[A project] is a temporary endeavour undertaken to create a unique product, service or result. A project is temporary in that it has a defined beginning and end in time, and therefore defined scope and resources. And a project is unique in that it is not a routine operation, but a specific set of operations designed to accomplish a singular goal. So a project team often includes people who don’t usually work together – sometimes from different organizations and across multiple geographies.

This is the definition of project given by the Project Management Institute. Now, the overall majority of NGOs working in emergency relief – and I would say the totality of those working with funds from large international donors like UN agencies – frame their work by projects, and virtually anyone who worked in an implementing role has had at some stage of their career the word “project” (be it officer or manager) in their job title.

And while this makes a lot of sense at times, it does not on many other instances.

Small technical clarification. The project cycle management approach that is most widespread in the development and emergency relief sectors is the so-called logical framework approach, a result-based project planning and management methodology whose principal tool is a matrix called, indeed, the logical framework (aka logframe).

Developed in the 60s by USAID and then further in the 80s by the German Development Agency, it was then adopted in the early 90s by the European Commission and then by most of the other donors, with major or minor variations (as one of my trainers once said, the sector creates a standard and then every agency slightly modifies it so that is no longer a standard). Some major UN agencies like UNHCR or UNICEF use a somehow different approach, but their frameworks still have strong similarities with the “plain vanilla” logical framework approach.

When done properly and in the right circumstances, it is a great approach, and the logframe is in fact a great tool. I happened to hear its praise sung from rather unexpected sources, and used it at times for personal endeavours with positive results.

However, there is a number of challenges.

The more mature version of the approach was conceived with the purpose of improving community participation and involvement in project design, unfortunately it became too often little more than a box ticking exercise. While the most important component should be the participatory work to create the logframe, much of the attention is devoted to the final product itself, rather than to how this was created. The debate about merits and limits of the approach is huge and there have been pressures for its demise for almost as long as it has been around. If you are interested in the question, just to name a few, you could have a look at Duncan Green’s blog, read something from Robert Chambers or by Ben Ramalingam (if you are interested in a in-depth analysis of a proper implementation of the LFA and of the most common misconceptions about it, and if you happen to know Italian, this is one of the best manual on the topic I had come across).

I will briefly mention here just two of the most common critiques moved to the LFA, namely that it is too rigid and simplistic.

First, the logframe approach is (supposed to be) a problem-solving technique, and a rather rigid one. Sometimes focusing on problems might lead to identify solutions that not always are the best, overlooking alternatives that are not really strictly related to it. It happens that in fragile contexts aid agencies miss interesting opportunities, because they are implementing a project aimed to achieve something and not something else, and they are bound by a contractual agreement that in fact offers limited flexibility (this is part of a wider problem I will deal with in a second post).

Secondly, the logical framework approach is inspired by a linear, cause-and-effect view of the world. This happens, hence this happens, then this happens. More often than not, things in the real world are not that easy. Ben Ramalingam in his book Aid on the edge of chaos (OUP, 2013) spends a good deal detailing how problems in the real world are often not linked by straight and clear cause-and-effect links, but rather they have different feedback relationship that influence each other. If the world that is being considered in the logframe approach looks a lot like the one in figure 1, most of the times it works more like the one in figure 2.

This is quite apparent, for example, in ecology. One of my favourite examples is the collapse of cod population in the north-western Atlantic in the eighties (I quote from Networks: A Very Short Introduction, by Guido Caldarelli and Michele Catanzaro, OUP, 2012).

At the time, the shortage of cod generated a massive economic crisis in the Canadian fishery industry. Canadian stakeholders asked for more expeditions to hunt seals, maintaining that controlling those predators of cod would help stop the collapse. Many seals were killed during the nineties, but the cod population did not recover. In the meantime, ecologists studied the different food chains that connected cod and seals. By the end of the decade, they drew a complete map ([fig. 3]), where a lot of different chains were found to connect the two species. In the light of this intricate picture, hunting for predators of cod does not necessarily help the fish. For example, seals predate about 150 species, and several of them are predators of cod: thus, reducing the population of seals can end up increasing the pressure of other predators of cod.

Casual relationships are not easy to find, and the logical framework approach might lead to a very rigid and possibly mistaken way to tackle a problem.

However, to me the most relevant problem, which in a way encompass most of the remarks that have been moved against the LFA, lies in the rather universal application of the approach no matter the size, scope, length, nature of the issue it is supposed to tackle.

I think everyone agrees that one thing is improving access to education in a remote valley in Ecuador, one thing is to improve livelihood opportunities for youth in Bali, one thing is to eradicate polio from the planet. All of these can be conceived as projects: it should however be matter of debate if the same approach can work in all these instances. Moreover, as I will try to clarify in a second post, not all the aid enterprises are best conceived in form of a project.

Somehow this point is often missed, and the industry tends to use a “one fits all” approach.

My next post will try to see how this approach is particularly unfit for emergency relief in complex and protracted emergencies – which happens to be the context about which I have direct work experience.