When something feels off with your team, the room may be quieter, estimates receive less challenge, or handoffs start slipping. People still say they are fine, yet the ordinary rhythm of the team has changed.
The dangerous move is to convert that pattern into a story: the team is disengaged, somebody is undermining you, or a resignation is coming. The equally dangerous move is to wait for a crisis because you cannot yet prove anything.
What you'll learn
- How to use intuition without treating it as evidence.
- A seven-day sense-to-evidence check for small teams.
- Questions that invite useful information without accusing anyone.
- A decision rule for monitoring, acting, or escalating.
- An observation-to-action log that avoids covert surveillance.
What does it mean when something feels off with your team?
Useful intuition is often pattern recognition: you know what ordinary challenge, pace, humour, and recovery look like in this team. But subjective confidence is not accuracy. People learn reliable intuition only in environments with repeatable patterns and timely feedback. Rare people events in a small company rarely provide that clean learning environment.
Which observations are specific enough to check?
| Vague story | Observable version | What remains unknown |
|---|---|---|
| The team is disengaged | Three risks that were usually raised in planning appeared only after work was late. | Whether the cause is workload, safety, role clarity, timing, or something outside work. |
| People are hiding conflict | Two handoffs were returned without the usual direct discussion between owners. | Whether there is relationship tension, unclear standards, or simple scheduling friction. |
| Energy has collapsed | Response time, rework, and after-hours messages changed during the last two weeks. | Whether the change is temporary, workload-related, personal, technical, or market-driven. |
| Nobody challenges me | The last four decision meetings produced no alternatives, questions, or recorded risks. | Whether people agree, lack context, expect challenge to fail, or prefer another channel. |
If your first sentence is still a judgement, use the same discipline as the guide to replacing vague manager feedback: name the gap, not the vibe. You are preparing to learn, not preparing a case.
How do you test more than one explanation?
- 01
Write the observation
Use dates, decisions, delays, rework, missed handoffs, changed participation, or repeated blockers. Remove claims about intent.
- 02
List competing explanations
Consider demand and capacity, unclear decisions, relationship friction, low voice safety, a project phase, an external constraint, and observer bias.
- 03
Name confirming and disconfirming evidence
Decide what would make each explanation more or less likely before you start asking questions.
- 04
Use data you already have
Review work flow, rework, blocked tasks, quality, absence, escalation timing, and follow-through. Do not create a secret monitoring system.
- 05
Check the leader variable
Review your last three responses to disagreement, mistakes, and bad news. Defensiveness or inaction can change what people bring you.
What should you ask without creating a fishing expedition?
A useful opening is: 'I have noticed two late handoffs and fewer risks raised during planning over the last fortnight. I may be connecting the wrong things. What has changed in the work, the decisions, or the support around it?'
| Area | Ask | Avoid |
|---|---|---|
| Workload | Which demand has changed, and what is being displaced to absorb it? | Why can nobody cope with the workload? |
| Decisions | Where are decisions returning, waiting, or being reopened? | Who is blocking the team? |
| Voice | When did you last hold back a risk or disagreement, and what made that reasonable? | What are people not telling me? |
| Support | What would make good work easier during the next two weeks? | Is everybody okay? |
If problems routinely arrive late, inspect the response pattern described in why teams bring problems too late. An open invitation does not undo a history of arguing, punishing the messenger, or failing to act.
How much should you trust silence in a small team?
Employee voice works through complementary routes: a manager conversation, a written channel, a team review, a trusted alternative contact, and a formal route when needed. One weekly meeting or one pulse does not prove that people had a usable way to raise the issue.
What should change during the first seven days?
| Point | Action | Output |
|---|---|---|
| Day 1 | Separate observations from interpretations and list competing explanations. | A short evidence log without motive or personality labels. |
| Days 2-3 | Check existing work signals and audit your recent reactions to challenge and bad news. | Evidence that supports or weakens each explanation. |
| Days 3-5 | Hold specific individual or team conversations and offer another safe route where possible. | Work conditions, friction, and uncertainty in people's own terms. |
| Days 5-7 | Choose one low-risk change, name an owner, and set a review point. | A testable action rather than a general promise to communicate better. |
If you ask for wider feedback, follow the rule in employee surveys without action: report what you heard, what will change, what will not, and when you will revisit it. Asking without movement teaches people that silence is efficient.
When should you monitor, act, or escalate?
| Response | Use it when | Example |
|---|---|---|
| Monitor | The signal is slight, consequences are low, and more normal work will clarify the pattern. | Track whether a temporary planning lull returns to baseline after the deadline. |
| Act | A reversible work gap is visible even though the deeper cause remains uncertain. | Clarify one decision owner, restore a missed handoff, or set a WIP limit. |
| Escalate | Facts indicate immediate risk, protected concerns, serious conduct, or expertise beyond the manager's remit. | Use the relevant HR, safeguarding, occupational health, legal, or emergency process. |
For low-risk decisions, use the latest safe decision point. Waiting for better evidence is responsible only while the team remains safe and the delay does not make the damage harder to reverse.
What belongs in an observation-to-action log?
- 01
Observation
Record the changed work pattern, timing, and consequence without a motive, diagnosis, or character judgement.
- 02
Alternatives
List plausible work, relationship, external, and observer explanations so the first story cannot dominate.
- 03
Safe evidence
Use existing work records and voluntary conversations. Do not inspect private messages, health information, or protected personal data.
- 04
Leader contribution
Name the decision reversal, hidden standard, pressure response, or missed follow-up that you can change.
- 05
Action and review
Choose one small operating adjustment, assign it, and decide what movement would make you continue, change, or stop.