Algorithmic Management

Signs it might be this

• Decisions about your hours, pay or standing are made by a system and nobody can explain them.
• You cannot see the data collected about you or the criteria applied to it.
• There is no working path to a human reviewer, or the path ends in an automated reply.
• Rules or rates change without notice, and your results shift as a consequence.
• Discipline or termination follows an automated flag without a conversation first.

Signs it is probably something else

• You can reach a person who looks at the data and can change the outcome.
• The rules, metrics and rates are shared with you in advance and in plain language.
• The system makes suggestions and a person decides, and uses judgment about context.
• Errors you reported were checked and corrected, and the record updated.
• The tracking covers only things that affect safety or pay accuracy.
• There is a clear, working route to a human who reviews decisions and can reverse them.
• Workers can see their own data and the rules that apply to it.
• Representatives of workers have a say in how the system is set up and changed.

Questions to ask yourself

• When a decision about you looks wrong, is there a person who can look at it and change it?
• Can you see what data is collected about you and what rules are applied to it?
• Have the rates, rules or targets changed without notice, and did your results change with them?
• Does a manager decide, or do they only pass along what the system produced?
• Do other workers describe the same unexplained outcomes?

What it is

Algorithmic management is the use of software to direct and judge work. It hands out tasks, sets schedules, measures speed and flags people for discipline.

Software on its own is common and often useful. The concern is when the criteria are hidden and the data cannot be seen or corrected.

Another concern is that no person can be reached to review a decision. Then workers face a boss they cannot question.

It shows up in delivery, warehouse, driving, call center and retail work, among others.

Why it matters

Big decisions arrive with no one to ask and nowhere to appeal. Workers adapt to a system they cannot see, and they blame themselves for outcomes they have no way to check.

Pay, hours and job security can turn on numbers nobody will explain.

How it shows up

• Software that assigns work, sets pace, tracks location or output and scores workers continuously.
• Decisions about scheduling, pay or discipline that come from the system, sometimes with a manager simply passing them along.
• Rules that are hidden or change without notice, so workers cannot tell what they are being judged on.
• Data collected about a worker that the worker often cannot see, check or correct.
• Found in gig platforms, warehouses, call centers, retail scheduling and delivery, among other places.

What you can do

Pick what fits. You do not owe anyone a confrontation.

A check to run before you share or act

“Which data and rules led to this decision, and who can review it? Could you send me that, along with a way to submit corrections, in writing?”

Ask about it gently

Ask a supervisor or the platform’s support team, in writing, what data and rules were used for a specific decision. Also ask who can review it.

Set a clear limit

Save screenshots, messages and records of hours, pay and warnings, with dates, outside the company’s own systems.

Step back

If the constant tracking is wearing you down, look at where you can safely ease the pace. Also look at your options for other work, and avoid sudden changes.

Get support

Compare notes with coworkers or a worker group. Seeing that others had the same unexplained outcomes can show whether it is a system issue and not something you did.

At work, school or with services

A union, a worker center or an employment attorney can explain what rights to information and review may apply where you work. They can also explain how to raise a dispute.

Using this label fairly

• Calling any software tool at work algorithmic management can overlook systems that help workers and that a person reviews. The concern is hidden criteria and no route to a human.
• Blaming an algorithm can also let managers off the hook when they chose the rules; the label should not hide who set them up.

How it differs from look-alikes

Performance Framework Manipulation is usually human judgment dressed up as measurement. Here a system makes or relays the decision and nobody can explain it.

Institutional Betrayal is about a failed response to harm. Silence Engineering teaches people to stay quiet by example.

This one controls pace, pay and hours through software that sees everything and answers no questions.