson’s Law” is from “The Lawyer’s Parable,” December 1987. Can you believe I’ve been doing this for such a long time?

Doug Samuelson ( is president and chief scientist of InfoLogix, Inc., in Annandale, Va., and a senior operations research analyst with Group W, Inc., in Merrifield and Triangle, Va., supporting the Marine Corps Combat Development Command (MCCDC).