Friction points in a price benchmark

The objective of a market price benchmark is to obtain a binding and neutral expert assessment of the market price for certain services. It also clarifies the question to what extent the agreed prices are (still) competitive in relation to the quality delivered. Therefore, most current sourcing contracts have a benchmark clause to establish parameters in advance for reviewing the agreement and to avoid friction points afterwards.

But despite a clause, there are many places where tensions between the parties can be unleashed. This is often the case when there are only statements of intent in the agreement instead of describing the control mechanisms and consequences in detail. The following points are potential sources of friction:

General conditions: First, it must be clarified who can initiate a benchmark at all - the client alone or only together with the service provider? In addition, it is important to address timing in the clause. For example, it has proven useful that the first benchmark can be started after twelve months at the earliest. It is also important to agree on a time period to which the benchmark should refer. After all, prices (which are usually in the contract) and quantities/volumes such as the number of changes or incidents also have to match the survey period. Sometimes this is difficult if you agree on a current year for which there is no data (quantities).

Scoping: What should actually be the subject of a benchmark: parts of the outsourcing agreement, the entire agreement, rate cards for time & material services such as for changes? It is also obvious that the benchmarker should not have a conflict of interest with one of the parties - a shortlist is a good idea here. However, acquisitions in the benchmarking market can quickly lead to lengthy arguments.

Participation: "Both parties shall use their best endeavours to assist the benchmarker in the preparation of its report". So far, so good. And what does it mean that requested documents are passed on "without delay"? In case the parties do not agree within 30 working days, the benchmarker can be appointed in advance to determine the services to be examined.

Peer group: The selection of the peer group is often reminiscent of the jury procedure in US courtroom films. A wide field opens up here, from the number of peers to general contractual modalities to trading conditions and tax issues as well as technologies or offshore quotas. Less would sometimes be more, especially since the many paragraphs are not always stringently drafted and thus open up further areas of uncertainty. An example: you have defined seven peer companies to be used for the calculation and later focus on the "better half" in the valuation. This then has to be clarified again in a time-consuming manner.

Reference values: If the reference value is determined by the "average of the better half of the comparison group", what is actually meant by this? The best fitting half in terms of service delivery (provider proposal) or the cheaper half of the comparison group (client proposal)? Here, both sides can build up arguments to spin the project in their direction. Therefore, it is important to determine in the clause what exactly the reference value refers to. In addition, it pays to decide on simple values such as the median and to dispense with individual formulas.

Implementation: After a benchmark, the parties often have 30 working days to comment on the outcome. But when does the (mostly) reduced price come into effect - from the time the result is communicated or when agreement is reached after the comments have been submitted? In large agreements, it can be a matter of millions that one side does not get and that the other side does not have to pay.

Adjustment: In general, new prices after a benchmark offer room for discussion. If, for example, the deviation is higher than five per cent from the value of the peer group, the service provider (if contractually negotiated) has to come down to the market level. However, it is often debatable whether this refers to the overall benchmark result, an IT tower or a single price point. We do not recommend the latter - prices are often a mixed calculation and subsidise other areas. It is fairer to choose sporting target values such as the first quartile (the cheapest 25 per cent of the comparison group) at tower or overall level.

A benchmark clause? It depends

Conclusion: Over the years, our consultancy projects have shown that benchmark clauses are becoming more and more detailed. Service providers and clients are increasingly insisting on their right to have a say, because after all, a lot of money is at stake. Thus, the parties involved give some thought to the benchmark in advance - but on the other hand, they do not define it precisely enough, for example in terms of reference values or dates. This then leads to lengthy discussions which often strain the relationship between the contracting parties. Best practice: If both sides absolutely cannot agree on a point, they should at least agree that a neutral authority will decide: either the benchmarker or take a chance and draw lots.

In a previous blog post, we put together the basic elements of a good benchmark clause.