How to Track Employee Productivity Without Distorting It

The first productivity metric a team adopts is usually task count, hours logged, or messages sent. All three share a flaw: a person can increase them within a week without doing any more useful work, and once they realise that, some of them will.

• Task count rewards whoever splits their work into the most items. You end up with more tasks and the same output.
• Hours logged measures presence and the diligence of timesheet completion, neither of which correlates with results.
• Message and update volume measures how much someone communicates about work, which is often inversely related to doing it.
• A single composite score hides its own formula, so it cannot be argued with, cannot be learned from, and gets gamed as soon as its inputs are guessed.

The pair to look at first is completion time against turnaround. If completion is ninety minutes and turnaround is six hours, four and a half of those hours are queue — time nobody was working on it. Pressing people to work faster cannot recover time nobody was working.

These are computed from stored task timestamps in TaskIt’s performance analytics, with nothing estimated or scored, and a duration is only averaged when both ends genuinely exist. Track employee task performance covers the full set and how attribution works.

1. 01Look at time to accept. If it is hours, work is sitting unclaimed. That is a routing or notification failure, not a person failing.
2. 02Look at the gap between completion and turnaround. A large gap is queue. A small gap with a long completion time is genuinely the work itself.
3. 03Look at workload distribution. One person holding eleven tasks while two hold three explains most overdue lists without any reference to anybody’s speed.
4. 04Read the blocked reasons. The same external blocker appearing repeatedly — waiting for a customer, waiting for approval, material not available — is a process problem wearing a performance costume.

The fixes for queueing are structural: route work to a team queue rather than a named person so whoever is free takes it, notify the channel people actually read, and rebalance a queue you can now see. None of them require anyone to work harder, which is why they are the ones to try first.

• Pending and in progress. What someone is currently holding. Look at this before adding to it.
• Overdue and blocked. Blocked is the more useful: it is a deliberate statement with a reason attached, not just a date that passed.
• Completed, over a window. Meaningful within one repeated kind of work, misleading across a mixed workload.
• Completion rate. Completed against assigned, with a floor — someone who finished their only task should not outrank someone who finished eighteen of twenty.

Be careful comparing counts between people. They are only comparable when the tasks are comparable in size, which they usually are not. A count describes somebody’s week; it is not a league table position.

The usual rule is to attribute a task to whoever currently holds it. That has one very good property: because a task has at most one current assignee, per-person counts summed across a company can never exceed the company total — which is the invariant a hand-written report eventually breaks.

Its cost is that someone who did most of a job and then handed it on earns nothing, while whoever finished it earns the completion. That is a limitation to compensate for rather than argue away: read handover history alongside the figures, and never open a performance conversation with a per-person average about somebody whose work changes hands frequently.

1. 01Say what you look at, before you look at it. People who do not know acceptance time is being measured will be measured badly for a behaviour nobody asked them to change.
2. 02Show the working. "Your average turnaround is eleven days" is an accusation. "These four tasks account for most of it, and two were blocked on the supplier" is a conversation.
3. 03Ask about the outliers. Averages hide the interesting cases. The three slowest jobs of the month usually contain the actual finding, and it is usually structural.
4. 04Never let the ranking make the judgement. The person at the bottom of a completion-rate ranking is sometimes the one being handed everything difficult, and no formula will tell you that.

Screen capture, keystroke logging, idle detection and location tracking measure presence rather than outcomes. They also produce their own distortion: people learn to look busy, which is a skill of no value to anybody.

Task-derived measurement is different in kind. The events being recorded — accepted, updated, handed over, blocked, completed — are things people did to the work, and they were worth recording anyway. TaskIt deliberately has no timer, timesheet, effort log or activity monitor of any kind, which is a design decision rather than a missing feature.

1. 01Overdue and blocked. Start with what is currently wrong, and read the blocked reasons rather than counting them.
2. 02Workload. Who is holding what. Rebalance before adding anything new.
3. 03Time to accept, per team. The most actionable figure most businesses have, and the one that responds fastest to a structural change.
4. 04The three slowest jobs. Not to attribute them — to find out what they had in common.