My number one piece of advice for new bloggers: Write whatever you would want to read if you weren’t writing it. Later on, you can worry about things like your writing style, scope, target audience, blog layout, etc. But to get yourself started, you have to stick with what you’re interested in writing and make a habit of it.

Well friends, I’ve fallen out of the habit recently. As the year starts, I’ll try taking my own advice in the hopes of getting back into it.

My professional and intellectual interests have always hovered around politicsmanagement and complexity, so I’m trying to focus my writing on those topics over the next few months. My current job offers a lot of food for thought, especially at the micro-level (individual manager or organization) but also at the system-level (aid sector, governmental, international, etc.). I won’t discuss my work directly though. My goal is to learn and process as much as share. I hope the result is interesting.

(Why am I telling you that I’ll be writing about this? Commitment mechanism! Some people start the new year by paying for a 6-month gym membership. I start by telling everyone that I’ll be blogging more. Let’s hope this works better than most people’s gym memberships.)

To kick things off, here’s a brain dump of ill-formed thoughts on the topic. I claim none of them as original.

  • Decision-making in situations of complexity requires checking our biases and also understanding nuance, which can be accomplished only through quantitative data that is informed qualitative data.
  • Induction is as important as deduction.
  • Intuition is unreliable without regular prior experience in similar situations — which is unlikely in a context of complexity.
  • Frequent communication among colleagues/partners is critical, as is direct contact with one another’s work (i.e. get out of the office to see what’s actually happening).
  • People management is hardest: select the right people and then macro-manage them. Give them flexibility within a framework.
  • Micromanaging tends to trickle down. Avoid it.
  • You get what you measure. But because what you measure is never more than a proxy for what you want, evaluation systems must be rigorous yet dynamic.
  • Pay attention to accountability relationships, types of accountability, and networks of accountability.
  • Death-by-committee and the heartbeat principle: In complex situations, it’s easy for individuals/organizations to avoid accountability (just look at US politics). Unless there is one person who is responsible for it (whose heartbeat depends on it) then it won’t happen.
  • Design feedback loops with selection processes.
  • There are no root causes. At best, there are points of leverage.

That’s all for now. You’ll see a guest post soon that relates to these questions, specifically on the topic of feedback loops and accountability. More coming after that.


Related reading: See Ben Ramalingam’s excellent blog Aid on the Edge of Chaos or Owen Barder’s blog, both of which cover these topics frequently. Or check out any of the following past posts on Find What Works:

  1. Looking forward to read the developments of these ideas. Specifically where you mention intuition as unreliable – that sounds intuitively (ha!) wrong to me, and my gut hasn’t failed me yet.

    Good luck with the blogging resolution!


  2. That was an idea that came out of Daniel Kahneman’s book “Thinking, Fast and Slow.” His argument for it is a bit academic though. Whenever I get around to writing on it, you’ll have to let me know if it’s convincing!


  3. I am also looking forward to seeing how you develop the above ideas, particularly the last one: There are no root causes. At best, there are points of leverage.

    I had a somewhat mixed reaction as I read that statement. I wonder if the two are mutually exclusive and if it depends on how you define “root”. As I said, I will look forward to a fuller treatment of the subject and hope to add my 2 cents.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: