Skip to content

Towards solving difficult problems in complex systems

  • Identify the problem. In complex systems the first problem you identify is probably just one of several symptoms of a network of contributing problems.
  • Become an expert: Learn as much about the problem-network as possible. In particular, spend time on the ground with the people who the problem affects, and where you can observe the key contributing factors at play.
  • Look at the wider system: what are the immediate causes of the problem? What process create the conditions for those causes? What are the key relationships? What is incentivised?
  • Look at the obvious solutions: find out why these haven’t worked.
  • Find friends: look for others who care about the problem. Share information. Look for ways to work together. If you can, build a team.
  • Start small: identify the smallest part of the problem that you might be able to address.
  • Look for leverage:  identify the fewest number of things you could do that might enable you to make things better.
  • Make it easy: try your first-pass solution in as close to ideal conditions as possible.
  • Spot resistance: what’s working against your solution? What pushes back harder the harder you push? Addressing what gets in the way of the mission is part of the mission.
  • Make it repeatable: if the solution seems to work, try to do it again.
  • Make it financially sustainable: is there a way that the people who benefit from the solution could pay for it? Ask people who profess to care about the problem to show you the money.
  • Move towards the margins: to work out how to scale your working solution, try it in the next-most-favourable context you can find. Find out what would make it work for the user who couldn’t quite afford it last time, or didn’t quite have the skills to be able to take advantage of it.
  • Build the wrapper: the “whole product” includes the system that administrates the solution.
  • If it’s still working, make it reproducible: share the word, write the handbook, make the training materials. Find others who can do it in parallel (or instead of) you.
  • Don’t expect it to be quick or easy: play the long game, have faith, be prepared to solve new problems of your own making, learn to play hurt.
  • Whether you succeed or fail – be prepared to do it all over again.

I'd love to hear your thoughts and recommended resources...