---

title: Alignment states and interest scales description: The evidence bar for each alignment state, and the anchored 1 to 5 scales for panel conviction and candidate intent.

Alignment states and interest scales

The four states

Every row on the map gets exactly one state. The state is decided by the evidence, never by how confident anyone sounded.

State	Meaning	Evidence bar	Action priority
BLIND	One side raised something the other has not registered at all	A quote or behaviour from the side that raised it, AND confirmation the other side never mentioned it when asked	First. Always.
DIVERGENT	Both sides engaged with the same thing and read it differently	A quote or behaviour from each side, pointing in different directions	Second. This is the workable zone.
UNCLEAR	Someone has a feeling but no evidence	A comment with no example behind it ("something felt off")	Convert to a question before the next step. Never decide on it.
ALIGNED	Both sides read the same thing the same way	A quote or behaviour from each side that agrees	Note and move on.

Rules for classifying

• Panel versus panel counts too. Two interviewers who read the same answer differently is a DIVERGENT row, even before the candidate is involved.
• BLIND beats DIVERGENT on ties. If a row could be either, and the candidate has shown no awareness of the concern, it is BLIND.
• UNCLEAR is honest, not weak. A gut feel from an experienced interviewer is worth chasing. It is not worth deciding on. The move is always to turn it into a specific question.
• Contaminated rows. Feedback given only in a group discussion, after a senior voice spoke, is tagged (group) on the map and cannot on its own make a row ALIGNED.

Examples

• Interviewer: "They never mentioned a number without me asking." Candidate: "I think I made a strong case on results." BLIND. The candidate thinks the results landed. The panel thinks they were withheld.
• Interviewer A: "Great at discovery, asked five layers deep." Interviewer B: "Talked over me in the role play." DIVERGENT. Same session, two reads. Likely both true: strong questioning, weak listening under pressure.
• Interviewer: "Not sure they'd fit the team." No example. UNCLEAR. The next move is "What would you have needed to see to be sure?"

Panel conviction scale

Rate the panel as a whole, then note any single interviewer sitting two or more points away from the rest. That outlier is usually the most important row on the map.

Score	What it sounds like	What it means
5	"I want them. When can they start? I'd be annoyed if we lost them."	Hire. Move today.
4	Specific praise, names one real concern, wants to see them again	Strong. Broker the concern and move.
3	"Solid. Good. Could work." Some specifics, no urgency	Undecided. Needs new evidence, not more time.
2	Praise with no examples. "Let's see who else is out there."	Slow no. Name it and force it.
1	Clear specific objection, or refuses to rate	No. Close the loop respectfully and quickly.

The disappointment test. When the panel sits at 3, ask each interviewer: "Would you be disappointed if this person accepted another offer this week?" Anyone who hesitates is not a yes.

Candidate intent scale

Score	What it sounds like	What it means
5	Asks about start dates, the team, the first 90 days. Tells you where you rank.	Committed. Protect against a counter-offer.
4	Specific enthusiasm tied to the role, names one open question	Keen. Answer the question fast.
3	"Really interesting, I'd like to learn more." Asks about package first	Open, not sold. Find the real motive.
2	Slow replies, vague on timing, mentions other processes without detail	Drifting. Ask directly where you stand against the others.
1	Reschedules, goes quiet, or says they are "keeping options open"	Gone or going. Ask once, directly, then release.

Ratings are always supported by a quote. A score without a quote is removed from the card.