|Update: This page has been substituted with completley new content.|International development project failure rates are variously estimated to be around 35% and a rough estimate of the losses arising from the $215 billion in private and international aid investment is roughly $75 billion. This “failure rate” is disputed by many in the larger aid organizations, to some degree motivated by concern for the image of their institution. Although this became a topic in the early 1990s that large international aid institutions stated they were addressing this problem, it would seem from the latest assessments and experience, that these failure rates appear to have remained the same.
A common media mistake is to associate project failures to be related to corruption as a diversion of funds. This is indeed a cause for leakage of funds and for project failures. However, the exposure of corruption is difficult when project failures are associated with project quality which are closely related to a lack of transparency.
The process of failure starts with the fundamental role of the professional competence of individuals who collaborate on the design, implementation and post-funding operations of projects. In our experience it is often the case that those assessing projects, to decide on whether or not to fund them, do not in fact know, looking at proposal documentation, if a project under consideration is over-ambitious or under-ambitious. This is because they have no effective means of imposing transparency on the design process. Whereas the theoretical probability is that an over-ambitious project will fail and the under-ambitious one will get by, there is here a fundamental problem in the system that this type of situation should never arise; it represents a failure. Project designs should have their objectives and operational design pitched to represent an optimised feasible solution that maintains risks within acceptable margins while maximising the development impact. Taking this as a key target, it then becomes apparent, given the relative scarcity of aid funding and the imperative of economic development for billions of people, that over-ambitious and under-ambitious projects are unacceptable. They both represent a failure in competence; both are not relevant, cannot be effective, efficient or economic and will have a disappointing development impact, if any, and doubtful sustainability. However those deciding on funding should have some means to evaluate project proposals so as to be able to identify those that maximise the feasible return to their investment, loans or grants; there is a need for transparency.
All of these process failures relate to project design deficiencies resulting in less than optimal design and lower than feasible potenial impact. So, for the donor who accepts a defective project cycle and portfolio management system, the perception can be that a project has been successful because it is complete on time, covering intended scope and within budget. However, deficient design can results in project objectives being set at a level well below what was achievable. On the other hand over-ambitious proposals inevitably result in disappointments for stakeholders and others.
The purpose of due diligence design procedures is to avoid failure arising from deficient design by making use of procedures, each with well specified information requirements and analytical methods, covering the whole cycle in a step-wise but re-iterative fashion. The outcome is a design process that generates precise Logical Process Options (LPOs) that contain all of the anticipated changes and impacts a project might face during implementation accompanied by decision support for those occasions when anticipated or un-anticipated1 changes take place during implementation. Overall the objective is to minimise the likelihood of the types of failures outlined above, from taking place.
1 anticipated changes that occur during implementation are identified by applying simulation techniques as a standard procedure within 3DP and un-anticipated changes can be analysed during implementation using the same simulation models but inputting data on the evolving conditions. In both cases decisions can taken in a more timely manner as well as being optimised.