Are we measuring work the wrong way?

Thinking about productivity in a workplace shaped by AI

When I looked over what was called a “final” deliverable for the first time, I knew something was wrong.

It looked finished on the outside. Slides that are well-organized, well-formatted, and submitted right before deadline. But when I looked closer, I saw something else: generic content, factual errors, and an unverified drag and drop tone that shows too much reliance on AI. It hadn’t been checked out. No one had owned it.

But it did come right before the deadline.

Such repeated experiences made me reflect on a growing problem in today’s workplaces, that we are still using outdated systems to measure work that don’t show how work is actually done.

Old Metrics Still used in a New World

Most businesses still use the same old metrics: turnaround time, hours worked, and number of outputs delivered. In a world where effort and time were directly linked to output, these metrics made sense.

We live in a new world now, with new AI tools everywhere.

AI tools have changed the game in a big way. Things that used to take hours can now be done in minutes. You can make content right away. The first drafts are no longer the problem.

But while things are moving faster, measurement systems have stayed the same.

This mismatch became clear to me when I was in charge of a creative vendor team for an information security awareness project. Even though AI cut down on the amount of time people actually worked, they still followed the usual deadlines and work was still received right around the deadline. What happened? On further scrutiny I discovered vendors started working on more than one project at a time, using AI to help them keep up.

On paper, output went up. But the reality was in the grey

The Illusion of Being Productive.

When time based systems don’t change in an AI-enabled environment, work expands to fill the timeline not because it has to, but because the system expects it to.

At the same time, teams focus on volume instead of validation. People often use faster ways to finish tasks to get more done, not to make the work they already have better.

This creates a dangerous cycle:

• AI speeds up execution
• More work to do to get the most out of your time
• Less focus on verification
• Timely delivery hid the drop in quality

The result is what I observed – deliverables that come right before deadlines but don’t meet standards.

In these kinds of systems, meeting deadlines is more important than making sure things are right.

AI Has Changed How We Work, But Not How We Think

The main problem isn’t AI itself. We haven’t changed how we think about and measure productivity.

AI has sped up the time it takes to run. But it has also made judgment, validation, and ownership more important.

It used to take a lot of work, but now it takes a lot of thought.

This is a paradox : As our tools get better it’s easier to make work that looks finished but doesn’t have much depth.

A New Way to Think About Productivity

In my opinion, productivity today should be measured in three ways:

1. Speed (AI’s Edge)

AI has made speed a fact. Being able to do things quickly is no longer a unique skill,  it’s a basic requirement.

2. Quality (The Gap in Verification)

Human verification is what really sets things apart. Is the work correct, useful, and aware of its surroundings? Has it been reviewed by experts?

3. Responsibility (Human Accountability)

In a workflow that uses AI, you can’t give tools the responsibility for accountability.

Real productivity isn’t about how quickly something gets done; it’s about whether it can be trusted and whether it makes a difference.

What This Means for the Future of Management

This is the question for CXOs and managers who are and will be leading in the world of AI:

You can’t just look at time or activity to see how productive someone is anymore.

What managers need to do:

• Change the timelines to show how much work is actually being done, not what was thought to be true in the past.
• Change the focus from quantity to quality
• Create systems that reward checking, not just finishing.
• Make sure that AI-assisted workflows are accountable

Conclusion: From Work to Effect

The only way to measure real productivity today is by the quality and impact of work.

In a world where AI can quickly come up with answers, the real benefit goes to those who can ask questions, improve them, and take responsibility for them.

And that is something that no metric has fully captured yet.

Credit : Megh Poddar is a PGDMGM student at XLRI Jamshedpur