Burnout early-warning checklist

Burnout doesn't arrive on a single day. It shows up as a slow drift — a fast rep who's suddenly slower, a quiet person who goes quieter, a team where the banter dies. By the time it reaches "I'm resigning" or "I'm signed off sick," the early signs have usually been visible for weeks. This checklist is for catching them while there's still room to act.

A warning up front, because it decides whether this tool helps or harms: this is a checklist for a manager, not a dossier on a person. Every signal below can have a dozen innocent explanations — a rough patch at home, a heavy release week, a new baby, a bad night's sleep. A ticked box means ask a question with genuine curiosity, not conclude something and log it. If you use this to build a case against someone, you have turned a care tool into a surveillance tool, and people will feel the difference immediately. Don't.

How to use this

• Cadence: skim the individual list for each direct report once a month — pair it with your 1:1 prep, not a separate audit. Run the team-level list monthly for the group as a whole. Do an off-cycle pass on anyone who's just been through something hard (a bad escalation, a bereavement, a reorg, a big launch).
• Threshold: one box is noise. Two or three ticked at once, especially across different sections, is a pattern worth a real conversation. The count matters less than the change — you're watching for someone moving away from their own normal, not measuring them against the team.
• The move is always the same: notice → ask, privately and without accusation → listen → agree one concrete thing → follow up. The interventions below are starting points, not scripts. The rep knows what they need better than the checklist does.

Golden rule: if you'd be uncomfortable telling the person you'd noticed the thing, you're either about to have an important conversation or about to misuse this list. Work out which before you open your mouth.

Individual signals — one rep at a time

Changes you can see in the work

These are drift from their own baseline, not the team average. A rep who was never fast isn't a warning sign; a fast rep who's suddenly slow is.

• Handle time or response time creeping up on someone who was reliably quick
• QA / quality scores sliding, or more small mistakes than usual (wrong macro, missed detail, misread question)
• Follow-ups getting dropped; tickets slipping SLA that used to get closed
• More bare copy-paste of macros, less of the personalisation that used to be their signature
• Reaching for escalation or "let me check with a colleague" on things they used to handle solo
• Cherry-picking easy tickets and leaving the hard queue for someone else (or the reverse — grinding only the hard ones and skipping breaks)

Try: Lead with the observation, not the metric. "I noticed billing tickets have been sitting longer than they used to for you — what's going on there? Is something making them harder right now?" Nine times out of ten they'll tell you the real cause (a tool that broke, a policy that changed, a personal thing) and the fix is obvious once it's named. Never open with the number on a dashboard.

Changes in how they show up

• The camera goes off, or they stop speaking up in team calls they used to contribute to
• Shorter, flatter messages in team chat; the emoji and the jokes disappear
• More irritable or short with customers, teammates, or you — a normally patient person with a shorter fuse
• Withdrawing from the social edges of work: skips optional calls, eats lunch at the desk, stops joining the non-work channel
• Cynicism creeping into how they talk about customers ("they're all idiots") or the company ("nobody up there cares") — a classic, well-documented burnout marker
• Visibly running on empty: tired, unusually flat, or wired and unable to switch off

Try: Make room, don't interrogate. A private "How are you doing — actually, not the work version?" and then silence does more than five questions. If the cynicism is about the company, resist defending the company; ask what specifically is wearing them down and whether any of it is something you can carry up the chain. Feeling heard is itself an intervention.

Boundaries and hours

• Online early, online late, online at weekends — creeping hours that nobody asked for
• Working through breaks; no real lunch; back-to-back with no gaps
• Not taking PTO, or booking it and then cancelling / working through it
• Answering messages at all hours — the always-on badge that looks like dedication and is often distress
• Apologising for being "behind" when they're not, or for taking normal time off

Try: Make the boundary your rule, not their weakness. "I don't want you online past your hours — if the queue needs someone at 7pm, that's a staffing problem for me to fix, not for you to absorb." Then protect it visibly: don't reward the weekend replies, and take your own breaks where they can see you. People copy what the manager does far more than what the manager says. If they have PTO banked, help them pick a date and book it in the conversation.

What you hear — language cues

• "I'm fine" delivered flatly, repeatedly, in a way that closes the conversation
• "I just can't keep up" / "there's no point" / "nothing I do lands"
• Talking about themselves in absolutes: always behind, never good enough
• Guilt about normal things — being ill, taking a day, logging off on time
• Going quiet where they used to ask for help — suffering silently instead of flagging blockers

Try: Reflect it back gently and specifically. "You've said 'I can't keep up' a couple of times this week — I want to take that seriously. What would keeping up actually look like, and what's in the way?" Naming it out loud, without judgement, often breaks the spiral. If you hear anything that worries you about their wellbeing beyond work stress, know your escalation route (EAP, HR, occupational health) before you need it — that's care, not passing the buck.

Team-level signals — the whole group

One person struggling is an individual conversation. Several of these at once is a system problem, and the fix is almost never "tell people to be more resilient."

The numbers

• Attrition ticking up, or the exit-interview reasons rhyming ("burned out," "volume," "no path")
• Unplanned absence / sick days rising, especially clustered around the busiest days
• Overtime or after-hours activity becoming normal rather than exceptional
• A widening gap between the top and bottom performers — a sign work is quietly landing on the same few people
• PTO balances stacking up unused across the team

Try: Treat these as a load problem first, a morale problem second. Pull the actual numbers: is volume up, is headcount down, has a product change dumped a new ticket type on the queue? Fix the load (staffing, deflection, a broken workflow, a bad macro set) before you ask people to feel better about it. If the top-vs-bottom gap is widening, look at who the hard tickets route to — burnout often hides inside "our best rep."

The room

• Team chat has gone quiet — less banter, fewer questions, more transactional
• Gallows humour tipping into genuine bitterness about customers or leadership
• People stop flagging problems ("why bother, nothing changes") — learned helplessness
• Retros and 1:1s surface the same complaints repeatedly with nothing moving
• New hires ramping slower or leaving early because there's no capacity to onboard them properly

Try: Close one loop, visibly. Learned helplessness comes from raising things that vanish; the antidote is picking one long-standing complaint and fixing it in the open, then saying "you flagged this, here's what changed." One visible fix rebuilds more trust than three surveys. If the same issues keep surfacing, the problem isn't that people won't tell you — it's that telling you hasn't worked.

Structural pressure — the things that manufacture burnout

Spot these before they show up in the signals above. They are the upstream causes.

• Chronic understaffing treated as the permanent normal
• Unrealistic or conflicting targets (fast and thorough and empathetic and upselling, all maxed at once)
• No control: rigid scripts, no autonomy, surveillance-style metrics
• The emotional load of the work unacknowledged — reps absorbing customer anger all day with no decompression (see: the empathy tax)
• No recovery built into the schedule: no queue-free time, no admin blocks, no rotation off the hardest channel
• Recognition and growth absent — people can't see a next step or hear that the hard work landed

Try: These are yours to fix, not the team's to endure. Pick the one with the most leverage and put a real change against it — a rotation off chat, a protected admin hour, a target you retire, a headcount case you actually make. You won't fix all six this quarter; fixing one and naming that you're working on the rest is the honest, credible move.

When several boxes are ticked — match the response to the signal

Interventions that backfire — don't do these

• A resilience webinar as the whole response. Teaching people to withstand a broken system, instead of fixing it, reads as an insult. Do the structural work first.
• A pizza / wellness perk instead of addressing load. Snacks don't offset an understaffed queue, and everyone knows it.
• Making the check-in feel like monitoring. "I saw your handle time and your camera's been off" lands as being watched. Lead with care, not evidence.
• Telling someone to take PTO and then messaging them on it. The message cancels the gesture. Protect the time you gave.
• Confusing burnout with a performance problem. They can look identical on a dashboard and need opposite responses — one needs less load, the other needs more support and clarity. Ask before you diagnose.
• Doing nothing because everyone's busy. "I'll deal with it after this crunch" is how the crunch becomes the reason someone leaves.

If it's you — manager burnout

Managers burn out too, and tend to hide it hardest because they feel responsible for everyone else's. Turn the list on yourself:

• Dreading the queue, the 1:1s, or the Monday you used to be fine with
• Snapping at the team or going numb to problems you'd normally jump on
• Absorbing every escalation yourself because it's "faster than explaining it"
• No recovery time; your own PTO permanently deferred to "after this"
• Nobody doing this check-in for you

Try: Say it out loud to your own manager or a peer — the same vulnerability you'd want from your team. Model taking the break. You cannot pour support out of an empty cup, and a burned-out manager is upstream of a burned-out team.

Running record — keep it lightly

Optional, and only for your own memory across months — not a permanent file, and not something to hold over anyone. One row per monthly pass.

If you keep this, keep it for you, delete it when it's stale, and never let it become the thing a review is built on. The point is the person, not the log.