What job specialization is
Job specialization (التخصص الوظيفي) is the division of work into narrower parts, so that each employee takes on a limited number of tasks and repeats them, instead of handling the work from beginning to end.
Job specialization is a design decision, not a description of how things are. An organisation chooses the degree of specialization in its jobs, and every degree has a gain and a price, both of them calculable. The gain can be visible and noticed, while the price may be deferred and appear elsewhere, in places where nobody attributes it to the way the work was divided.
The first gain from job specialization: time saved on switching between tasks
Take an employee whose working day is 480 minutes and who moves between four kinds of work. Each switch costs 10 minutes before the employee has recovered the context of the new work. Suppose the day contains 12 switches:
- At 12 switches, the time lost to switching is 120 minutes, or 25% of the day.
- If the work is narrowed to a single kind and the switches fall to 2, the time lost is 20 minutes, or 4.17%.
The difference between the two cases is 100 minutes in a single day. That is the gain job specialization buys, and it is a real gain that shows from the first day. The switching cost of 10 minutes, however, is assumed in this example, and it varies with the nature of the work.
A second gain comes on top of this one and is harder to calculate: repetition shortens the time a task takes and reduces the errors made in it. We give no figure for its size, because it differs from one task to another and we have no published measurement to base a figure on.
The first cost of job specialization: every division creates a handoff
Work done from start to finish by one person involves no handoff. Split it into 3 steps across three people, and 2 handoff points appear. Split it into 6 steps, and there are 5.
At each handoff point the work waits until the next person is free. If each handoff added half a day of waiting on average, the added time would be a full day in the first case and two and a half days in the second.
It follows that job specialization reduces working time and increases waiting time at the same moment. An organisation that measures working hours alone sees a steady improvement every time it divides the work further, and does not see that the time a request spends between arriving and being completed is growing. These are two different figures moving in opposite directions, and anyone who measures one and ignores the other is deciding on the basis of half the picture.
The second cost of job specialization: a fragile chain
When each step has a single person who has mastered it, the whole chain stops if any one of them is absent. The effect grows with the number of steps, not with the number of absences.
Assume that the probability of any one employee being present on a given day is 95%, and that individual absences are independent of one another. That is a simplifying assumption, not a description of reality:
- A chain of 3 steps, each held by one person, has an 85.7% probability of being completed on the day.
- A chain of 6 steps, on the same terms, has a probability of 73.5%.
Doubling the number of steps held by a single person lowered the probability of completing the day by more than 12 percentage points, without any change in anyone’s attendance. The response to this is not to reverse job specialization but to have more than one person master each step, which is one of the purposes job rotation serves.
The 95% attendance figure is assumed in order to show the calculation, as are the 10 minutes of switching cost and the half day of waiting at each handoff.
Where job specialization stops paying
Dividing the work buys a saving in switching and pays a price in handoffs and fragility. As long as the saving is larger than the price, the division pays. When the two are equal, the division has reached its limit, and every further division is a net loss.
Where that limit falls depends on two things. One is the cost of switching in that particular work: the higher it is, the more room there is for division. The other is the cost of waiting at the handoff points: the higher it is, the less room there is. That is why work that needs a long setup for each kind of task can take a degree of division that work cannot take when employees move between tasks at almost no cost.
It is a mistake to regard division as a steady improvement, in which every narrowing of a role raises efficiency. That holds up to a point and then reverses, and the reversal does not show in the productivity indicator, because it happens in waiting time and in coverage.
The definition sets no optimal degree of job specialization. The right degree differs from one kind of work to another, as the two costs above show, and we have no published value for it.
How job specialization changes what hiring asks for
Take work in which a skilled person needs 6 weeks to reach full productivity. If it is divided into 3 equal steps, a person filling a single step needs 2 weeks. The period a new hire needs to reach full productivity is covered in employee onboarding.
That is a direct gain for hiring. The position is filled faster, the candidate needed has less experience and is easier to find, and the disruption when an employee leaves costs less. It is an argument for job specialization in an organisation that is growing quickly.
The price appears years later. Someone who has worked on a single step for three years has not learned what the other four weeks would have taught, because they never went through them. When someone is needed to oversee all three steps, nobody working in those steps knows the whole of the work. The organisation gained speed in filling positions and lost its source of successors for the positions above them, and the loss shows only when a successor is needed. Preparing replacements for critical roles before they fall vacant is the subject of succession planning.
Two forms of job specialization with different results
Work can be divided on two different bases. Divided by task, each employee handles one step of the work for every customer. Divided by customer or case, each employee handles the whole of the work for a group of customers.
Division by task brings the largest saving in switching and the highest cost in handoffs, because every case passes through every pair of hands. Division by customer keeps handoff points to a minimum and leaves some switching in place, because the employee performs several kinds of work within each case.
So the question to ask is not “should we divide the work?” but “on what basis?”. The answer depends on which of two costs is higher in that work: the cost of moving between kinds of task, or the cost of moving the case between people.
The effect of job specialization on the employee
A narrow, repetitive role reduces the variety of skills the employee uses, makes it harder for them to move later into a broader job, and weakens their connection to the outcome of the work, because they see only part of it.
Three tools have developed in this area that address the effect without abandoning job specialization: job enrichment, which adds depth to the role; job rotation, which moves employees between roles; and job crafting, which keeps the task and changes the way it is performed. Each one deals with a different side of the same effect.
What job specialization is not
- Job analysis. Analysis describes a job as it stands and gathers its tasks and requirements, whereas job specialization is a decision about how the work is divided in the first place. Analysis works on what exists; job specialization decides what will exist.
- Span of control. That measure concerns how many employees report to one manager, while job specialization concerns how many tasks one employee holds. The two can move together: narrow, similar roles allow a wider span, and broad, varied roles narrow it.
- Efficiency in itself. Job specialization is a means that carries a price, not a virtue without qualification. An organisation that divided its work until every step had its own holder, and then found its requests running late, has not gained efficiency. It has moved the cost from one column to another.
What undermines job specialization
- Dividing around people rather than around the work. The division is built on whoever is in the team today, which produces a role that only its holder understands and that cannot be filled once that person leaves.
- Setting handoff points without defining what is handed over. A handoff at which nobody knows what counts as complete produces work that is sent back repeatedly, and that can cost more than the switching time saved. How the acceptable level of output in a job is written down is covered in performance standards.
- Extending the division when the volume of work is low. A step held by one person needs enough work to keep that person occupied. Without it, the holder is left waiting, and job specialization becomes a cost with nothing in return.
- Measuring each step apart from the chain. When every specialist is measured by the output of their own step alone, each of them has an interest in passing the work on quickly, whatever its state. Errors then travel to the final steps and are found where they are hard to correct.
- Keeping the division after the tools have changed. A new tool removes a step or merges two. If the division stays as it was, it keeps handoff points that no longer have any reason to exist.
- Leaving a step with no one to cover it. The effect of this on the probability of completing the chain is set out above.
Before work is divided
Two figures are calculated before the division, not after it: how much switching time is saved each day, and how much is added to the time a request takes between arriving and being completed. If the first is larger, the division pays. If the second is larger, the gain is on paper and the loss falls on the customer.
A third question is then asked before going ahead: who does each step if the person who holds it is absent? If the question has no answer at every step, the division has created a chain that breaks at a single point. The answer is not to give up job specialization but to train a second person on each step, even one who does not work on it every day.
This is an explanation of the concept, not legal advice.
A standalone Saudi HR system
One employee file holding the contract, the documents and their expiry dates, the attendance record, leave, salary and end-of-service entitlements. End-of-service, overtime and leave-balance calculations are built into the system.
A standalone system on its own subscription. The connection to Qoyod Accounting is now available.