A manager installs software that shows when each team member is active, how many tickets they close, how long they were away from the keyboard. The reason given is reasonable: the team now works partly from home, reports are due to clients, and nobody wants to be surprised. Within a month the team has quietly learned to keep the screen moving. The numbers look better and the work, in small ways, is worse.
This piece looks at that pattern: what happens when visibility into work is delivered as monitoring of people, why it tends to wear away the very trust it was meant to replace, and what a business can do to get reliable reporting without making every employee feel watched. It is a hypothesis drawn from general workplace research, not a finding about any particular South African employer.
Why monitoring is tempting
Reporting is a real obligation. Directors need to know whether the month is on track. Clients want service-level reports. A team lead who cannot see what is happening is carrying risk without information, and that is uncomfortable. When work moved out of a shared office, the informal signals that used to fill the gap (who is at their desk, who looks overwhelmed, who is quietly finishing early) disappeared, and activity data looked like a substitute.
The trouble is that activity and outcome are different things. Keystrokes, logged-in hours and response times are easy to count. Whether a customer's query was actually resolved, or a reconciliation was actually correct, is harder. Systems tend to report what they can measure cheaply, and managers tend to manage what is reported.
What it does to the people being measured
The World Health Organization's guidance on mental health at work lists low job control and authoritarian supervision among the working conditions that pose a risk to mental health. It also names unclear job roles and limited support from colleagues. Monitoring that is imposed rather than agreed removes a little of the control people have over how they organise their day, and tells them, without a word being said, that their judgement is not presumed.
The UK Health and Safety Executive's management standards for work-related stress come at the same point from a different direction. They treat control (how much say people have in the way they do their work), support and role clarity as core conditions for a healthy workplace. A dashboard that counts minutes of activity does not supply any of these. It supplies a number and a reason to manage the number.
Gallup's research on employee burnout identifies unfair treatment, unmanageable workload, unclear expectations, lack of manager support and unreasonable time pressure as the leading causes. Monitoring does not automatically cause any of them, but poorly explained monitoring can feel like unfair treatment, and a metric that rewards constant visible activity can encourage the unreasonable time pressure that people then describe as burnout.
An illustrative case
Consider an illustrative example, not a real case: a Johannesburg accounting practice with twenty-two staff. During a period of rotating load-shedding schedules, several accountants began working from home on the days their suburb had power. The partners, worried about deadlines for provisional tax and VAT returns, introduced software that logged active time and took periodic screenshots.
Within weeks, two things happened. A senior accountant who did her most difficult review work by reading printed schedules away from her screen found her logged hours dropping, and began opening spreadsheets she did not need so that the tool would register her. A junior employee, afraid of looking idle, stopped asking colleagues questions by phone because a call produced no activity at all.
The partners now had a tidy weekly chart and a team that was more tired, more guarded and less likely to flag a problem early. The thing they actually wanted to know, whether the returns would be filed correctly and on time, was no clearer than before.
The human tax: autonomy and dignity
The first cost is autonomy. Employees lose some of the discretion to decide how and when to do their work, even when the outcome is identical. The second is dignity: being watched as a precaution carries an implied judgement that one could not be relied on. Both costs are felt by diligent people at least as much as by the ones the monitoring was meant to catch, because diligent people are the ones who care about how they are seen.
There is also a quieter cost to the business. When people believe a number is being used to judge them, they have a reason to hide errors and to avoid raising problems, which is the opposite of what a reporting system is for. Google's site reliability engineers make a version of the same argument in their writing on postmortem culture: if people fear blame, they will conceal information, and the organisation learns less. A reporting regime built on suspicion tends to produce the same concealment.
There is a legal dimension in South Africa too. Employee monitoring generally involves processing personal information, and the Protection of Personal Information Act (POPIA) places obligations on employers as responsible parties. The Information Regulator publishes guidance on POPIA, and a business that wants to monitor staff should be able to say what it collects, why, and how long it keeps it. The earlier article on when access controls feel like distrust covers the related case of how security measures are received by the people they apply to.
What could realistically change
The alternative is not to stop knowing what is happening. It is to measure the work, and to be open about what is measured and why.
- Define the outcome first. Decide what each role is accountable for in terms a client or director would recognise: returns filed, queries resolved, invoices issued. Report those, not hours of activity.
- Monitor the process, not the person. Where a workflow runs through systems, track whether items are stuck, late or waiting for approval. The guidance on what to monitor in an automated business process shows how a queue can reveal a bottleneck without anyone being singled out.
- Agree the rules with the people they affect. If any individual-level data is collected, explain it in plain language, limit it to a stated purpose, and give staff a way to see their own data and raise concerns.
- Replace surprise with a rhythm. A short weekly check-in, where each person states what is on track and what is blocked, gives a manager most of what activity data promised, and gives the employee a voice. What real delegation requires from systems sets out why clear expectations and visible status matter more than supervision.
- Make it safe to report a problem early. Early warnings are the most valuable thing a team can give a manager, and they only arrive if the person raising them is not penalised for it.
The business outcome
Managers get reporting they can trust because it describes results rather than appearances. Problems surface sooner because raising them is not risky. The business is also better placed to answer a regulator or a client who asks how staff information is handled, and it reduces the chance that good people leave because they feel they are being policed. Where burnout is already a concern, the article on why burnout often starts as an operations problem is a useful companion.
The human dividend
The dividend is autonomy: the freedom to organise a working day around the work itself, with trust that the result will be judged on its merits. For a person who does careful, unglamorous work, that can mean reading on paper, stepping outside to think, or phoning a colleague without worrying that the time will count against them. It will not look the same for everyone, and it is not a promise of a gentler job. It is the conditions for being treated as a competent adult at work.
The trade-offs
- Less visibility at the individual level. Managers lose a convenient proxy and must do the harder work of defining outcomes and having conversations.
- Some misuse will go unnoticed. Trust-based arrangements accept a small risk that a few people will take advantage. Whether that risk is acceptable depends on the work.
- Some monitoring is justified. Security logging, safety-critical roles and regulated activities may need individual records. The question is whether they are proportionate, explained and limited.
- Outcome measures can also be gamed. Any number used to judge people can distort behaviour, so the measures need review.
Why it matters
A business that replaces trust with monitoring usually ends up with less information than it started with, because the people who hold the real picture stop sharing it. Reporting that rests on agreed outcomes, visible process and psychological safety gives managers what they were after and leaves people their dignity.
Sources and further reading
- Mental health at work (last verified 10 October 2026)
- Management Standards - HSE (last verified 10 October 2026)
- Employee Burnout, Part 1: The 5 Main Causes (last verified 10 October 2026)
- Google SRE - Blameless Postmortem for System Resilience (last verified 10 October 2026)
Related Mellorca reading
- The Difference Between Being Informed and Being in Control
- What Changes When People Can See Their Own Numbers
- When the System Decides What You Can Do Today