Ladder of Inference
What is the ladder of inference?
The ladder of inference is a model introduced by Chris Argyris that explains the often subconscious process we each go through when forming conclusions. While we may all observe the same event or look at the same set of observable data, we each select different data to pay attention to based on our own knowledge, perspective, and cultural background.
From that selected data, we then add meaning, make assumptions, and draw conclusions—which in turn influence our subsequent beliefs and actions. This model helps explain why two people viewing the same customer interview or situation can walk away with very different interpretations.
Why does the ladder of inference matter in product work?
The ladder of inference reveals a critical insight: so much of our understanding comes from projecting our own meaning onto data rather than observing what was actually said or happened. Often, what we hear in a customer interview is not what the customer actually said—it's what we think they said, filtered through our own perspective.
This matters because:
Teams can view the same data and reach different conclusions. When a product manager, designer, and engineer watch the same interview, they might each select different moments to focus on and interpret them differently based on their roles and experiences.
Disagreements often stem from different starting points. When team members disagree about what to build, they may have started by selecting different data from the same source, not because one person is right and the other wrong.
Projecting our assumptions can lead to misunderstanding customer needs. If we're not careful, we end up building products based on what we think customers meant rather than what they actually experienced.
How can teams use the ladder of inference more effectively?
Understanding where divergence happens on the ladder helps teams improve their discovery work:
Practice empathy to see data from the customer's or teammate's perspective rather than immediately filtering it through your own experiences. This helps at the critical first steps where you select data and add meaning to it.
Question your own filters by actively examining the ways you add meaning to what you observe. What assumptions are you bringing to the interpretation?
Make your thinking visible by explaining not just your conclusions but the data you selected and the meaning you added along the way. This helps teams understand where their interpretations diverge.
Seek to understand before being understood by asking teammates or customers to share their perspective before jumping to explain your own interpretation.
Learn more:
- How To Find Common Ground When Engineers Don't Like Features
- How To Develop Your Active Listening Skills
Related terms:
- Customer Interview
- Collaboration
- Confirmation Bias
Last Updated: October 25, 2025