First impressions are unreliable as verdicts but genuinely useful as data — if you know what they can and cannot show. Most people make one of two errors: they trust a first meeting too much, or they dismiss it entirely. The accurate position is somewhere in between, and it depends on which layers of a person are actually observable in that first hour.

Layer one: how they interpret new information

Within the first minutes, you can begin to measure how someone's attention selects and processes signals. When you introduce a new fact, a problem, or an unexpected change, watch what they reach for first:

  • Opportunity signals — "what could we build from this?"
  • Risk signals — "what could go wrong, and where?"
  • People signals — "who is affected, and how?"

The questions they ask in the first ten minutes reveal this layer more reliably than anything they say about themselves. Someone who asks "who else is involved" is showing a different default filter than someone who asks "what's the upside" or "what's the failure mode."

Layer two: how they position themselves

The second observable layer is stance: how they take space in the conversation and respond to hierarchy. Watch for the recurring pattern rather than the single moment:

  • Advance and claim — moves the agenda forward, appropriately asserts themselves, volunteers first.
  • Align and support — builds on others' points, makes the group's plan work.
  • Observe and wait — collects information before committing, speaks late and deliberately.
  • Define and defend — establishes boundaries early and maintains position under pressure.

One interaction is not enough to score this layer, but the direction is usually visible: how they enter the room, how they react when interrupted, how they handle a disagreement in the room.

Layer three: what they optimize over time

This layer is not visible in a first meeting, and pretending otherwise is where most professional misjudgment happens. What a person actually optimizes — autonomy versus security, speed versus durability, individual fairness versus system fairness — appears only in the pattern of tradeoffs they make across repeated decisions. Charm, articulateness, and confidence are not evidence of this layer. They are evidence of presentation skill, which is a different variable.

What does your own read run on?

Three questions — that's how long it takes to see the decision profile behind how you read people and trade-offs. TangoEra's free check maps how you weigh risk, people, and time horizons.

See my decision profile

The practical protocol

  1. Use the first meeting for layers one and two only. Note the signal class they reach for first, and the stance they default to. Log it as a hypothesis, not a conclusion.
  2. Design a follow-up loop for layer three. After three to five real interactions — or, better, after observing them make actual decisions under tradeoffs — check whether the pattern matches your initial read.
  3. Track decisions, not statements. The reliable data on layer three is what they chose when the options were both good and mutually exclusive. Statements about values are cheap; choices are data.
  4. Re-check the hypothesis after real stakes. First reads are calibrated by reality, not by more impressions. The second and third interactions matter more than the first, because they reveal consistency.

The useful question after any first meeting is not "what kind of person is this?" It is "what did this meeting actually measure?" If the answer is only layers one and two, then the rest is still data to be collected — not a judgment to be acted on.

Your data indicates the layers. The follow-up schedule is up to you.