Productivity Is Not What We Think It Is
Productivity is often seen as a question of doing more: more tasks, faster execution, fuller calendars. But activity alone does not tell us whether useful work is happening. This article takes a closer look at what productivity means in practice, examining how context, attention, tools, interruptions, and system design shape the quality of work.
FIG 1.0 - An arrangement of multiple different working outfits.
You can possess many faces and many outfits, but you can only ever operate in one of them at a time.
Related Insights
Should we Rethink How to Use AI?
MXI.Studio Inauguration and History
It Is Time to Return Engineering to Engineers
MXI.STUDIO EDITORIAL
Productivity | Work | Metrics | Operations
The MXI.Studio Editorial is a collection of articles, ideas, and reflections written by the people behind MXI.Studio. We explore emerging technologies, innovation, industry trends, and the current challenges of an ever-evolving world.
Productivity Is Not What We Think It Is
T hroughout the years, productivity has been presented through an endless succession of techniques: work faster, complete more tasks, structure the calendar, organize the inbox, automate repetitive work, or find the right application and tool for the job.
The tools change. The underlying promise does not: somehow, there must be a better way to get more done. The problem is that productivity itself is rarely examined with the same attention given to optimizing it. We have spent years refining the machinery around work without necessarily agreeing on what the machinery is supposed to produce. That is where the subject should start.
Productivity depends on the work
"There is no universal productivity method because there is no universal form of work."
Writing software, answering support requests, designing a product, studying a technical problem and managing a project all impose different constraints. The same method can improve one activity and make another worse. A tightly scheduled day may help routine execution and be counterproductive when the work requires extended exploration. Productivity therefore has to be evaluated in context.
The useful question is not whether a method is inherently productive. It is whether the method produces a better result for the particular task, given the capabilities of the person performing it and the conditions around that work. This is why seemingly minor sources of friction matter and why these minor sources can accumulate to become a major issue. Consider what happens when an important task is interrupted. The cost is not necessarily captured by the time spent on the interruption itself. Work may have to be reconstructed: what was being considered, which decisions had already been made, what information was relevant, and what the next step was supposed to be.
Research repeatedly demonstrates that knowledge work is rarely continuous, occurring instead in brief, broken intervals. In a field study of 24 information workers, 57% of observed “working spheres” were interrupted, and interrupted work was typically resumed after more than two intervening activities. The effects are not quite as simple as “interruptions make everything slower.” Mark, Gudith and Klocke found in one controlled study that people sometimes compensated for interruptions by completing interrupted tasks faster, without a measured loss in quality. The compensation came with higher stress, frustration, time pressure and effort.
That distinction is important. A workflow can appear efficient on a stopwatch while being substantially more expensive in cognitive effort. The same principle applies to the systems built to organize the work. A task manager, note system or elaborate organizational hierarchy can reduce friction when it removes ambiguity. It can also create another layer of work when maintaining the system becomes a recurring obligation.
"There is a point at which organizing the work becomes work."
The objective is therefore not to create the most sophisticated system. It is to create enough structure to support the work and no more. This is not to say sophisticated systems are obsolete. They are simply built for scale. Without a high volume of tasks or a complex team structure, their overhead outweighs their benefit.
More activity does not necessarily mean more productivity
This leads to a more fundamental problem: what exactly should be counted? Productivity is often approximated through visible activity. Number of tasks completed. Hours worked. Messages answered. Documents produced. Tickets closed. These are useful operational measures in the right context, but none of them is productivity by itself.
The Bureau of Labor Statistics defines productivity fundamentally as the relationship between output and the inputs used to produce it. Labor productivity, for example, compares output with labor input. That formulation is simple, but its implications are easy to lose in day-to-day work. More output is not automatically better output, and more activity is certainly not the same thing as more useful output.
A person can complete twenty low-value tasks while making no meaningful progress on the one problem that actually matters. The same problem appears when several projects are pursued simultaneously. Context switching is not free. Controlled experiments on task switching have demonstrated measurable switching costs, with the cost increasing as task rules become more complex. Workplace research underlines how rarely knowledge workers get to experience sustained, uninterrupted focus. In a two-week study of 40 information workers, the median duration of online screen focus was 40 seconds; shorter focus duration was associated with lower self-assessed productivity at the end of the day. This isn’t to say that the average attention span is 40 seconds, it indicates how segmented online work can become under normal workplace conditions.
The important point is not that every task must be worked on continuously. Some work genuinely requires coordination, parallelism and rapid switching.
"The actual point is that switching should have a reason."
When several projects receive small amounts of attention without reaching a meaningful milestone, the work acquires another cost: maintaining awareness of where everything stands. More switching, more state to remember, more progress to reconstruct, and more opportunities to lose the thread. At some point the better productivity question is no longer:
How can this workflow be optimized?
It becomes:
What is more important to do?
Productivity is not speed
Metrics Needs to matter more.
No tricks
Just like a lot of things in life, work is often not epic, and there is no easy, low-effort way to get everything done.
Quality Metrics
Meaningless metrics are often very present in the discourse. In software, metrics like code coverage and total lines of code are not a direct translation of quality.
FIG 2 — Collaboration is key when building large-scale projects.
Speed itself is attractive because it is easy to observe.
-
a configuration file for metadata and runtime behavior
-
a CSS file for encapsulated styling
-
an HTML template for structure
-
a JavaScript module for logic and lifecycle handling
It feels productive to answer an email in thirty seconds, complete a ticket quickly or close a task before lunch. But speed only describes how quickly something happened. It says almost nothing about whether the thing was worth doing, whether it was done correctly, or whether it created a useful result. As management theorist Peter Drucker famously noted, efficiency is simply doing things right, whereas effectiveness is doing the right things, making speed alone a poor measure for value.
The same applies to utilization. A fully occupied calendar can indicate a person with too much work, poor prioritization or a badly designed process. Empty time can indicate wasted capacity, or it can represent the space required to think, learn, plan or solve a difficult problem. Indeed, operations research shows that knowledge work requires deliberate 'slack', unscheduled cognitive space, to adapt and solve complex problems.
The number of tools used tells us even less. There is no intrinsic productivity gain in having a task manager, note-taking application, calendar system, automation platform and AI assistant all participating in the same workflow. Tools are inputs to a system. They become productive only when they improve the relationship between effort and result. Industry research has also reported substantial costs associated with fragmented software environments. A Qatalog/Cornell Ellis Idea Lab report, for example, examined time spent moving between digital tools and searching for information across them.
Research on personal productivity reflects this ambiguity. A 2019 CHI study of knowledge workers found that people conceptualize and evaluate their own productivity in diverse ways, while productivity-tracking tools often rely on much narrower signals such as time spent interacting with work applications. This gap is worth taking seriously.
What can be measured easily is not necessarily what matters most.
Productivity is a system
A more useful model is to treat productivity as a system composed of the individual, the task, the environment, the available tools and information, the surrounding constraints, and the method used to connect them. The individual is at the center, but the individual is not the entire system.
A capable person can be inefficient because the task is poorly defined. The same person can become efficient after the information is clarified, the unnecessary coordination is removed, or the toolchain is simplified. This is also why productivity discussions about time management, AI, automation, communication or knowledge management should not be treated as separate universes. They are different parts of the same system.
The important variables are familiar: effectiveness, efficiency, focus, clarity, quality, consistency, speed, effort, friction, autonomy and learning. Effectiveness asks whether the intended result is actually being achieved. Efficiency asks how much resource is consumed to achieve it. ISO's terminology makes a similar distinction: effectiveness concerns achieving planned results, while efficiency concerns the relationship between the result achieved and the resources used.
ISO terminology is found in the ISO 9001
(Quality Management) as follows:
Effectiveness (Clause 3.7.11): Defined as the "extent to which
planned activities are realized and planned results achieved."
Efficiency (Clause 3.7.10): Defined as the "relationship between
the result achieved and the resources used."
Neither is sufficient by itself. Doing the wrong thing efficiently is still a failure of productivity.
Reduce cognitive transitions
One of the simplest ways to improve a work system is therefore to remove unnecessary transitions. Every transition creates some amount of overhead: a new context, a new set of rules, a new information state, a new decision about what should happen next. This is particularly relevant to complicated work.
A difficult task can easily feel impossible before the problem has been properly understood. Sometimes the apparent lack of competence is actually a lack of sustained exposure to the problem. Once enough uninterrupted attention is given to understanding it, the problem becomes structured, and the work begins to flow. That is why focus matters.
Not just because focus is morally superior to distraction, but because difficult work often requires accumulating enough context for the problem to become tractable. The objective is not to eliminate every interruption or work indefinitely on one thing. It is to protect the periods in which continuity is genuinely valuable.
The highest-leverage improvement is often subtraction
There are two broad ways to change a system. Something can be added to it, or something can be removed. Productivity culture tends to favor the first. A new application. A new framework. A new automation. A new dashboard. A new AI agent. Humans usually overlook subtractive changes (e.g., removing a meeting, cutting a step, ditching an app). Instead, people tend to default toward adding elements. Across eight experiments, researchers found that participants were less likely to identify advantageous subtractive changes when subtraction was not explicitly brought to mind. (Adams, Converse, Hales & Klotz “People systematically overlook subtractive changes,” Nature, 2021.)
Precision That being said, later research also show that the subtraction effect can vary with task, age, culture, background, etc. making it less of an absolute law.
FIG 3 - Prioritizing one outfits over the other.
When you're overwhelmed by assets, the most overlooked solution isn't to add more, it's to subtract down to the things you actually own and master.
But subtraction is perhaps the
higher-leverage move.
Before optimizing a process, ask what should exist at all.
Does this task need to exist?
Does it need to happen now?
Does this step need to be performed manually?
Does this complexity solve a real problem, or is it solving a
hypothetical one?
Large systems especially invite premature complexity. A substantial
application can accumulate subtle improvements, layers of tooling,
abstractions and edge cases long before the basic workflow has
demonstrated what users actually need.
A simpler starting point is often more productive.
It is easier to understand, easier to change and easier to evaluate. Through use, constraints become visible. Experience reveals what deserves to become specialized. The system can then evolve into something tailored to its actual priorities rather than its imagined requirements. Productivity is therefore less about finding the perfect method in advance than about building a work system that can continuously remove what does not contribute and reinforce what does. That is the foundation for everything that follows. Productivity can be examined through different lenses: time, attention, tools, automation, information, AI, coordination and learning. The subject is not how to do everything faster. It is how to make useful work happen with less unnecessary effort.