Psychology · Leadership
Locus of control: why contact center agents disengage, and what it costs you
Every operations leader has watched it happen. A capable, motivated agent joins the team. Six months later they are doing the minimum, deflecting feedback, counting the hours. Leadership calls it a motivation problem and responds with incentives, gamification, or a pep talk. The pattern continues, because the diagnosis is wrong.
Agent disengagement is usually a design problem. The concept that explains it best comes from psychology, and it is called locus of control.
The concept in plain terms
Locus of control describes where a person believes the causes of outcomes sit. With an internal locus, I believe my actions influence what happens to me. With an external locus, I believe outcomes are decided by forces outside me: the system, the boss, luck, the script.
This belief is shaped by experience. If my environment keeps showing me that my effort changes nothing, I update my belief accordingly. That update is rational. It is also the beginning of disengagement.
What a contact center teaches its agents
Now look at a traditional contact center through this lens. I can, because I sat in that chair before I ever ran one.
An agent identifies a product bug generating hundreds of contacts. They flag it. Nothing happens. They flag it again. Nothing. Meanwhile their AHT is up because the bug makes every call longer, and their scorecard turns red for something they diagnosed and reported. What is this environment teaching them about the connection between their actions and their outcomes?
Disengagement is not a character flaw. It is a learned, accurate response to an environment where effort and outcome are disconnected.
Repeat that lesson daily across an entire floor, and you get a team with a collectively external locus of control. They stop reporting bugs. They stop suggesting improvements. They follow the script even when they know it will fail, because deviating carries risk and conforming carries none. The customer feels all of it.
What it costs
The costs hide in metrics nobody connects to their cause. Repeat contacts rise, because closing tickets became more rational than resolving problems. Attrition rises, and every departure carries recruitment and training costs plus months of lost productivity. Product intelligence dries up, because the people who talk to customers every day have learned that reporting upward is wasted effort. Quality drops precisely where AI cannot compensate: judgment, empathy, initiative.
When leadership responds with tighter monitoring, the external locus gets reinforced. More control produces less ownership. The spiral is fully self-sustaining at that point, and no engagement survey will fix it.
Designing for internal locus
The good news is that locus of control responds to environment, in both directions. Managed across teams of over 1,850 agents in 4 countries, I have seen the shift happen when the design changes. Four moves matter most:
- Close the loop, visibly. Every escalation gets an answer, even when the answer is no. Effort that disappears into silence teaches helplessness. Effort that gets a response, any response, teaches agency.
- Give real decision space. Define the boundaries, then let agents choose within them. Trusting people to go beyond the script when a customer needs it is a locus intervention, not a compliance risk.
- Measure outcomes, not obedience. Metrics like First Contact Resolution reward judgment. Metrics like AHT alone reward compliance. Your scorecard tells your team which belief to hold.
- Put agents in rooms where decisions happen. In the Customer Happiness Center model, Advocates join sprint planning and present evidence to product teams. Nothing rebuilds internal locus faster than watching your input change something real.
Why this matters more in the AI era
As AI absorbs repetitive volume, the interactions left for humans are the ones requiring initiative and judgment. Those are exactly the capabilities an external locus of control shuts down. A disengaged team plus a deflection layer produces a support operation that is efficient at the easy work and broken at the hard work. I wrote about that combination in AI deflection fails when it is implemented under pressure.
The leaders who get ahead of this will treat psychology as an operational discipline, not a soft topic for the offsite. That is the bet behind everything I build.