Two Factors Responsible for Project Failure Today

It is an unfortunate fact that in business today, many projects run over-budget and over-due. This is true both in the public and private sector, and these mistakes cost the affected businesses and the taxpayer an enormous amount of money; in some instances, these failures to complete a project lead to a direct threat to public safety, as in the case of the Department of Defense’s failure to comply with its own FIPS-201 mandate.

Put simply, the two factors that cause these projects to fail, balloon beyond normal budget constraints, or take longer than necessary are fear and integrity.

The business world is full of people who know “something about a lot of things,” but who lack real depth in a given subject. Such people are often aware of this fact, and will perform their given responsibility, whether as a manager or an “individual contributor” as best they can. Problems arise, however, when these people are asked to work in an area outside their area of expertise, or beyond their level of understanding of a given subject. A typical example is the Systems Administrator who is experiencing a network failure, presented as a large number of workstations unable to reach the Internet or internal assets and cannot determine the cause of the service interruption. When asked, the employee is given two choices: admit they do not know what the problem is, or to “stretch the truth” and say that “it’s a network problem” and “they’re working on it.”

As the employee searches for the root of the problem, they may not know enough about the protocols spoken on the network to realize that a server that was recently deployed in another area of the building has been configured with a dhcp daemon that is sending out incorrect offers. Much time will be spent “googling,” talking to friends via instant messaging, desperately trying to find out if anyone else knows what the problem is. In the meantime, the employee may be questioned if progress is being made, and they will, almost without fail, lie, stating that progress is being made, that they are “tracking down the problem,” and may even give an entirely contrived estimate time of service restoration for the affected workstations.

Employees are often very afraid to admit ignorance on a given subject, or indeed any given subject. The truth of the matter is, however, that indicating at the onset of a problem or outage that there is a knowledge gap allows pooling of resources. In the case of the rogue dhcp server, it is likely that other people in the organization may know about it, or other people may know about tools such as tcpdump and snoop. Corporations tend to foster this culture of fear, primarily by not training managers to ask the right questions – “do you think somebody else could help with this?” or “is this something we should call our vendor about?” – and by not teaching them that employee growth is predicated on deficits in employee knowledge (which all employees have). The last contributor to this employee-fear-culture is a poor hiring process, in which people are hired because they said the right words in an interview, or wore a prettier tie, or answered the proverbial “how many gas stations are there in the country” question best. Better hiring practices and better training for managers can make an enormous difference in the productivity of an organization, simply by removing the implicit “threat” employees feel.

The other problem impacting productivity in industry is the lack of integrity of both individual employees and management. While the notion of “integrity” is a fairly simple term, and easy to spot in the archetypal example of “I tried it once, I didn’t like it, and I didn’t inhale,” there are many other, far more subtle ways to deceive and deflect responsibility in the corporate environment.

This latter problem usually comes from middle management, and somewhat more rarely, individual employees (their concern is largely that of fear, rather than deflection or deception). When budgets are exceeded, or projects fail to meet deadlines, or service outages occur, a very common response is to speak in such vague terms about the state or cause of the problem, what is being done to provide a remedy. Often, the individual responding to the event will make copious use of words directly from vendor marketing literature, either correctly or incorrectly, and sometimes both. The net effect is the receiving parties of this communication become confused as to what the actual state of the event is, and what is being done. All that is clear is somebody has crafted a very technical-sounding response, and that understanding the problem may not actually be possible.

In general, it is always possible to explain even very complicated situations to non-technical or un-briefed staff. Consider the loss of the Columbia orbiter; within hours of the breakup of the orbiter, the public had been informed that she had entered the high atmosphere at 17,000 miles per hour, and encountered tremendous heat from resulting friction. A defect in a control surface caused the orbiter to “tumble” and move about in ways such that surfaces, which were not prepared to sustain such heat and friction, were exposed to tremendous levels of both. As the craft tumbled, she began to break apart at a very high altitude, and that “The Columbia is lost.” The fact that the general public does not understand the behavior of ionized gas, escape velocity, control surfaces, or even the construction of the orbiter does not matter in this case. The situation is simple: a flaw in the vehicle caused it to overheat and break apart in the very high atmosphere.

If we re-visit the example of the rogue dhcp server, the local IT manager or director may send out an “all-hands” email filled with adjectives and future-perfect phrases, but very few actual present-tense descriptions of the problem, steps being undertaken to remedy the problem, or how long the outage will last. For the manager in this case, the message serves as interference, a foil, with which to dodge responsibility and parry any assertions of blame. The aim of these messages is to create such a complex set of unrelated questions and variables that addressing them individually, or even making sense of the individual terms becomes nearly impossible.

The term used to describe this tactic is utraquistic subterfuge, described in 1923 in The Meaning of Meaning, by Charles Kay Ogden and Ivor Armstrong Richards. The phrase has not made its way into management training courses, or into the guidelines of human resources departments, and indeed those employing it are rarely aware of the fact their behavior has been analyzed and described as, indeed, deceit.

Both fear and integrity are tightly coupled in this scenario. On one side of the equation, we have a party fearing repercussion or discipline, and on the other we have a party convinced that they can muddy the waters so thoroughly that they will not be disciplined or even implicated. After all, it is a very complicated situation, goes the logic.

There are simple cures for both of these character flaws in employees and management. The first and most important is to have an environment throughout the organization, from the lowest individual contributors, to the highest management, which encourages cooperation to overcome failures, rather than competition to avoid blame. Hiring practices which involve the staff whom new employees will be working with (versus a simple phone screen and interview with Human Resources), so the marginally skilled or “something about a lot of things” people can be spotted and eliminated from the process. There are of course many other curatives for these problems, with the least invasive and boat-rocking being simply providing training to employees so that they do not find themselves in situations where they are unprepared for problems which come up in their day-to-day duties.

The Columbia orbiter disaster is a particularly good example of how both of these character defects can lead all the way to the death of innocent people. Edward Tufte has commented on this in great depth in his The Cognitive Style of Powerpoint. Prior to re-entry, the engineering staff had sent many emails to middle management expressing concern. This management proceeded to deflect and avoid any responsibility for damage to the orbiter (in particular, the use of the phrase “not within scope of failure” versus the reality, that no testing had ever included that scope and thus results were at best unknown). Staff unsure of whether there was a failed component on the orbiter dithered and attempted to determine whether there was danger to the orbiter, without having any data with which to do so. Lastly, after the orbiter was most certainly lost, Boeing engineers were similarly without any data to deduce the failure modality from, and worked tirelessly and fruitlessly for literally a month before a wing panel was taken from Endeavour to perform in situ testing.

The fact that people have lost their lives to these two flaws is tragic. What is more tragic, however, is that they are basic character flaws in people, and that they can be recognized, educated, and grow out of them. Organizations with a culture of transparency where “blamestorming” (Frank Roche, iFractal) is not a “corporate skill” are by definition more agile and can respond to failure more rapidly and more effectively. The very culture in these organizations itself is the problem. The word itself, culture, derives from the Latin colere, meaning “to cultivate,” and “to grow.” In a word then, the best way to prevent these failures in organizations is to grow.