Organization Design

Information Flow and Where a Company Goes Blind

Find the specific places your organization systematically cannot see (bad news, slow drifts, and anything that implicates the person reporting it) and install the three cheap mechanisms that fix each one.

  • Advanced
  • 14 min total
  • 14 chapters

What decision this helps you make: What to change about how facts reach you, so that the next problem arrives in week two rather than in the quarterly pack, and so that the person who spots it is not required to be brave.

What this topic is

Information flow is the set of paths by which facts about what is actually happening reach the people who decide. A company going blind is not a company short of data. Most are drowning in it. It is a company whose paths systematically drop certain classes of fact: unwelcome news, changes too slow to notice week to week, anything a reporter would have to implicate themselves to say, and everything nobody thought to measure.

Why it matters

The expensive organizational failures are almost never bad decisions made on good information. They are ordinary decisions made on a picture that had quietly gone out of date, or that never contained the relevant fact at all. Because each individual filtering step is honest and defensible (summarising, prioritising, waiting for the monthly cycle) the failure has no culprit, which is exactly why it recurs.

Who should learn it

Executives who keep finding out about things late and have started to wonder whether that is a coincidence, managers designing reporting for a growing team, and founders at the point where they stop personally seeing everything and have not yet replaced their own eyes with anything.

What you will understand

  • The four filters (compression, incentive, selection and measurement) and which one your blind spot is behind
  • Why exception reporting works and status reporting does not, and how to convert one into the other
  • How to price a detection lag in cash, using the numbers you already have
  • The specific mechanisms that get bad news to travel, and why asking for candour is not one of them

Prerequisites

Common misconception

"If we had better dashboards, we would see problems sooner." More data usually makes this worse, not better, because the limiting resource is attention rather than availability. A number sitting in a table nobody has a reason to look at is not visible, and a dashboard with sixty tiles is a place for a problem to hide in plain sight. The second and more damaging misconception is that bad news travels slowly because people are dishonest. It travels slowly because every honest person in the chain has a good reason to check first, gather context, and present it properly. Four honest people doing that in sequence produce a thirteen-week delay with no culprit anywhere in it.