This seems an obvious suggestion when written down, but it's usually not followed - which is why basic and fundamental things like 'human centered design' are so radical to the public service.
Government approaches delivering social value quite differently to how a startup does. A social enterprise startup may have a high-level social objective, but it will fail if it tries to achieve everything at once (known as 'boiling the ocean'). No matter how central that objective may be, it needs to start by being successful delivering value at the smallest scale, to a small set of individuals. For a startup, this is an important way to validate hypotheses about the business model - and in particular the experiences and interactions - minimise risk, and progressively seek funding for expansion. It also fosters a completely different perspective on social change to that of the bureaucrat.
There is no one approach to developing policy that will foster a better understanding of impact at all different scales. Enterprises scale up because they need to, they operate and grow very differently to government, and the different actions of government can't be shoved into a single model for implementation. Nevertheless, there should be a better way to make policy than to design a magnificent national system, then fill in the unpredictable and inconvenient gaps known as 'people'. It may be hyperbolic, but this is closer to reality than it should be.
Some elements of human centered design can definitely help, particularly to understand how individuals interact with numerous different government systems (think MindLab). But in other contexts, it might be more useful to rethink how programs are implemented, inspired by enterprise approaches and led by the need to understand impact at different scales. A program will first test the most central hypotheses that are the easiest and lowest risk to trial - not necessarily the smallest scale - focus on getting this right, and progressively build on the growing foundation.
Thinking like this requires a conscious approach to risk, iteration and learning-through-doing that is confronting, unsettling - far from straightforward.
Some elements of human centered design can definitely help, particularly to understand how individuals interact with numerous different government systems (think MindLab). But in other contexts, it might be more useful to rethink how programs are implemented, inspired by enterprise approaches and led by the need to understand impact at different scales. A program will first test the most central hypotheses that are the easiest and lowest risk to trial - not necessarily the smallest scale - focus on getting this right, and progressively build on the growing foundation.
Thinking like this requires a conscious approach to risk, iteration and learning-through-doing that is confronting, unsettling - far from straightforward.
Nevertheless, thinking about 'progressive' policy as an option, to deliver a broad-impact policy - particularly on a difficult issue - without testing hypotheses about how it works for actual people would be crazy. It's a large investment and has high risk.
Flicking back over old posts, I realised this one was a bit limited.
ReplyDeleteThe point is still completely valid (I am somewhat surprised that so much of what I wrote so long ago still is!)
But it's only half the point.
The other half is that we fail to understand how impact scales UP from micro interventions to the whole, and take a holistic perspective.
So the real problem, it's not really about looking up or looking down, but about thinking through implications through different scales - understanding how actions work and are meaningful at all scales, to develop a genuine picture for how our actions will influence the world around us.
This depth of understanding is of course generally really hard, if not impossible!
But we still need to aim over time to do so.
Because without it we are ignorant of what we are really achieving.