A resignation letter is the last signal, not the first. The first signal arrived weeks or months earlier, and you probably did not notice it because it did not look like a problem. It looked like someone having a quiet week.
Researchers call this "quiet cracking," a constant feeling of workplace unhappiness that guides workers toward poor performance and eventual departure (InFeedo, 2026). With employee engagement at an 11-year low of 31% in the U.S. and 21% globally (Gallup, 2025), this is not a rare occurrence. It is the default state of most teams.
Here are the seven behavioral signals that precede the resignation you did not see coming.
1. Questions Replaced by Execution
Engaged employees ask questions. Why are we building it this way? Have we considered this alternative? What happens if this assumption is wrong? When someone shifts from asking questions to just executing tasks as assigned, they have stopped investing in outcomes and started investing only in completion.
This is one of the earliest and most reliable signals. Qualtrics research identifies "lack of enthusiasm for new projects" as a primary behavioral indicator of disengagement (Qualtrics). The shift is subtle because the employee is still producing. Output looks normal. Ownership does not.
2. Camera-Off Creep
One person turning off their camera in a meeting is a personal preference. Three people on the same team turning off their cameras over the same month is a pattern. The APA's 2024 Work in America survey found that 1 in 5 employees feel overlooked during virtual meetings (APA, 2024). Camera-off behavior can be both a cause and a symptom: feeling invisible leads to turning the camera off, which reinforces invisibility.
The signal is the trend, not any single instance. Track whether optional visibility (cameras, participation in collaborative docs, voluntary contributions) is declining across the team.
3. Standup Monosyllables
A standup contribution that used to be "I am working on the auth migration, hit a snag with the token refresh flow, going to pair with Jordan on it this afternoon" becomes "Still on auth migration. No blockers." The information density collapses. Slack research highlights that "passionate employees turning monosyllabic in meetings, responding with brief yes or no answers" is a signal of diminished engagement (Slack).
The monosyllable pattern matters because it indicates withdrawal from shared context. The employee is not just doing less. They are sharing less. That means the team has less visibility into their work, which increases coordination cost and reduces collective intelligence.
4. Declining Optional Meetings
Every team has mandatory meetings and optional ones. Retros, brown bags, team socials, architecture discussions. Engaged employees attend optional meetings because they want to influence outcomes and stay connected. Disengaging employees start declining because the return on investment feels too low.
FranklinCovey research identifies "social isolation" as a critical sign, noting that team members who were once social begin limiting contact with peers, avoiding team tasks, and skipping events they once enjoyed (FranklinCovey). Optional meeting attendance is one of the most quantifiable versions of this pattern.
5. Shorter, Less Detailed Written Communication
This one is measurable without any special tools. Pull up a disengaging employee's Slack messages or PR comments from three months ago and compare them to today. You will often see: shorter messages, fewer explanations, less context, and fewer questions or suggestions in code reviews.
Research on communication changes in disengaging employees found that "shorter, less detailed emails and minimal participation in collaborative documents" often precede more visible signs of workplace detachment (Teramind). Communication quality is a leading indicator. Output quality is a lagging one.
6. Zero Pushback
A team that never disagrees is a team that has stopped caring about outcomes. Healthy teams have friction. They debate technical decisions, challenge timelines, and raise concerns about scope. When that friction disappears, it does not mean the team is aligned. It means the team has given up on influencing direction.
This maps directly to psychological safety research. Edmondson's work shows that safety decays when the "performing" process stops. When people stop pushing back, they have exited the learning zone. They are executing, not collaborating.
7. Output Without Ownership
The most advanced stage of disengagement before departure is output without ownership. The employee delivers what is asked, on time, to spec. But they never volunteer for anything beyond the spec. They do not propose improvements. They do not flag risks proactively. They do not mentor junior teammates. They are a function, not a teammate.
This stage is the hardest to catch because it looks like reliable performance. Many managers mistake it for maturity. It is not. It is the final stage before the employee starts passive interviewing. Research shows disengaged employees are 18% less productive and drive 43% higher turnover on their teams (WellSteps, 2025).
Why Managers Miss These Signals
Three reasons:
First, they are looking at output, not behavior. Most management systems track what people produce. Almost none track how people collaborate, communicate, and engage with each other. Slack activity is not a reliable signal. Neither is meeting attendance. Behavioral quality, the richness of participation, is what matters.
Second, they are overwhelmed. Gartner research shows 75% of managers are overwhelmed by expanded responsibilities, with the average manager shouldering 51% more than they can effectively handle (Gartner). When you are stretched thin, subtle behavioral shifts in individual team members become invisible.
Third, they lack a baseline. If you do not know what "engaged" looks like for each person on your team, you cannot detect the deviation. This requires attention to individual patterns over time, not just aggregate team metrics.
What to Do Instead
You need a system that generates behavioral data continuously, not annually. QuestWorks, the flight simulator for team dynamics, runs teams through scenario-based challenges on its own cinematic, voice-controlled platform. Each quest generates behavioral data: who communicated, who stepped up, who pulled back, where patterns shifted. QuestDash surfaces these trends in real time. Leaders see aggregate team dynamics and strengths-based highlights. HeroGPT provides private AI coaching that never shares upstream.
The point is not surveillance. The point is visibility into team dynamics that would otherwise be invisible until it is too late. Annual surveys tell you what happened. Behavioral data tells you what is happening.
Every sign on this list is something QuestWorks surfaces through actual teamwork, not monitoring. When someone who used to be vocal in quests starts going silent, that shows up in the data weeks before it shows up in a survey or a resignation letter.
Try it free
