The Frequency

I built a simple Slack workflow that asks one question every Friday at 3 PM. “On a scale of 1-5, how sustainable was this week’s workload?” Everyone on the team answers anonymously, and the results go to department leads in real time.

Here’s the practice that matters. If any department averages below a 3 for two consecutive weeks, we pause all new projects for that team and do an emergency workload audit within 72 hours. We look at what can be automated, what can be delayed, and what needs to be reassigned to contractors or other teams.

We’ve run this for 18 months across 100+ SSD law firms using our platform, and we’ve seen it work internally too. The key is making the rebalancing automatic and non-negotiable. People won’t ask for help when they’re burned out. You have to build the ask into your system and act on the signal before they start updating their resumes.

Burnout doesn’t usually announce itself. It creeps in. In my experience, you see it first in small things: people who used to volunteer stop speaking up, response times slow down, quality dips in subtle ways.

The biggest mistake leaders make is waiting for someone to formally complain. By then, they’re often already halfway out the door.

What we watch closely isn’t just hours worked, but load distribution. During peak periods, there’s always a tendency to lean on the most dependable people. They don’t push back, so they get more. That works for a few weeks. It hasn’t worked for a few months.

One rule I’ve used consistently is simple: if someone is carrying the highest-pressure work two cycles in a row, we rotate it. Even if they can handle it. Especially if they can handle it. High performers are the ones who burn out quietly because everyone trusts them to “figure it out.”

We also force clarity around priorities during peak times. We’ll sit down and ask: what truly has to be done now, and what are we pretending is urgent? When the team sees that leadership is willing to deprioritize something, it lowers the psychological pressure immediately.

Burnout prevention isn’t about adding wellness programs during busy seasons. It’s about disciplined workload management and paying attention to patterns before they turn into exits. If you intervene early, most people don’t leave, they recalibrate.

Burnout rarely shows up as someone saying, “I’m overwhelmed.” It shows up as subtle shifts such as slower response times, less input in meetings, emotionally reactive communication, or a drop in initiative. One of the earliest signals I look for is when high-performing team members stop asking thoughtful questions. When employees who were previously engaged start doing the bare minimum, that’s often a sign they’re operating in burnout mode.

To address burnout risk before it escalates, we use a simple rule during peak periods: nothing new gets added without something else being deprioritized. When workloads spike, we focus on trade-offs. If a new priority comes up, I’ll ask, “What is getting deprioritized as a result?” That question alone prevents unsustainable workloads.

In my experience there really isn’t a magic bullet sort of sign for spotting burnout risk, so what has worked for me is to look for the subtle changes before burnout becomes obvious. This is made possible by my relatively small and tight team, but things like shorter replies, slower decisions, or less initiative are what seems to be the most obvious to me.

I like to follow one simple rule with my teams, namely that no one stays above planned capacity for more than two cycles. If someone is stretched beyond that, it is an immediate flag to rebalance or risk burnout.

Some crunch is unavoidable, but systemic crunch is a failure in leadership so I’d much rather shift work early than wait for frustration to build. Overall, I find that burnout doesn’t start with some extraordinary event, it starts with quiet withdrawal and subtle behavioral change.

I’ve spent enough time in startups to know that burnout rarely looks like someone working too hard. It looks like someone carrying too many decisions. During peak periods, the people most at risk aren’t the ones logging long hours. They’re the ones silently trying to figure out what to prioritize when everything feels urgent. That paralysis is the early warning sign most leaders miss.

We operate with a simple rule here: when workloads spike, leadership decides what gets deprioritized. Not the individual. The moment you leave that call to the person already underwater, you’ve set them up to fail. They’ll either try to do everything and burn out, or make a tough call and wonder if it was the wrong one. Neither is fair to them.

The other thing I watch for is when someone who usually contributes ideas goes quiet. That silence tells me more than any productivity metric ever could. It means they’ve shifted from engaged to surviving. By the time exhaustion is visible, you’ve already lost months of their best thinking.

Burnout is a leadership failure that shows up as an individual problem. If your people are consistently overwhelmed during peak periods, the issue isn’t their capacity. It’s that no one above them is making hard calls about scope. Protecting your team’s energy isn’t soft management. It’s how you keep your best people from walking out the door.

I identify potential burnout by observing behavioral changes rather than waiting for complaints, like decreased participation in meetings, longer response times from previously high-performing team members, or a general lack of innovative ideas. We also analyze workload distribution data on a weekly basis to identify individuals who have an unusually high number of pressing tasks. We know burnout often lives inside high performance.

We have one simple rule for times when we need to be ultra-productive: no one can have more than one high-priority initiative in their queue without us intentionally redistributing lower priority work to someone else. This ensures we don’t just pile priority on top of priority, and it helps us keep performance and retention high.

Engineering team burnout typically presents as “communication silence” and precedes performance degradation in code. We’ve noticed major telltale indicators, including dramatic drops in peer review engagement or “velocity drift,” where standard workloads take significantly longer than historical averages. Upon detecting these signs we begin by auditing their context-switching load to determine if the difficulty arose from the number of hours alone or due to the ramifications of mentally switching between numerous unrelated issues.

During periods of high demand for delivery we implement a “One-In, One-Out” policy. When a stakeholder provides an urgent “must-have,” an existing similarly-sized item must be pushed back to the backlog. Essentially, this takes the onus for judging priority off the developer and puts it on to the executive leadership team. As such, peak demand cycles do not become permanent over-capacity conditions which contribute to chronic attrition.

To scale sustainably “treating developer energy” is critical; it is a limited resource as opposed to an unlimited one. Sustaining developers’ focus during high velocity sprints is critical for maintaining quality of work produced by your best developers while also preventing them from seeking employment elsewhere. Remember every line of code was written by an individual attempting to balance providing professional quality output against maintaining personal wellness. Building a culture where individuals are comfortable voicing their capacity constraint prior to reaching a trigger point is key.

From my experience, burnout shows up in energy changes before it shows up in results. When someone who used to be engaged becomes quieter or less proactive, I pay attention right away. Over time, I also study how their workload has been stacking up because patterns never lie. If they have been operating at max capacity for too long, I see that as a signal for leadership to step in. So I deal with it directly through open conversations and by reshuffling priorities before frustration builds.

During busy seasons, we keep things very straightforward. We prioritize based on revenue impact and long term strategic importance, and the rest goes on hold. If a task does not clearly support growth, we postpone it without overthinking it. You know, that clear filter reduces stress quickly and allows the team to concentrate on the work that truly moves the needle.

Before a person quits, they are likely to show signs of distress due to burnout. Some examples of those distress signs include decreased physical energy, increased reaction times, or an increase in the number of errors made than usual. Because of this, completing regular check-ins is critical to the well-being of your team, and we place a high emphasis on talking about workload at the very beginning stage of an employee career (before the buildup of stress can make them disengaged from their work).

During busy times, one of the rules we follow is that, if a new task arrives that needs to be considered immediately (a new urgent task), then something else must slide to accommodate this. You cannot have everyone working on every priority at once. Following this rule allows for a focused workplace, keeps difficulties with overall work volume within realistic limits, and demonstrates to employees that we value them and their contributions to the success of our company.

To spot burnout risk early, we monitor workload distribution and project intensity across teams. One simple rule we implemented is a “two-major-project limit” per team member during peak cycles. If someone is assigned two high-priority projects, they cannot take on a third without redistributing tasks. This rule helped maintain productivity while protecting energy levels. Preventing overload is far easier than repairing burnout after it appears.

Burnout often shows up when employees spend too much time on repetitive administrative work instead of meaningful projects. Our team watches for situations where HR staff or managers begin spending large portions of their week manually updating systems or reentering the same data across platforms.

Data integration helps us prevent that problem by automating the flow of information between HR, payroll, and other systems. Removing those repetitive tasks frees up time in employees’ schedules so workloads stay balanced even during busy periods.

A simple rule we follow during peak periods: if a task requires people to move the same data between systems, we automate it. That approach keeps teams focused on work that actually drives the business forward and helps prevent burnout before it starts.

The red flag nobody warns you about is calendar practices. When an individual begins marking their calendar solid with opaque “focus time” or “planning sessions” that last from 3-4 hours they are avoiding new work. They have been subsumed by the tsunami and are desperately trying to hold onto any semblance of control over their calendar. If someone goes from 60% availability to 90% booked within 2 weeks please raise the alarm. There is a fracturing that occurs when someone feels like they can’t say no anymore, so they fabricate busy periods.

The productivity hack that actually helps is mandated task delegation during crunch time. If someone hits 55 hours/week for 3 weeks straight they are required to hand off their lowest priority project to a teammate working 40ish hours. No questions asked. This rules out consistently having the same people shouldering everything when crunch hits.

Burnout does not develop overnight; typically, it develops gradually through small changes in behaviour, for example, someone having very short responses or having fewer ideas to discuss in meetings, and it can also be evidenced by people working longer hours than they mention. The importance of recognising these signals helps, as often the decision to leave is already forming before an individual has acknowledged their experience of being burned out. During peak periods of delivery work, our teams have a simple approach to reduce pressure on our team members, which is to stop new initiatives internally as delivery workloads increase. If the increase in delivery workloads becomes excessive, we pause or reduce our internal project load, reduce the number of reporting layers, or eliminate any experimental initiatives that have not been identified as critical.

This approach provides teams with some breathing room, and does not set the expectation for team members to work harder. Typically, teams respond positively to this clarity. When teams see that their leadership is actively trying to rebalance expectations rather than maintain the status quo of adding pressure quietly, it enhances the level of trust and relationships between the teams and the leadership. Furthermore, sustainable performance is generated through maintaining team member’s capacity, not by continuously extending it.