If The Horse Is Dead, Dismount
In two decades working on all kinds of projects—some of which should have never ever seen the light of the day under any reasonable circumstance—I have never had the pleasure to experience a project being properly canceled. I mean: shelved, terminated, scrubbed for good. I have seen projects being manipulated and rescoped on the fly, this is, its objectives being shrunk or expanded as feces were impacting at terminal velocity the ventilation device. I have seen projects being merged—a set of ill-fated smaller adventures being opportunistically combined together to make them share their misery. I’ve seen projects being rebranded, which is an internal marketing stunt consisting in only changing their name in order to bring a breeze of fresh air in the otherwise reigning fetid atmosphere. I have seen projects being reassigned to another project manager with theoretical healing hands. And I have seen projects being slowly left to rot, incrementally deprived of resources until its flame extinguishes away only to disappear into the hall of shame of unremarkable unfinished endeavors. But I have truly never seen a project manager saying “people, the dream is over, project is kaput, let’s focus on some other things now”, and everyone moving on.
An obvious problem pops up: how do you diagnose that a project is terminally ill? It all boils down to metrics. As things go bigger, the demand for a constant stream of reports and data have the effect, intended or not, of diminishing the clarity for making decisions. A fatally wounded project might look nice and healthy as seen from reports created by those whose jobs depend on the project being nice and healthy.
Are consistent budget overruns enough symptoms to kill a project? Not really, because it can be claimed the project was initially underestimated by those who allocated the money in the first place. Skunkworks projects might be difficult to estimate if what’s being pursued has little or no heuristics to compare with. Consistent schedule overrun then? That must be it, right? Not really either: we all know everything takes longer than expected. And, in some cases, cool stuff might be late. See the JWST case.
And there is the famous problem of “efforts already made” or “pains already felt” that makes everything more difficult to stop. A project that has consumed brain power, money, blood, sweat, tears and time for a long time is more likely to be kept rolling to justify the ordeals suffered, only to make the problem worse: more resources are burned, making it even harder to be called off.
And last but not least, there is a question of pride: no project manager wants to add a canceled project into his or her track record. What transpires from this last point is how much it (still) prevails to correlate project cancellation to individual people’s incompetence instead of seeing it as beneficial to the organization at large: canceling a project increases the probability of succeeding in some other project. The longer you stay marching like a zombie, the longer you are not working on something that could be a jackpot. We culturally struggle when coping with premature ends, and our natural reaction is to try to find culprits.
Death sentence is something barbaric we humans are—alas, still—learning to leave behind. There are clearly better means to impart human justice. But, when it comes to ill-fated projects—as inanimate inventions of ours as they are, incapable of feeling anything—there is nothing especially wrong with going a bit medieval and keeping the guillotine sharp. Not to cut any human heads—as it’s metaphorically and frequently done in organizations—but to cut the crap keeping a lot of souls riding a dead horse that is going nowhere, no matter how much you kick it.