When it comes to making things happen the model I was implicitly taught in local government required a committee report for decisions above a certain scale; even those that didn’t still followed a kind of ‘lite’ version of the same thing, and it looked quite simply like this. Set out the context and the problem, the options you have identified that might address the problem and then make a recommendation for the preferred option to be implemented. This was usually, but not always, the one that balanced lowest cost with lowest risk. The recommendation was usually agreed and, after a period, work started to implement the solution. This would require a project to be initiated and managed, whether large or small, whether in-house or through procurement of external expertise, whether it was a scheme to put new play equipment in a community setting or a piece of research about local quality of life. Depending on how frequently the decision-making body met and whether this required some form of procurement process prior to project initiation and implementation the whole process, as you can begin to imagine, could take over a year before from inception to the work actually starting.  

I call this the Decide and Deliver approach. It is a process that takes a broad range of information, determines a few solutions to the presenting issues or opportunities and leads to a decision about which one to implement. It’s like a right facing chevron narrowing down to the point of decision. It is a valuable and appropriate approach but it’s not the only one. It works well when we know we can’t hedge our bets, when we need to cut off alternative options and commit to a specific path, when we are working with a repeated or familiar problem, or when we have a mature market to provide solutions to our problem.

Decide and deliver searches for the silver bullet. We do it in procurement, where we narrow down potential suppliers to the one best balancing quality with our available budget; in recruitment we narrow down the field to a final appointment; VC investors have a range of options and curate a process whereby those options are narrowed down until you are left with the best bet. It’s usually one big bet too. A silver bullet approach assumes that there is a best option or best practice to be found and seeks to do so. It works well in ordered systems where we can understand the variables. 

In running those processes you are often left with the sense of ‘if only we could have the best of each supplier/candidate/start-up’. Clearly this is not appropriate for you are already committed to your linear selection path. Interestingly one of the ways to overcome bias in recruitment, where appropriate, is to recruit through a pool where you make two or three appointments to similar roles through the same process – perhaps you need three teachers or two service designers. It can enable you to take a more rounded view of the appointments and blend different skills and experiences and backgrounds to bring more diversity into the team. This approach hints at the alternative to decide and deliver, which I call Explore and Experiment. 

Explore and Experiment works well in the opposite circumstances – broadly categorised as those where there is no clear solution that we can specify, indeed there may be no market or solution at all. The context into which we want to intervene may be unclear or uncertain and therefore an environment where picking a single winner – placing one big bet – is higher stakes. The likelihood of failure, given the ambiguity involved, is that much greater. These are characteristics of a complex environment. The Design and Deliver approach doesn’t work here because it is based on an inappropriate paradigm, one that assumes we can implement a solution to solve the problem. Yet we know, as I have written about before, that the only certainty about interventions in a complex system is that they will generate unintended (and unpredictable) consequences. 

Once we have evaluated the challenge or problem and determined the nature of the situation we wish to intervene in we can choose the most relevant approach. If the situation is an ordered system we can use Decide and Deliver. For example, there is an established market to deliver a multi-use games area in a community and we can go through our standard selection process to procure the most appropriate provider at the price point we are able to afford. But if we want to address knife-crime, encourage staff to return to the office or improve educational attainment that’s a different story. Here implementing a single solution – programme, policy, product etc – isn’t an option because we are in a complex system and despite our best intentions we are likely to make matters worse. We need to Explore and Experiment instead.

Consultants selling solutions in complex systems are playing the same game as the commissioners who are buying those solutions. They know that you know that they can’t really promise a solution that will work because you both know that’s out of the sphere of your control due to the very nature of the problem. But they make promises with polish and certainty and you need someone with credibility to do the work and take the blame if it fails and so you both perpetuate this dance macabre. 

Instead we need to follow the alternative path, that of Explore and Experiment. We convene our team, whether in-house or with consultancy support, with citizens or service users, perhaps with input from all. In complex systems we start where we are and seek the potential for change in the present. Then we can proceed accordingly, exploring the situation, learning about it, gathering insight and evidence that can inform our next actions. We might test out a number of ideas and see how the situation changes. Indeed, this is the very real value of this approach. It’s the opposite of placing big bets: we place numerous small ones. This maximises the opportunities for learning from the feedback we gain and minimises the risks of catastrophic failure. It means we can test multiple ideas in a short period of time, each time adjusting them based on the insights. 

In my next article I’ll explore how we might know which approach to take so that we maximise the likelihood of success.