One-on-one

Measures whether the first minutes established why you are meeting, what the other person should leave with, and how it connects to their work. Not whether a written agenda existed.

1. 5

   In the first two or three minutes you name the purpose, the outcome you both want, and what changed since last time. You set context (why it matters, what is fixed, what is open) and then hand the floor: “what’s on your list?” You show you read their notes or remember their last commitments.

2. 4

   Purpose and outcome are both stated, but one is implied rather than said, or you reach their list only after covering your own items.

3. 3

   An agenda exists, yours or theirs, but nobody says what the meeting is for. Items are topics (“the Ohio memo”), not questions or decisions.

4. 2

   The conversation opens on whatever comes up first. Nobody says what the hour is for; the other person has to guess.

5. 1

   You are visibly unprepared: you ask what was agreed last time, misremember their project, or spend the opening on logistics and your own news.

Evidence: CIPD 2023 (goal clarity r = .51 to .54; agenda presence r = .14 to .18); Rogelberg 2022; Hastings and Meyer 2020 (lead with context) for the language only. All sources

Measures whether your turns are short, on topic, headline first, and stop when the point lands.

1. 5

   Your turns are shorter than theirs. When you speak you lead with the point (“my read is X, here’s why”) and stop. Advice comes once, in one or two sentences. Any term they may not know is defined the first time it appears.

2. 4

   Mostly tight. One turn runs long or restates a point that already landed; the message survives.

3. 3

   You explain the same point two or three ways, or stack a second idea on the first before the other person can respond. Several turns run past the point.

4. 2

   Long monologues. The other person’s replies shrink to “yeah” and “makes sense.” Points blur into each other.

5. 1

   The other person cannot tell what you asked or recommended: they ask you to repeat it, or the topic changes without landing.

Evidence: CIPD 2023 (focused communication r = .28 to .59; avoid monologues); Kauffeld and Lehmann-Willenbrock 2012; Oppenheimer 2006. All sources

Measures whether every item leaves with an owner, a date, and a first step; whether the other person reads back their own commitments; and whether one named person decides anything still open.

1. 5

   In the last minutes you restate every commitment with owner, date, and first concrete step, and the other person says theirs back in their own words. Anything undecided has one named decider and a date it will be decided by. The next 1:1 is confirmed. You end on time.

2. 4

   Owners and dates on every item, but no readback, or one item is left as “let’s keep an eye on that.”

3. 3

   A list of actions exists but half lack a date or an owner. The close is “good, anything else?” rather than a summary.

4. 2

   Actions are implied in the conversation but never named as commitments. The meeting ends when time runs out.

5. 1

   Soft close (“great, let’s keep talking”). Nobody could reconstruct the commitments from the transcript.

Evidence: Gollwitzer and Sheeran 2006 (when/where/how plans, d = .65 across 94 tests); Härgestam et al. 2013 (closed-loop readback); Lehmann-Willenbrock et al. 2013; Hastings and Meyer 2020 (informed captain) for the language only. All sources

Measures how much of the meeting the other person speaks, whether your questions are open and build on their answers, and whether you paraphrase before you advise.

1. 5

   The other person speaks for most of the meeting (roughly two-thirds of the words or more). Your questions are open (“what have you tried?”, “what would good look like?”) and each builds on their last answer. You paraphrase what you heard before offering a view, and you let a silence sit. Any feedback you give names a specific moment, the behavior, and its effect.

2. 4

   The other person speaks more than you do. Questions are mostly open; one or two are leading (“don’t you think...?”). You paraphrase at least once.

3. 3

   Talk time is near even. Questions are mostly yes/no or status checks (“is X done?”). You move to advice before confirming you understood the problem.

4. 2

   You speak more than the other person. Questions are rhetorical or steer to your preferred answer. You interrupt or finish their sentences.

5. 1

   The other person gets short turns between your monologues; their concerns are deflected or turned into your topic.

Evidence: Itzchakov et al. 2018 (attentive listening raises the speaker’s clarity, lowers anxiety); Grant 2012 (solution-focused questions raise self-efficacy); ICF 2019 (Listens Actively, Evokes Awareness); Center for Creative Leadership SBI model, with Hastings and Meyer 2020 (4A feedback: actionable) for the language only. All sources

Measures whether the other person’s topics fill most of the meeting and whether they leave with next steps they set themselves. When the 1:1 exists to give them an update they asked for, the test shifts: the update should be headline-first and short, and the question “what do you need from me?” should get real time.

1. 5

   Their items come first and take most of the time; yours wait or go to email. They name their own next steps and you add at most one. Where they are stuck you ask what they would try before you propose. They leave with something they chose to do. In a status call: your update takes a third of the time, headline first, and their needs and reactions take the rest.

2. 4

   Their items dominate, but you set one of their next steps for them, or your one item ran long.

3. 3

   Half the time is your updates or your asks. Next steps exist but you dictated them.

4. 2

   The meeting is your status check: you run your list, they respond. Their topics get whatever time is left.

5. 1

   Their items never come up, or they raise one and it is parked for “another time.”

Evidence: Rogelberg 2022 (1:1s succeed when the report’s topics dominate); Theeboom et al. 2014 (coaching’s largest effect is on the coachee’s own goal management, g = .74); Grant 2012. All sources

How the total is built

Weighted = raw × weight × 20, with the weight as a decimal (a raw 5 at 20% contributes 20 points).

What is never scored

• Whether the other person’s work, plan, or decision was good. Score the conversation, not the work.
• Whether your advice was right. Score whether the other person was heard first and how the advice was delivered.
• Personal check-in time at the top. Rapport is part of a 1:1; count it against the score only if it displaces the other person’s items.
• Feelings inferred from a transcript. Score observable moves (paraphrase, open questions, talk share), not guessed emotion.
• Transcript artifacts, dropped connections, or the other person’s tangents.