It is a pleasure to have Ahmed Aboubacrine as guest blogger on the interesting and pertinent issue of how to think about ‘failure’ from different perspectives. Our paths have crossed many times at evaluation conferences and I always appreciated both his input and commitment to high-quality evaluation. He is an experienced evaluator who has worked among others for CARE International and Islamic Development Bank. Here he shares some useful lessons from his own experience. They also resonate well with the engaging book Evaluation Failures edited by Kylie Hutchinson.
**4 min read**
Experience of failure is a double-edged sword. On the one hand, one can hide it, blame it on others or just ignore it. On the other hand, it could be transformed into an incredible learning opportunity provided that it is studied, its underlying causes are identified, and more importantly, open discussions are held and corrective actions are taken.
We can have an enabling environment where leadership encourages and rewards learning from failure. Yet the challenge remains how to identify failure timely, or detect early signs of it – especially in development contexts that combine complexity and lack of solid evidence for an evaluation.
To address this challenge and prevent such situations, I want to share here from my experience as well as from the literature four lessons that can be useful for development and evaluation professionals.
Lesson 1: All that glitters is not gold: When assessing development outcomes, failure can look like success, and cognitive bias is often the most difficult to avoid. A number of water wells built and being used by communities may mislead the evaluators to conclude that there has been a huge increase in access to potable water and henceforth a contribution to reducing potential water-borne diseases. However, this actual reality may differ depending on the quality of water and its possible health and social consequences.
Lesson 2: Proper triangulation limits the stakeholders’ influence and opportunity to hide failures: During an evaluation the available data upon which the analysis is built may be misleading. A famous case was Abraham Wald’s work on aircraft survivability during World War II. The study aimed at enhancing the US Navy’s aircraft based on the returning planes after missions. Wald studied not only the returning planes as sample, but estimated the whole distribution of damage in order to ensure exhaustive, more accurate decisionmaking.
Similarly, when evaluators collect data, they are not always able to meet all the intended beneficiaries or observe the actual changes due to seasonal factors such as migration, the rainy season or shocks. Data on failure are not always available on the spot. That is why evaluators have to dig deeper and find out about the farmers that migrated or the pregnant girls that are not any more attending schools. The quest for truth has to guide them always to search for any relevant data, information or missing stakeholders.
Lesson 3: Consider carefully the timing of the evaluation: When undertaking an evaluation, choosing the right time matters. One cannot assess outcomes and impact immediately after delivery of project or program outputs. Besides, ascertaining sustainability needs time to ripen the project results and to test the maturity level of the effects and impact.
This is extremely important for infrastructure projects and programs in which results may differ depending on the time of the evaluation. One can observe one of two scenarios: either it is too early and there has not been enough time to mobilise resources for project operation, or the project is working because the installations are brand new and do not face any maintenance problem. Therefore, evaluations gathering data points at multiple times during the lifetime of the project and afterwards may be more accurate than grand one-time evaluation endeavors.
Lesson 4: Creating a learning culture goes beyond gathering knowledge, and requires balancing M&E efforts for analysis: Most organisations produce huge amounts of data and information, generating basic explicit knowledge that can inform an evaluation. But due to resource constraints – time, staff and budget – development evaluators often do not have the luxury to go deeper and uncover more tacit knowledge, intelligence and wisdom to substantiate the available evidence. Hence for better decision-making, resources should be allocated to create a learning culture by investing enough in the analysis steps of the process, rather than rushing and limiting the findings to description only.
Based on the above, evaluators should ideally always consider the following four dimensions to ensure that they are able to identify and learn from failures in order to enhance the quality of future interventions: (i) cognitive bias, (ii) selection/design bias, (iii) the timing of the evaluation, and (iv) stakeholders’ influence.
Failure is an opportunity for future success – provided that we identify it and understand it.
Henry Ford wrote that failure is the best opportunity to begin again more intelligently.
And Paolo Gallo said “We learn by making mistakes. If I’m wrong, I do not forget. I learn and I can explain it to others”.