We don't need "what works". We need to know what is working.
Many claim to be seeking "what works" when looking for the solutions that will meet certain social goals.
This quest for evidence-based policies, programmes and practice has gained a renewed momentum, with the Cabinet Secretary Jeremy Haywood's calls for a "NICE for social policy" and the forthcoming creation of an Early Intervention Foundation.
Yet we don't want "what works". What works is bad for innovation, and by extension, bad for service users.
We have argued before that knowing what is effective improves outcomes, and that we should be drawing upon the best approaches and practices available to us from across the world.
But finding "what works" implies that once identified our work is done.
Instead we should be looking at "what is working" now, recognising that the collation of convincing evidence is just the beginning
Science Solves Problems
Now this might just seem like semantics, but it's important for service users and it's important for innovation.
At an event we hosted last year, evidence guru Sir Michael Rutter emphasised that science is a way of solving problems, not providing facts.
This is a crucial point.
Our problems change, we change, and prevailing wisdom will be overturned. We aren't searching for the definitive truth but what is effective for all our problems right now.
This has important implications for the ways in which we categorise knowledge. Many would argue that truth is always relative to the discourse of the moment - the power relations which produce "truths" of one period - which will then be superseded as society changes and knowledge evolves.
From the perspective of service users this is important. Although producing definitive lists of "what works" or "proven practice" can signal what has strong evidence behind it now, it could also reinforce the status quo, meaning we don't question the rationale of decision making - preventing alternative voices and issues from being heard.
This may lead to problems of services users going unnoticed, with new solutions not developed or identified to solve them.
Innovation is Evolutionary
Not identifying problems is bad for innovation.
Innovation is a transient, evolutionary, non-linear process. We need to continually be opening up the ways knowledge and data are generated and used. If it is supposed that we have identified "what works", then it could signal that there is nothing left to achieve.
This means we won't be experimenting with new solutions and the new, better approaches developed will struggle to gain the recognition they deserve.
The identification of "what works" also supposes rationality in decision making. That once we know what works or what is proven, it will be adopted and scaled.
Yet we know that the identification of a successful intervention doesn't equate to mass use, nor indeed does the identification of failure ensure that there is a change in delivery.
‘What Works’ Must Become ‘Still Working’
It should be emphasised that I don't want the quest to look for "what is working" to mean that we can't make recommendations and decisions based on current levels of effectiveness.
Decision makers - encompassing a spectrum of policy makers, practitioners, service commissioners, end users and beyond, - require timely, useable and accessible information.
This means that as well as looking for impact evaluations, we need to be seeking to improve the processes in which this information is orchestrated and used.
It is too obvious a point to note that selecting the most effective approaches will mean better outcomes for users.
However, we do not want our services to remain static. As long as our tastes, lifestyles, problems and needs change, we need to be experimenting, testing and developing new approaches that best suit them.
This means the "what works" is only relevant as long as it is "still working".