Communication isn't one skill: reading behavior through relationship, flow, and rule

When a company tells a manager "you need to improve your communication skills," it has not really said anything. Communication skill is not one skill. The same person can listen well and then fall apart in a hard decision. They can show empathy and never bring a conversation to a close. They can know the policy and skip steps under pressure.

The trouble in most corporate training is rarely missing content. It is that behavior gets measured at too coarse a grain. The moment you say "good communication," measurement goes blurry. Ask instead whether the person paraphrased the other side's concern, kept the purpose of the meeting clear, recognized the escalation threshold, and the behavior becomes something you can actually see.

This is old ground. Aviation, surgery, anesthesia, and emergency medicine have assessed performance through observable behavior rather than vague traits for a long time. The umbrella term in that literature is non-technical skills: the cognitive and social skills that support safe and effective performance alongside technical knowledge. Flin, O'Connor, and Crichton's (2008) Safety at the Sharp End is the standard reference, and it treats communication, teamwork, leadership, situation awareness, decision making, and task management as separate behaviors to be observed and trained.

What we want to do here is adapt that behavioral-assessment logic to corporate conversations. EVRE's three-way distinction does not copy any single academic model. It is a simplified version of the same tradition, scaled down to three working dimensions for workplace conversations: relationship, flow, and rule.

Three dimensions, one conversation

Splitting behavior into observable sub-categories rather than personality is what aviation and healthcare have been doing for decades. The NOTECHS framework, developed for pilots, breaks non-technical skills into co-operation, leadership and managerial skills, situation awareness, and decision-making (Flin et al., 2003). ANTS, developed for anesthetists, uses task management, team working, situation awareness, and decision-making (Fletcher et al., 2003, British Journal of Anaesthesia). NOTSS, developed for surgeons, separates situation awareness, decision making, communication and teamwork, and leadership (Yule et al., 2006, Medical Education).

The categories are not identical across frameworks. But they all do the same thing: instead of a single "good pilot" or "good surgeon" score, they break behavior into observable parts and pull evaluation away from character judgement. EVRE's three dimensions are a corporate translation of that pattern, the same behavioral-decomposition logic applied to workplace conversations rather than a relabel of any one framework.

Carry that logic into a performance review, a hard feedback session, or a customer complaint and you end up with three questions:

• How does the person manage the relationship? What are they doing while someone is in the room with them?
• How do they structure the flow of the conversation? Can they hold the shape?
• Can they apply the rules under pressure?

The three dimensions are not independent in practice; they interact. But pulling them apart in training design matters. Replacing "weak communicator" with "loses the close of the conversation" gives the person something concrete to work on.

1. Relationship: what happens when the other person is in the room?

This dimension watches the person's behavior while in contact with the other party. It is not about "being warm." It is observable behavior:

• Can they paraphrase what the other party just said?
• Can they hold the conversation without dismissing the emotion in it?
• Can they take feedback without sliding into defense?
• Can they sit with silence instead of filling it immediately?
• Can they hold a limit without making the language harsher as pressure rises?

None of that is a personality trait. The AHRQ TeamSTEPPS framework treats communication, leading teams, mutual support, and situation monitoring as teachable team skills with training modules and assessment tools. The medical communication literature is more granular still. The Kalamazoo consensus statement, summarized by Makoul (2001, Academic Medicine), defines seven communication tasks: building a relationship, opening the discussion, gathering information, understanding the patient's perspective, sharing information, reaching agreement on a plan, and closing the encounter. Kurtz and Silverman's (1996, Medical Education) Calgary-Cambridge guide layers relationship-building and structuring the consultation onto that same task flow. Both make the same point: relating to the other person is a set of teachable tasks, not a fixed trait.

When this dimension is weak, the conversation tips one-sided. The other party's resistance grows; the conversation drags or breaks. The manager is rarely "ill-intentioned." More often they simply do not have a reflex that tracks contact quality.

2. Flow: where is the conversation going?

The flow dimension shows whether the person can manage the work structure of the conversation. Hard conversations are not only emotional; they are also a structure problem:

• Which topic comes first?
• If new information arrives, will the decision change?
• Where does the conversation land?
• What is still open?
• Who is going to do what, by when?

In aviation and anesthesia this lives under task management. ANTS treats it as a separate category, with sub-elements like planning and preparation, providing and maintaining standards, prioritising, and identifying and utilising resources (Fletcher et al., 2003). It is not a technical call; it is managing the structure, timing, and sequence of the conversation. The Calgary-Cambridge guide carries the same logic into a clinical interview by separating initiating, gathering information, explanation and planning, and closing as explicit phases of the encounter.

When flow is weak, the conversation drags, the topic scatters, the two sides leave with different conclusions. The manager may be well-meaning; the team still cannot get clarity out of them.

The strongest tool for training flow is not a one-shot explanation. It is running the same scenario over and over with different disrupting variables: new information arrives, time gets cut, the counterpart shifts position, priorities collide. What you are rehearsing is not the right sentence. You are rehearsing how to hold the shape when the shape keeps moving.

3. Rule: are the limits being held?

The rule dimension has less to do with whether the person knows the policy and more with whether they can apply it under pressure:

• What needs to be on the record?
• When does the conversation have to be paused?
• At what point should legal, HR, security, or a senior manager be brought in?
• Which promises should not be made?
• Which phrasing is risky under regulation or company policy?

This dimension matters most in compliance, customer complaints, harassment reporting, data protection, dismissal, and crisis communication. Picking the right rule in a calm module is not the same skill as applying that rule mid-argument.

The cleanest example of procedural assessment in the medical literature is OSATS. Martin and colleagues' (1997, British Journal of Surgery) Objective Structured Assessment of Technical Skill scores surgical procedures with both an operation-specific checklist and a global rating scale. What it tests is whether the surgeon applies the steps, in order, under load, not whether they can recite the rule. OSCE, the Objective Structured Clinical Examination (Harden & Gleeson, 1979, Medical Education), follows the same idea for clinical competence: performance is observed across structured stations with standardised patients, not inferred from a single global impression.

The strongest single piece of evidence for the rule dimension is the WHO surgical safety checklist. Haynes and colleagues (2009, New England Journal of Medicine), working across eight hospitals, found that introducing a 19-item checklist was associated with a drop in post-operative complications from 11.0% to 7.0% and in death from 1.5% to 0.8%. The authors do not claim a single lever; they read the effect as the product of several mechanisms working at once, team communication and systematic verification among them. For corporate training, the lesson is not that checklists fix everything. It is that making the critical steps visible reduces the chance they get lost under pressure.

Why these three dimensions are useful

The same conversation can succeed or fail along three different axes. A manager may have been respectful to the employee but left the conversation without a resolution. Another may have followed policy to the letter but used language so cold the other party's resistance grew. A third may have managed both the relationship and the flow well, but never raised a topic that should have been on the record.

That is why a single "good conversation" score is limited. The more useful move is to break the feedback into dimensions:

• Where did the relationship hold; where did it break?
• Where did decision clarity and closure weaken in the flow?
• Which limits were held in the rule dimension; which ones were missed?

That split also sharpens the training recommendation. Instead of "go work on your communication," you can say "work on closing the conversation and naming the decision." This is consistent with the core principle of the non-technical skills literature: assessment is anchored to observable behavior, not to character.

Reliable measurement has a precondition. Behavioral marker systems work because they rest on observable behavior, but for that to scale you need clear behavioral definitions, rater calibration, and ongoing validity checks. Tagging the transcript is not, on its own, a reliable measurement; behind it has to sit a behavioral dictionary, a calibrated rubric, and routine audit.

How does practice actually happen?

The hardest part of training these three dimensions is that no single explanation will move them. Relationship, flow, and rule are practiced, not discussed. In the medical and aviation literature this point sits on top of decades of work on deliberate practice. McGaghie and colleagues' (2011, Academic Medicine) meta-analytic comparative review found that simulation-based medical education combined with deliberate practice produced better results than traditional clinical education on specific competency outcomes. Their own caveat is worth keeping in view: the effect comes from careful design, targeted feedback, and graduated difficulty, not from the format itself.

A broad "communication workshop" rarely targets any of the three dimensions. Relationship benefits from short, repeated contact drills. Flow benefits from the same scenario rerun with disrupting variables. Rule benefits from applying procedure under emotional load. Running them as a single homogenized module fails to move any of them.

How EVRE uses this model

EVRE applies this distinction in corporate role-play simulations to make feedback more legible. The goal is not a single overall score; it is to surface, separately, how the participant managed the relationship, the flow, and the rules. That turns the development recommendation from "your communication is weak" into something trackable: "relationship held at the opening, the flow scattered in the middle, you missed the record-keeping threshold at the close."

The aim is not to copy any of the high-stakes assessment systems directly. It is to carry the same underlying principle into corporate conversations: evaluation is anchored to behavior, not to the person, and what can be observed can be trained.

EVRE's AI role play simulation turns these three dimensions into something practical for workplace conversations: every session produces separate feedback on relationship, flow, and rule, so the area of growth shows up as a concrete behavior.