BookYol

Concept

Dragonfly Eye

Dragonfly eye is the habit of building a judgment from many different perspectives, such as data, outsiders, critics and other models, and blending them into one view, the way a dragonfly's compound eye fuses thousands of lenses into a single image.

Philip Tetlock's research on superforecasters found that the best predictors rarely trusted a single lens. Any one viewpoint has blind spots you can't see from inside it, whether it's your own expertise, your team's mood or one striking data point. A dragonfly's eye has up to 30,000 facets. Each one catches a slightly different slice of the world, and the brain stitches them into one picture. Applied to judgment, this means deliberately collecting views unlike your own: the base rate, the skeptic's case, the person closest to the customer, the analyst who disagrees. Then you weigh them and land on a single number. This isn't consensus or splitting the difference. Some lenses deserve more weight than others, and the blend should shift only as far as each new view actually earns. Done well, it guards against two errors: overreacting to one vivid fact and ignoring the inconvenient one.

A real-life example

Maya, a product lead, puts a 70% chance on her app shipping by March. Before committing, she asks three people who see things differently. An engineer flags an untested payment integration. A support rep says similar launches slipped twice. A finance analyst notes that past projects ran 40% over schedule. None of them alone changes her mind. Blended, they move her to 45%, and that's the number she gives leadership, with her reasons.

How to use it

  1. 1Before settling on a view, list at least three perspectives that differ from yours: the outside-view base rate, a skeptic, and someone closer to the ground.
  2. 2Write down what each lens predicts and how much weight it deserves. Then combine them into one probability instead of picking a favorite.
  3. 3When a new view or fact arrives, adjust the blend in proportion to its strength: a small nudge for a weak signal, a bigger shift for strong evidence.