Why Two-Thirds of All Change Projects Fail
There is a stigma attached to failure. Few managers would openly admit that their reorganization projects failed miserably. It is a rare exception for a consultant to report at a conference that the merger she oversaw did not achieve its intended goals. Managers and consultants fail all the time, but no one talks about it.
There is, however, one striking exception to this rule. Almost all quantitative investigations of the success or failure of new management concepts, change processes, or IT projects conclude – as if steered by an invisible hand – that a large number of such projects fail. As early as the 1990s, the ‘inventor’ of business process reengineering – a now almost forgotten management fad – declared that more than two-thirds of reengineering projects did not achieve their ambitious goals. There are regular reports saying that 70% of all rationalization measures come to very little or nothing. Studies today have established that at least two-thirds of change management projects are unsuccessful. In some studies the percentage may be a bit lower (about 60%), but sometimes it is also higher, around 70% or, in rare cases, even 80%. The message of all the studies that have measured the effectiveness of projects is unambiguous: the likelihood of failure is high. What is the reason for the fact that the high rate of failure seems almost to be a natural law? Why do around two-thirds of projects always fail, almost independently of the management fad involved?
The message of these studies is that reorganization and change projects are very tricky affairs, that organizations that seek to implement fundamental changes are usually out of their depth. The managers lack the necessary know-how to push through reorganization; the employees are not sufficiently included in the process; and senior management does not offer enough support. At first, this message sounds plausible – after all, there is never enough know-how, inclusion, or support.
But there is an altogether different way of interpreting these studies. The two-thirds figure says very little about the performance of the businesses discussed in the studies but a lot about the people who publish these studies. The main objective of the studies which magically produce failure rates of between 60% and 80% is to prolong the life of an organizational concept that is beginning to seem unattractive or a management tool that has received a lot of premature praise. The studies create some passion for fashions that are dying out. Admittedly, this is a strong claim, but there are several points that make it plausible.
It is interesting to note that the studies that establish these high failure rates are usually carried out by the same consultancies that advocate and use the failing management methods – and in many cases the same consultancies that even ‘invented’ the method at issue – or by research institutions with close links to the consultancies. These institutions usually send a short questionnaire to the businesses, ask a student on a work placement to call all of the businesses, or send their consultants to the businesses with a short list of questions. Then they summarize the results of this information gathering in the form of a short study, which they send to the businesses concerned and also try to place in some economics journal.
In this way, they kill two birds with one stone. On the one hand, they get in touch with potential customers. The people responsible for the budgets of change management projects are constantly deluged with consultancy firm advertising brochures, so most of these are chucked in the bin; however, questionnaires address these people as experts, and so by sending questionnaires the consultants increase their chances of getting to talk to them. On the other hand, the consultancies and research institutions use the study to present themselves to an expert audience as specialists in the management method under investigation. Those in the specialist media like these studies because they use exact figures – say 68.5% – and thus create an impression of scientific precision.
But the situation is nevertheless astonishing. Why is it that the very consultancies and research institutions that are prominently associated with the development or implementation of the management tools carry out these studies? Are they not shooting themselves in the foot by announcing that ‘their’ management tools often fail? This all becomes more comprehensible when we consider the way the studies are designed.
Even though these studies will mention a likelihood of failure of two-thirds or three-quarters without the slightest hesitation, it is almost impossible to identify even one concrete project that failed. Journalists seeking to add some colour to their pieces about these studies by citing some actual cases often despair because they cannot identify any of the businesses that experienced the failures.
The consultancies are also little help. Their studies never say that three-quarters of their own projects failed. Instead, the stress is always on how wide-ranging the sample was, and on the fact that it was selected independently of their own customer database. After all, it would be quite embarrassing for a consultancy to have to admit that most of its projects fail.
Those behind the studies thus always imply that the failures are attributable to others, and they suggest that the businesses which approach them will find themselves among the third of projects that defy the general trend and enjoy success. The investigations therefore also almost always contain a list of rules for the avoidance of failure.
In the case of evergreen issues such as change management and mergers, potential customers will have seen many studies, over many years, on the tendency of such projects to fail. But in the case of the new management methods that periodically get introduced with great ballyhoo, such as lean management, business process reengineering, or virtual teams, the studies documenting their tendency to fail appear two, three, sometimes five years after these trends arrive, and just as they are starting to be forgotten.
When these studies are published, they tally with the observations of the many employees who say that the management method has not delivered what it promised. Two or three years after the launch of a new management fashion, no one would believe a report that alleged a failure rate of only 8% or 10%. But the percentage must not be too high either. While a rate of 92%, 97%, or 100% would indicate that there was no point in employing the method any longer, a rate of 60%, 66%, or 70% means there is still hope the method might work. And commissioning the consultancy which authored the study will increase the likelihood that your business will not be among the large number of failures – that, at least, is what the study suggests.
These reports promise to help businesses learn how to succeed. After all, they present not only the figures relating to how many projects failed but (Spicer, 2017) also the main factors that contributed to the failures. Among them are mistakes due to insufficient skills, insufficient involvement of employees, and a lack of commitment from senior management. The reports convey the idea that one can learn from these failed projects – ideally with the help of the consultancy that carried out the study, of course.
But this learning ultimately never takes place, because information about the projects mostly takes the form of superficial and anonymized figures, while the explanations for the failures are mostly commonplaces. Only if the likes of Daimler, Boeing, and McDonalds were prepared to discuss their failed projects, rather than only reporting their successes, would it be possible to learn from failure. But here we run up against the stigma that attaches to failure, and it is very difficult to overcome it.
In the end, studies of project failure are therefore simply a form of ‘contraception against knowledge’. They tell us very little about the allegedly failed change projects but a lot about the way consultancies, management gurus, and research institutions try to inject a little bit of extra momentum into fading management fads. But what is the use of such knowledge?
