Check your interpretation

At Testagility, we're big fans of Virginia Satir, in particular her Interaction model & how it helps us improve our communication.

The model suggests ways interactions can go astray & ideas for getting them back on track. Check out this great resource for more details on the model.

During the interaction, we can check back with the sender what they sent in order to try & prevent the interaction off track.

This post came about after a conversation I had with a client focusing on the different ways we receivers can misinterpret a message (in the Meaning stage).

Self (receiver)

  • Our understanding of a definition or Acronym (e.g. TDD, BDD, DevOps) differs from sender
  • Missed translations from languages other than our own
  • Interpretations from a message we mis-heard / mis-read 
  • Our relationship with the sender skews our interpretation of the message they're sending
  • We're having a bad day which leads us to jump to conclusions & incorrect interpretations 

Others (senders)

  • Senders understanding of a definition or Acronym (e.g. TDD, BDD, DevOps) differs from ours
  • Different translations of our language
  • Senders relationship with us skews the message they're sending
  • Sender is having a bad day which results in a different message being sent to the one that was intended

Context (situation)

  • A stressful encounter prevents us from thinking clearly (e.g. blame game in meetings)
  • Noisy environments leading to obscured messages being received
  • Cocktail party effect prevents us from attending to the current interaction
  • Rushed exchanges


  1. Check back on the intake you received & see if you make the same interpretation again
  2. Check back with the sender on the interpretation you've made based on the intake to see if that's what they meant
  3. Think of 3 interpretations you think the sender could have meant
  4. Goldilocks - play back 3 interpretations; best meaning, worst meaning & a feasible meaning 

You can find out more about the Interaction Model here: