During the past few years, I've noticed that there has been considerable attention paid to the importance of gathering real-time feedback during project implementation, identifying ways to 'course correct,' and learning from failure as tools to improve development outcomes. This all sounds worthwhile and is consistent with what World Bank President Jim Kim calls 'science of delivery.'
What's not clear to me is how this is different from good and timely program monitoring? Ideally, monitoring should be built into project implementation, influence key decisions during execution, and ultimately improve development outcomes. If this is not happening already what is the reason?
In recent years, I've noticed a crowding out of conversations about monitoring in favor of a heated debate about evaluation (call it the Randomized Control Trial (RCT) revolution). Groups like MIT's JPAL and Yale's Institute for Social and Policy Studies argue we should use randomized control trials much more 'to know' what really works (and doesn't) because development practitioners haven’t been as rigorous in evaluating development failures. Others argue that RCTs are just not practical in many contexts, particularly when projects are already 'in process.' They would argue that what's more important than 'what works' is the 'why' 'how' and 'under what conditions' given the critical importance of adaptation to local conditions. Ben Ramalingam's work on complex adaptive systems suggests that adaptation is critical to scaling up what works and requires a better understanding of positive deviants in a given system.
So what happened to the 'M' in 'M&E'? Isn't the whole point of monitoring to adjust during implementation because evaluation is simply too late to influence project outcomes? Isn't real-time feedback just monitoring by a different name?
Perhaps what's new is our ability to collect much more data, from many more people, faster and at a lower cost. Project Mwana in Zambia uses SMS messages and citizen engagement to expedite early infant diagnosis of HIV resulting in a 50% quicker cycle time from diagnosis to treatment. That's a great thing and we should take full advantage. Companies in the private sector show us that firms companies who 'listen' to their customers and act quickly do much better than those who don't. The lesson: we need to continuously collect information from those impacted by our projects, listen to client feedback and make changes and improvements more often and more quickly.
So let's get back to basics. Let's listen and learn as we implement, leave room for adjustment, and course correct based on the data we gather. In short, let's put the M back in M&E.
What's not clear to me is how this is different from good and timely program monitoring? Ideally, monitoring should be built into project implementation, influence key decisions during execution, and ultimately improve development outcomes. If this is not happening already what is the reason?
In recent years, I've noticed a crowding out of conversations about monitoring in favor of a heated debate about evaluation (call it the Randomized Control Trial (RCT) revolution). Groups like MIT's JPAL and Yale's Institute for Social and Policy Studies argue we should use randomized control trials much more 'to know' what really works (and doesn't) because development practitioners haven’t been as rigorous in evaluating development failures. Others argue that RCTs are just not practical in many contexts, particularly when projects are already 'in process.' They would argue that what's more important than 'what works' is the 'why' 'how' and 'under what conditions' given the critical importance of adaptation to local conditions. Ben Ramalingam's work on complex adaptive systems suggests that adaptation is critical to scaling up what works and requires a better understanding of positive deviants in a given system.
So what happened to the 'M' in 'M&E'? Isn't the whole point of monitoring to adjust during implementation because evaluation is simply too late to influence project outcomes? Isn't real-time feedback just monitoring by a different name?
Perhaps what's new is our ability to collect much more data, from many more people, faster and at a lower cost. Project Mwana in Zambia uses SMS messages and citizen engagement to expedite early infant diagnosis of HIV resulting in a 50% quicker cycle time from diagnosis to treatment. That's a great thing and we should take full advantage. Companies in the private sector show us that firms companies who 'listen' to their customers and act quickly do much better than those who don't. The lesson: we need to continuously collect information from those impacted by our projects, listen to client feedback and make changes and improvements more often and more quickly.
So let's get back to basics. Let's listen and learn as we implement, leave room for adjustment, and course correct based on the data we gather. In short, let's put the M back in M&E.
Join the Conversation