Benchmarking

Benchmarking: a definition

Benchmarking (المقارنة المعيارية) is the comparison of the performance of an organisation, or of one of its processes, with an external or internal reference. Its aim is to identify the gap and where that gap comes from, not merely to learn where the organisation ranks.

The definition has three parts. The first is what is compared: the performance of the organisation as a whole, or the performance of one of its processes. The second is what it is compared with: a reference, which may lie outside the organisation or inside it. The third is the purpose. Benchmarking sets out to identify the gap between the organisation and the reference, together with the source of that gap. Knowing the ranking, that is, whether the organisation stands above or below the reference, is not by itself the purpose that the definition gives to benchmarking.

The types of benchmarking

Benchmarking takes three forms, which are distinguished by the reference against which the organisation compares itself:

• Internal benchmarking. It compares one branch with another, or one team with another, inside the same organisation. Of the three types, it is the one for which data are easiest to obtain, and the one whose figures are the most precisely comparable.
• Competitive benchmarking. It compares the organisation with a direct competitor. Data for it are the hardest of the three types to obtain, because a competitor may not publish them.
• Functional benchmarking. It compares one specific process with an organisation that has mastered that process, in any sector, not only in your sector.

In internal benchmarking the reference lies inside the organisation, while in competitive and functional benchmarking it lies outside. What separates the last two is the choice of reference. Competitive benchmarking chooses a direct competitor. Functional benchmarking chooses whoever has mastered the process in question, whether or not that party works in the same sector as the organisation making the comparison.

The condition on which benchmarking depends

A benchmarking comparison is valid only if both parties measure the same thing in the same way. Take two organisations that each calculate their cost per hire. One counts external costs alone, and the other counts the same external costs together with internal time. Their two figures cannot be compared, however similar the name of the measure may be.

The example turns on the difference between the name of a measure and its definition. Both organisations report a figure under the same name, cost per hire, but the two figures are built from different contents. The first contains external costs only, whereas the second contains external costs and internal time as well. Setting one figure beside the other therefore compares two different calculations, not two levels of performance. The similarity of the name does not make the figures comparable, since the name does not say what each figure contains.

For this reason, one question comes before any benchmarking comparison: how does the other party define the measure? If that definition is not known, the comparison is not carried out.

The question concerns the definition used by the other party, whose figure is the one being set against the organisation’s figure. Where the answer is not known, carrying out the comparison regardless would place two figures side by side without knowing whether both of them measure the same thing in the same way. That is the situation the condition rules out.

Why benchmarking does not end at the number

Learning that another organisation completes a process in half the time is of no use in itself. What is useful is the source of the difference. Is it an extra step in your procedure? Is it a centralised authority that slows down approval? Or is it a tool that shortens manual work?

Each of these questions points to a different source. The first places the difference in the steps of the procedure: your procedure contains a step that the other organisation does not carry out. The second places it in where the authority to approve sits: when approval rests with a central authority, that authority can slow the process down. The third places it in the means used to do the work: the other organisation uses a tool that shortens work which would otherwise be done by hand. The finding of half the time is the same in all three cases, but what it points to differs from one case to the next.

So a benchmarking exercise that ends with a table of figures has been wasted. One that ends with a list of procedural differences that can be tested is the one that is worth the time spent on it.

A table of figures records the gap: it shows that the other party is faster, and by how much. A list of procedural differences records where the gap may come from, item by item, and each item on that list can be tested. The list is what serves the purpose stated in the definition, which asks for the gap and its source rather than for the ranking alone.

External benchmarking against a sector figure

A figure published for an entire sector brings together organisations that differ in size, in activity and in operating model. For that reason, internal benchmarking over time remains the more reliable reference available in practice: your performance today against your performance a year ago, on the same measure and with the same definition.

A sector figure combines organisations of different sizes, engaged in different activities and run on different operating models, into a single number. An organisation that sets its result against that number is comparing itself with a mixture of organisations that differ from one another, not with one party. Comparing the organisation with itself over time keeps that mixture out of the comparison, because both sides of the comparison are the same organisation. It also meets the condition on which benchmarking depends: the measure is the same on both sides, and so is its definition.

How benchmarking differs from key performance indicators

A key performance indicator measures your performance against your own goal. Benchmarking measures it against the performance of another party. The first tells you whether the plan was achieved, and the second tells you whether the plan was ambitious in the first place.

The two differ in the reference against which performance is set. For a key performance indicator, the reference is a goal that the organisation set for itself. For benchmarking, the reference is the performance achieved by another party, which may be another branch or team inside the organisation, a direct competitor, or an organisation in another sector that has mastered the process. A key performance indicator can show that a goal was met, since it compares performance with that goal. Whether the goal was set high enough is the question that benchmarking answers, by setting the plan beside what another party has actually achieved.

This is an explanation of the concept, not legal advice.