Monday morning. You approved the project plan, confirmed the timeline, told the team you trusted them to run with it. By Friday, you have no signal. No red flags, but no data either. You didn’t want to hover. You wanted to give space.
Now it’s the following Monday and you’re sitting in a status meeting, and the update feels off. Not wrong, exactly, but lighter than expected. Someone mentions a dependency that wasn’t in the original plan. Another person phrases their progress in terms of effort (“we’ve been working hard on it”) rather than output (“here’s what shipped”). You don’t have evidence of a problem. You have a feeling. And you have zero data from the last five business days to tell you whether that feeling is right or a projection.
This is what we’ll call the monitoring gap: the space between setting expectations and the next formal checkpoint where no active observation happens. Most performance management advice focuses on the checkpoints (the reviews, the one-on-ones, the retrospectives) and skips the space between them entirely. That space is where calibration drift starts, where small misalignments compound, and where the feedback you eventually give arrives too late to be useful.
Monitoring and feedback, as a daily skill, lives in that gap. And for most people at work, the gap is wide open. This post focuses primarily on managers (who monitor teams), but monitoring and feedback is a dual-track skill. If you’re an individual contributor, the same calibration loop applies to your own work: tracking your quality, seeking feedback proactively, and giving useful input to peers. We’ll cover the IC angle later in the post.
What monitoring actually means as a daily skill
Monitoring sounds like watching, but the two are fundamentally different. Watching is passive. Monitoring is comparative: you’re holding what you see against what you expected to see, and updating your mental model when the two don’t match.
Maintaining a reference point
Every person on your team has a behavioral baseline: how they communicate when things are on track, how they show up in meetings when they’re confident, what their work cadence looks like during a normal sprint. You built this reference point through exposure, even if you never wrote it down.
Monitoring starts there. When you check in, you’re comparing what you observe against what you’d expect from this person in this context. Without a reference point, you’re just collecting impressions. With one, you’re reading signal. This is why setting clear expectations matters so much. You can’t spot deviation if the expectation was never specific enough to serve as a baseline.
The three things managers track, and the one they miss
Most managers, if they’re paying attention at all, track three things:
- Task progress. Is the work moving? Are deadlines being hit? This is the easiest to observe and the most common default.
- Team dynamics. Are people collaborating well? Is there tension? Who’s pulling away from group conversations?
- Energy and engagement. Does the person seem invested? Are they showing initiative or just completing assignments?
All three are valid signals. But there’s a fourth category most managers miss entirely: calibration drift, the slow, incremental shift in a person’s baseline that no single observation would flag. We’ll come back to this one. It’s the most important signal and the hardest to catch.
The calibration loop
The calibration loop is the process that turns perception into useful action (or useful inaction). Four steps, and the third is where most managers make their biggest mistakes.
Observe, compare, decide, act
- Observe. Collect data through structured channels (one-on-ones, project updates) or ambient ones (Slack comments, standup tone, who’s asking questions and who isn’t).
- Compare. Hold what you observed against the reference point. Expected behavior? Note it and move on. Unexpected? Flag it internally.
- Decide. A mismatch does not automatically mean you should give feedback. Is this a data point or a pattern? One quiet meeting is a data point. Three in a row is a pattern. The decision here is whether to act or whether to update your model and keep watching.
- Act. The action isn’t always feedback. Sometimes it’s a question. Sometimes it’s adjusting your support. Sometimes it’s realizing your expectation was off and the person is fine. The loop doesn’t always produce output. Sometimes it produces a better-tuned instrument, which is you.
Signal threshold: when observation becomes feedback-worthy
A useful rule: don’t give feedback on a single data point unless it’s a safety or ethics issue. Everything else gets the pattern check.
A team member missed a detail in a deliverable. That’s a data point. They’ve missed similar details in three of the last four deliverables. That’s a pattern. The first gets filed. The second gets a conversation.
Acting on single data points is what makes monitoring feel like micromanagement: flagging every small thing, every time, regardless of context. Acting on patterns is what makes it feel like support: “I’ve noticed something over the past few weeks, and I want to talk about it.” The difference comes down to signal threshold, the bar for when observation becomes action.
Two failure modes
When monitoring and feedback breaks down, it almost always falls into one of two patterns. Both come from a reasonable instinct taken too far.
Under-monitoring: the “autonomy disguise”
This one is common among managers who value trust. You set the expectation, delegate, and step back. Reasonable instinct, but trust without calibration is abdication. You can trust someone completely and still maintain a reference point. Trust means you’re not second-guessing their decisions. Calibration means you’re paying enough attention to know whether the conditions around them have changed.
Gallup’s research found that 70% of the variance in team engagement is attributable to the manager. Under-monitoring shows up as a slow fade: the team member who needed a course correction three weeks ago didn’t get one, the dynamic building between two people went unaddressed, the drifting project kept drifting.
Managers who under-monitor discover problems only at the performance review stage, where feedback feels retrospective and the person feels blindsided. It’s also where bias creeps in, because you’re reconstructing a picture from sparse data instead of referencing a pattern you watched develop.
Over-monitoring: when support becomes surveillance
Same behavior (observing, checking, asking questions), but the frequency, intent, or relationship context turns it from support into control.
Deci and Ryan’s Self-Determination Theory explains why. When people perceive monitoring as controlling, it undermines autonomy and intrinsic motivation. When they perceive the same monitoring as informational (“someone checking in because they care”), it supports autonomy. A manager asking “where are we on the deliverable?” can land as genuine interest or suspicion. The difference sits in three things:
- Frequency. Asking once a week is informational. Asking every morning is controlling, unless the project is in a crisis state that justifies it.
- Context. Asking after a known risk point (“I know the vendor deliverable was due yesterday, how did that land?”) is contextual. Asking with no specific trigger (“just checking in”) too often reads as distrust.
- Relationship. A manager who has built trust over months can observe closely during a crunch without it feeling invasive. A new manager doing the same thing with a team that doesn’t know them yet will trigger defensiveness.
Over-monitoring costs you the thing you need most: honest signal. People who feel surveilled start managing their visibility instead of managing their work. They give you the update you want to hear, not the update you need.
The calibration drift problem
This is the signal most managers miss, and it’s the one that causes the most damage precisely because it’s invisible.
Every person has a behavioral baseline
You probably already know your team’s defaults. The person who always has three questions ready in planning. The one who sends detailed Slack messages with bullet points. The one who ships slightly ahead of deadline. The one who pushes back on scope early and then commits fully.
These patterns form a baseline. You never formalized it, but you know it from exposure. That baseline is your most sensitive instrument for detecting change.
Drift signals vs. noise
Calibration drift is what happens when that baseline shifts incrementally, so slowly that no single observation triggers concern.
The person who used to have three questions ready now has one, or none. The detailed Slack messages get shorter. The early shipper starts hitting deadlines exactly, then starts asking for extensions. Each change, taken alone, looks like nothing. A quiet day. A busy week. A distraction.
The question that separates drift from noise: is this a trend or a moment?
One off day is noise. Three weeks of gradually reduced engagement is drift. The tricky part is that you can only see the difference if you’re comparing against the baseline, not against yesterday. Yesterday’s slightly-quieter version of this person looked normal because the day before was only slightly quieter than the day before that.
This is why Amabile and Kramer’s research on The Power of Small Wins matters for managers, not just for the positive side (small wins compound into sustained motivation), but for the inverse. Small losses, small withdrawals, small reductions in the conditions that keep someone engaged, compound too. And just like small wins, small losses are invisible unless someone is tracking them.
The manager who catches calibration drift early can have a low-stakes conversation: “I’ve noticed a shift over the past few weeks. Nothing alarming, I just want to understand what’s going on.” The manager who misses it has a high-stakes conversation six months later when the person is disengaged, looking for another role, or underperforming in a way that now requires formal intervention.
Designing your monitoring rhythm
Knowing what to watch for is half the skill. The other half is building a rhythm that makes consistent observation sustainable without turning your calendar into a surveillance apparatus.
Different people, different frequencies
One of the most common monitoring mistakes is applying a uniform cadence to every person on the team. Weekly one-on-ones for everyone. Same format, same questions, same depth.
The problem: different situations demand different levels of attention.
- New hire (first 90 days): High-frequency structured check-ins. They’re building the baseline you’ll compare against later. You need to understand their normal before you can detect their drift.
- Senior person on a familiar project: Lower frequency, more ambient. They have a long track record, and the reference point is well established. A brief check-in and some Slack observation may be enough.
- Anyone on a stretch assignment or during a difficult period: Increased frequency, but with explicit framing (“I’m checking in more because this is a big project, not because I’m worried about your work”).
- Struggling project with visible risk: High frequency, higher structure. This is where closer monitoring is justified, and where being transparent about why prevents the surveillance read.
Structured touchpoints vs. ambient observation
Your monitoring system has two channels, and you need both.
Structured touchpoints are scheduled. One-on-one meetings are the most obvious example: predictable space for signal exchange where both sides can prepare. But structured touchpoints have a limitation. People perform differently in formal settings. The person who seems fine in a one-on-one may be struggling in the daily work.
Ambient observation fills that gap: how someone shows up in group meetings, the tone of their Slack messages, whether they’re asking for help or going silent. This is where active listening becomes critical. You’re noticing patterns in how people communicate, what they mention and what they avoid, and where the gaps are between their formal updates and their informal signals.
When both channels align, your reference point is confirmed. When they diverge, that divergence is itself a signal worth exploring.
From observation to feedback: the pattern check before you act
You’ve observed something that doesn’t match the baseline. Before you turn that into constructive feedback, run the pattern check:
- Have I seen this more than once? A single instance is a data point. Multiple instances are a pattern.
- Is this about the situation or the person? The team member who’s quieter than usual might be processing a difficult client interaction, not disengaging. Context matters.
- Do I have enough data, or am I filling gaps with assumptions? If the answer is assumptions, collect more data before acting.
- What does this look like from their perspective? The behavior you’re reading as disengagement might look like focus from their end. Test your interpretation before acting on it.
If the pattern check confirms a real signal, that’s when observation becomes feedback that matters. And because you’ve been calibrating all along, the feedback arrives with specifics (“I’ve noticed X across three situations over the past two weeks”) rather than vague impressions (“I feel like you’ve been a bit off lately”).
The specifics are what make real-time feedback land. Not the speed. The precision.
If you’re recognizing gaps in how you read and respond to team signals, a conversation with Merlin can help you pinpoint where your calibration breaks down and what to adjust first.
The IC side: monitoring your own work
Everything above frames monitoring from the manager’s chair. But monitoring and feedback is a dual-track skill, and the IC version is just as demanding.
When you’re an individual contributor, you are both the monitor and the performer. Nobody else is watching your quality indicators day to day. Nobody else notices when your output starts drifting from your baseline. The five behaviors that define IC-level monitoring and feedback are: tracking the quality of your own work, giving useful feedback to peers, actively seeking feedback from others, converting the feedback you receive into visible changes, and correcting course before someone else has to tell you.
Quality tracking: your own calibration loop
The IC version of the calibration loop points inward. Instead of “is my team member’s baseline shifting?”, the question is “is my own work meeting the standard I set, or am I drifting?” This means maintaining your own quality indicators (not just task completion, but whether the work is actually good), checking them regularly, and adjusting before the gap becomes visible to others.
A practical version: at the end of each week, review your three most important outputs. Were they at your standard? If you submitted them last month, would you be proud of them? If the answer is “they were fine,” that might be the early signal of drift. “Fine” is where calibration starts slipping.
Peer feedback and feedback seeking
ICs who are strong at monitoring and feedback don’t wait for their manager to surface problems. They ask colleagues: “What am I missing in this analysis?” or “Did that handoff give you what you needed?” They also give peer feedback when they see something drifting in someone else’s work, because waiting for the manager to notice costs the whole team time.
The skill is the same whether you’re monitoring a team or monitoring yourself and your peers. The calibration loop (observe, compare, decide, act) runs the same way. The only difference is the direction.
Feedback uptake and self-correction
Seeking feedback is half the loop. The other half is doing something visible with it. Feedback uptake means converting input into a concrete change: adjusting a process, reworking a deliverable, shifting a communication pattern. The person who gave you feedback should be able to see the change without you announcing it.
Self-correction is the version that doesn’t require outside input at all. You catch your own drift, diagnose the cause, and adjust before anyone needs to flag it. This is the IC equivalent of a manager noticing a team member’s output has shifted and intervening early. The difference is that you’re intervening on yourself.
The strongest ICs close the loop publicly. When someone gives them feedback and they act on it, they circle back: “You mentioned the handoff docs were missing context. I restructured the template. Does this version work better?” That single move (act, then confirm) builds the kind of trust that makes peers more willing to give honest input next time.
Self-assessment: are you calibrated?
Before working on your monitoring system, it helps to know where you actually stand. Four questions to surface the most common blind spots.
Four audit questions
- Can I describe each team member’s behavioral baseline in specific terms? Not “she’s a strong performer.” Can you describe their default communication patterns, their meeting energy, their work cadence? If you can’t, your monitoring has no reference point.
- When was the last time I noticed something before it became a problem? If your feedback consistently arrives after things have gone wrong, your cadence is too slow or your signal threshold is too high.
- Do I adjust my monitoring frequency based on the situation, or do I treat everyone the same? Uniform monitoring feels fair but isn’t effective. Same format for a new hire and a five-year veteran means you’re under-monitoring one and over-monitoring the other.
- Would my team describe my check-ins as support or surveillance? The clearest test: do people share problems with you early, or do you find out about them late? People who feel supported bring concerns forward. People who feel watched hide them.
Take the assessment
If those questions surfaced some gaps, a structured assessment can help you see the full picture. Risely’s monitoring and feedback assessment gives you a baseline score across the core behaviors that make up this skill, from signal-reading to feedback timing to cadence design. It takes a few minutes and gives you a specific starting point rather than a general sense of “I should probably get better at this.”
If you’re finding that the observation side is solid but the response side (coaching conversations, developmental feedback, helping people act on what you’ve noticed) is where things stall, Risely’s coaching skills assessment covers that complementary skill set.
Back to Monday morning
Same scenario. You approved the project plan and trust the team. But this time, you designed the week differently.
Tuesday, you read the project channel’s last 24 hours. The conversation was active, the right questions were being asked, one person flagged a dependency early. Reference point held. No action needed.
Wednesday one-on-one with the project lead. One calibration question: “What’s the one thing that could knock this off track this week?” The answer told you more than a status report would have.
Thursday, you noticed a usually-active team member had gone quiet for two days. Single data point. Filed it.
Friday standup, same person was back to normal. Noise, not drift. Model updated.
By the following Monday, you weren’t starting from zero. Five days of low-effort, high-signal observation gave you context. You could ask precise questions instead of vague ones. That’s a calibrated week. Not more meetings. A manager whose instrument is tuned.
Monitoring and feedback isn’t a performance management process. It’s a perceptual skill that makes every other management skill work better. The feedback is only as good as the observation behind it, and the observation is only as good as the reference point it’s compared against.
If you want to build this skill with structured support, try a conversation with Merlin. Risely’s AI coach can help you identify where your monitoring gaps sit and build a cadence that fits your team’s actual needs. Not a template. A calibration tuned to your context.
Start by noticing what you’re not noticing. That’s where the skill begins.
Frequently asked questions
What is monitoring and feedback as a management skill?
Monitoring and feedback is the practice of actively observing your team’s work patterns, comparing what you see against expected baselines, and deciding when that observation warrants a conversation. It’s a perceptual skill before a communication skill. The monitoring side involves reading signals (task progress, team dynamics, energy shifts) in real time. The feedback side is the output, but only when the pattern you’ve noticed is confirmed, not every time something catches your eye.
How do I monitor my team without micromanaging?
The difference between monitoring and micromanaging sits in three variables: frequency, intent, and what you do with the information. Monitoring means maintaining a reference point for each person and checking observed behavior against that reference at a reasonable cadence. Micromanaging means checking so often, or acting on so little data, that the person feels controlled rather than supported. A useful test: does your observation lead to better-timed feedback, or does it lead to more instructions? The first is calibration. The second is control.
What is calibration drift in team management?
Calibration drift is the gradual, unnoticed shift in a team member’s behavioral baseline. Like instrument drift in engineering, no single reading triggers an alarm. The person who used to speak up in meetings goes quiet over three weeks. The one who shipped early starts shipping exactly on deadline. Each individual data point looks fine. The trend tells a different story. Managers miss calibration drift because they compare today against yesterday, not today against the person’s actual baseline from months ago.
How often should I check in with my team members?
There’s no single right frequency. A new hire in their first 90 days needs more structured touchpoints than a senior person running a familiar project. Someone working through a difficult stretch needs more ambient observation than someone in a steady rhythm. The principle is: match your monitoring frequency to the person’s current situation, not their tenure or title. When things are stable, lighter touch. When something shifts, closer attention, and tell them why so it reads as support, not suspicion.
What’s the difference between monitoring and surveillance at work?
The same monitoring behavior can read as support or surveillance depending on three things: whether the person knows why you’re paying attention, whether you act on what you observe or just collect it, and whether the relationship has enough trust to absorb close observation. A manager who says “I noticed you’ve been quieter in standup this week, wanted to check in” is monitoring. A manager who tracks the same thing silently and brings it up in a performance review three months later is surveilling. The behavior is identical. The context changes everything.