Benchmarking continues to be an emotive and, at times, difficult issue for customer and supplier in the outsourcing process, both during contract negotiations and when it comes to implementation. This article identifies the key issues and concerns that should be considered and addressed in the benchmark terms of the outsourcing contract, and suggests approaches that can be taken to achieve a more realistic and balanced benchmark process.

For most customers, benchmarking is seen as an important contract tool to help ensure competitive pricing in long-term outsourcing contracts. As competition between suppliers increases, and technology costs continue to fall, customers are increasingly seeking benchmark terms "with teeth", which require suppliers to reduce prices during the contract term in line with market pricing.

Suppliers, unsurprisingly, view benchmarking differently. While their general perception of the benchmark process has improved in recent years, suppliers continue to approach benchmarking with some suspicion, concerned that the benchmark process is often poorly constructed and implemented, and that it fails to strike a fair balance between customer and supplier interests.

As a result, benchmarking continues to be an emotive and, at times, difficult issue for customer and supplier, both during contract negotiations and when it comes to implementation of the benchmark process.

This article seeks to identify the key issues and concerns that should be considered and addressed in the benchmark terms of the outsourcing contract, and suggests approaches that can be taken with a view to achieving a more realistic and balanced benchmark process.

It considers the following topics:

  • Preliminary considerations, such as whether benchmarking is necessary at all in view of the contractual term.
  • Considerations on the appointment of the benchmark adviser.
  • Agreeing the benchmark target.
  • Selecting the benchmark comparison sample.
  • Agreeing the frequency of the benchmark.
  • The techniques used and the parties' main obligations during the benchmark process.
  • Consequences of benchmarking (for example, whether the benchmark is binding or non-binding).


Preliminary considerations

Contract term

The average duration of outsourcing contracts has fallen in recent years: whereas, in the past, a term of ten or more years was not unusual, in the current market, an initial term of between three and five years is much more common.

As a result (and given the time and effort that is often involved in negotiating and agreeing benchmark terms), customers should consider carefully, at the start of the contract negotiations, whether formal benchmark rights are required given the contract term. Customers will normally have market tested the supplier's charges as part of the initial contract tender process and, for many customers, therefore, formal benchmarking is seen as a 'nice to have' rather than a 'must have' in contracts with an initial term of three or four years. For contract terms with a longer initial period, most customers will generally seek to include an ability to benchmark the supplier's charges at regular intervals during the term.

Status of benchmark results

A second preliminary consideration, which will have an important influence on the parties' approach to negotiation of the benchmark provisions, is the contractual status of the benchmarker's report and recommendations.

Customers are increasingly seeking benchmark provisions which require suppliers automatically to adjust their charges to match the market. This can be one of the most contentious aspects of the negotiation of the benchmark terms and, particularly where the amount of any automatic reduction is uncapped; can be a walkaway point for some suppliers. It is also likely to lead to protracted negotiations on some of the other key commercial issues relating to the benchmark provisions (such as the benchmark target, for example).

Suppliers, of course, prefer that the results of the benchmark report are not automatically binding, and that they simply act as a trigger for price renegotiation. Although this amounts to nothing more than an unenforceable 'agreement to agree' in legal terms, some customers consider it to be of value from a commercial standpoint and, because the approach generally leads to an easier negotiation of the other benchmark terms (and quite often a more favourable position on those terms), some customers are prepared to agree to non-binding benchmarking, especially where the services are highly commoditised, and the costs of switching to a replacement supplier (in the event that the price renegotiations are unsuccessful) are low.

Because the approach chosen will have an important influence on how negotiation of the benchmark terms is conducted, customers should decide their preferred approach at the outset, prior to the start of contract negotiations, taking into account the relative pros and cons of each, and their objectives in including benchmark provisions in the contract (see below, Consequences of benchmarking: Binding or non-binding).

What can be benchmarked?

A final issue for early consideration is how much flexibility the customer requires in relation to the services to be benchmarked. This is important in contracts involving the outsourcing of more than one related service line. In that case, customers will often wish to retain the ability to benchmark individual service lines. This can be problematic for suppliers, for a number of reasons. Chief among these is that, in contracts involving the outsourcing of multiple service lines, there is normally an element of cross-subsidisation of pricing between service lines, and to allow the customer to "cherry pick" the services that are benchmarked would, therefore, result in an inaccurate and unfair comparison.

This is a legitimate concern on the supplier's part. Customers who wish to retain the flexibility to benchmark by service line will need to make the supplier aware of this early on in the tender process, to allow the supplier, where possible, to price the service lines on a standalone and unsubsidised basis.

Appointing the benchmark adviser

Identity of the adviser

It is in both the supplier's and customer's interests to appoint an experienced benchmarker with access to the most up-to-date market data. Suppliers will also wish to ensure that the benchmarker is independent of the customer and is not a competitor. The appointment of an independent benchmarker who the parties feel can carry out the benchmark in a fair and even-handed manner is essential if the parties wish to minimise the risk of dispute at the time a benchmark right is invoked.

The benchmarker is therefore generally appointed by mutual agreement. In some respects it makes most sense to appoint the benchmarker at the time of the benchmarking, as this will allow the parties to select the benchmarker with the most up-to-date and accurate data. Customers, however, are understandably concerned that this approach could lead to a delay in the benchmark process, and potentially be used by the supplier as a stalling tactic. To address these concerns the parties will generally agree, in the contract, a pre-approved list of acceptable benchmarkers from which the customer can choose at the time of the benchmark exercise.

Terms of appointment

In early outsourcing transactions, it was not unusual for the customer to instruct (and pay for) the benchmark adviser. Suppliers are understandably concerned that this approach affects the benchmarker's independence and objectivity, and may influence the benchmarker's approach to the benchmark exercise and the conclusions reached.

For this reason (and because the supplier will also derive value from the benchmark exercise in understanding how competitive its charges are), it is now common for the benchmarker to be jointly instructed by customer and supplier and, in many cases, for the benchmarker's charges to be shared equally between the parties.

The benchmarker will generally be appointed under a tripartite agreement with the supplier and customer, which sets out each of the parties' respective rights and responsibilities. Suppliers will want to ensure that the tri-partite agreement contains robust confidentiality terms, limiting the use that the benchmarker can make of any supplier pricing information disclosed during the benchmark process. The benchmarker will generally seek to retain the right to add the data it gathers from the supplier, on an anonymised basis, to its database of contracts to:

  • Build its market cost estimates.
  • Help improve the accuracy of the benchmarker's findings in future benchmarks.

Agreeing the benchmark target

The benchmark target is the price point that the supplier's charges must meet in order to be considered competitive. Market practice on this issue continues to evolve, as customers, suppliers and their advisers learn from earlier outsourcing projects.

Until recently, the benchmark target in many outsourcing contracts has been based on a percentile measurement, with the lowest quartile (25%) being one of the targets most commonly used.

This approach assumes that the benchmarker will have a large number of data points against which to make a comparison. The reality, however, is that the benchmarker will rarely have a sufficient number of data points available to make a percentile target a meaningful measurement. Once the benchmarker has taken into account any relevant normalisation factors to help ensure it is comparing like with like, the size of the comparison sample quickly shrinks (see below, The benchmark process: Normalisation and price adjustments). In all but the most commoditised of outsourcings, the benchmarker will generally work from a comparison sample size of between four and ten.

For this reason, most benchmarkers now prefer to work using a benchmark target of the mean average price of the comparison sample. This approach is slowly finding its way through to the benchmark provisions of outsourcing contracts.

There is also an increased recognition among customers that benchmarking is an art, not a science, with the outcome inevitably dependent, to an extent, on the benchmarker's own judgement (particularly when it comes to selecting the comparison sample and applying any normalisation factors (see below, Selecting the benchmark sample and The benchmark process). For this reason, the parties will sometimes agree a tolerance threshold to the benchmark target, so that the supplier is only required to adjust its charges where they exceed the benchmark target by an agreed percentage.

Selecting the comparison sample

Final selection of the comparison sample (peer group and sample size) will normally be left to the benchmarker, and will depend on the data points available to the benchmarker at the time. However, the outsourcing contract will normally set parameters for selection of the comparison sample within which the benchmarker must operate. These parameters will generally include a:

  • Defined minimum sample size.
  • Definition of the organisations that are capable of forming part of the peer group.

Sample size

Most benchmarkers will generally seek to identify a minimum of four comparators in any benchmark exercise, and this is often the minimum number stipulated in the outsourcing contract. While suppliers generally prefer to specify a larger number as a minimum, this can be counter-productive for the supplier, in that it may force the benchmarker to include within the comparison sample contracts that are not appropriate comparators.

Defining the peer group

Approaches to defining the peer group vary, and the following approaches are common:

  • Defined by reference to those outsource providers that the supplier considers to be genuine competitors (in terms of size, scale, geographic footprint and reputation).
  • Defined by reference to the customer organisation, with the benchmarker instructed to take into account pricing in contracts involving customers of a similar size, geographic spread, and services service levels of a similar nature to those being provided under the contract in question.

Frequency of the benchmark

Suppliers will typically seek to agree a "honeymoon" period following contract signature during which its prices cannot be benchmarked, and to limit the number of benchmarks that can be carried out once that period has ended.

Customers, on the other hand, like to retain maximum flexibility and the right to benchmark as often as they consider necessary during the contract term.

While this issue can often be the subject of extensive negotiation, in practice it need not be. As the customer is likely to have benchmarked the supplier's pricing as part of the original contract tender process, in reality the customer is probably conceding little by agreeing to an initial honeymoon period (provided of course that the period is reasonable).

Following any initial honeymoon period the frequency with which benchmarking may be carried out will depend, among other things, on:

  • The contract term. A right to benchmark only once in a contract of five years or longer is unlikely to be acceptable to most customers.
  • The complexity (and duration) of the benchmark process. The duration of the benchmark will vary depending on the services and contract concerned, but for some benchmarks, it can take as long as six months from the start of the process to finalisation and implementation of the benchmarker's recommendations. In these cases, the reality is that the customer will probably not want to carry out benchmarking annually (particularly when duration is coupled with the costs of the benchmark exercise). In these cases, customers should consider if a right to benchmark every couple of years is sufficient as a form of price protection. In other cases, where the market is mature, the services are commoditised, and prices are dropping quickly, annual benchmarking may be more realistic and valuable to the customer.
  • Whether the customer can benchmark by service line. Where the customer is entitled to benchmark by service line, it should ensure that any limit on the frequency of benchmarking applies per service line only.

Customers should also bear in mind that, if the supplier is agreeing to share the benchmarker's charges, the supplier may factor those costs into the charges the customer must pay, and that the inclusion of a right to benchmark annually may therefore result in an increase in the charges that the customer is required to pay to the supplier.

The benchmark process

Normalisation and price adjustments

One of the key tasks that the benchmarker will be required to carry out, to ensure that the benchmark is carried out as fairly and objectively as possible, is price normalisation. This involves the benchmarker taking into account a range of factors that are likely to affect pricing under the contract in question or the contracts within the comparison sample.

While the normalisation process (and the factors to be taken into account) will normally be left to the benchmarker to determine, the parties will often agree in the contract a non-exhaustive list of factors that the benchmarker must take into account. These will often include some or all of the following:

  • Service volumes. Greater service volumes are more likely to result in efficiencies, economies of scale and, therefore, more competitive pricing.
  • Service levels and other performance standards. The service levels that the supplier must meet may have a direct impact on its costs and therefore its charges. For example, if (in an IT outsourcing) the customer requires a high availability solution, this may mean the supplier providing "hot" back-up arrangements or a dual data centre solution, both of which are likely to have a material impact on price.
  • Other contract obligations. Other contract obligations may also have a direct impact on the supplier's charges. If, for example, the supplier has been asked to commit to technology refresh obligations during the contract term, the supplier will make provision for this in its pricing. Any contractual obligations of this nature which have a material impact on charges should be expressly provided for in the contract. Additionally, suppliers should, where possible, keep a record of any additional costs included in their charges as a result of these contractual commitments.
  • Financing arrangements. It has become increasingly common for suppliers, in outsourcing transactions, to finance some or all of the customer's transition costs, and to recover those costs during the contract term as part of the regular service charges. This has a number of advantages for the customer, including allowing it to achieve cost savings early on in the contract term (which often it would not otherwise be able to do). Where the supplier finances transition in this way, it will want to ensure that this is taken into account as part of any benchmark (as the suppliers' charges during the contract term will be higher than would otherwise have been the case). For suppliers, it is important to keep a clear record of those financing costs, so that they can be provided to the benchmarker during the benchmark process.

Obligations during the benchmark process

The outsourcing contract should capture the parties' key obligations during the benchmark process. These will typically include:

  • Co-operation and assistance. The contract should include an obligation on both parties to co-operate with the benchmarker, including by providing the benchmarker with the information it needs to carry out the benchmark exercise. Customers should be wary of any general qualifications or carve-outs to the information that the supplier is required to provide, as these can give rise to disputes at the time of the benchmark exercise.
For most suppliers, the key concern will be disclosure of information about its cost base, as it will not wish to disclose the profit margin it is making under the contract. In most cases, this should not be an issue for the customer or the benchmarker, as it is not the supplier's margin that is being benchmarked, but the charges overall. If the supplier is able to provide the services more efficiently than its competitors, allowing it to make a higher margin, then the supplier should not be penalised for doing so if the overall charges meet the benchmark target. (An obvious exception to this is 'cost-plus' contracts, where the supplier's costs and profit margin may well be the subject of the benchmark.)
  • Adequate resourcing. A benchmark exercise will require both parties to devote resources to allow the benchmarker to carry out and complete the benchmark exercise. The contract should include an obligation on both parties to do so, including the appointment by each of a benchmark manager, who will act as primary point of contact to the benchmarker, and co-ordinate the other resources required.
  • Timely input. The benchmark provisions should require both parties to provide any input required (for example, feedback on the benchmarker's draft report) within clearly specified (but realistic) timeframes.

Consequences of benchmarking

Binding or non-binding?

This is often the most emotive and heavily negotiated aspect of the benchmark terms of the contract. Approaches vary, and much will depend, ultimately, on the parties involved and the relative strength of bargaining positions during contract negotiations.

Customers will, ideally, want the results of the benchmark report to be binding, such that if the supplier's charges fail to meet the benchmark target, the supplier must reduce its charges to bring them in line with the market. Suppliers are understandably concerned with this approach, in that it potentially hands over control of the margin they make to a third party. Although market practice continues to evolve, for some suppliers an uncapped obligation to reduce charges is a deal-breaker in negotiations.

The supplier's preference, of course, is for any benchmark report to be indicative only and to be used (if at all) to frame discussions or negotiations on pricing, with any adjustment to pricing to be subject to agreement. For many customers an agreement to agree of this nature will be unacceptable, and will not offer the degree of price protection they need to commit to a long-term deal.

There are a number of alternatives to these two positions. These include:

  • Termination rights. One approach that is often proposed (particularly by suppliers) is for the benchmark results to be non-binding, with a commitment on the parties to reach agreement on an adjustment to the pricing within a set period following finalisation of the benchmark report, and a right, on the customer's part, to terminate the contract (either as a whole or in relation to the services that have been benchmarked) if the parties are unable to agree a price adjustment within that period.
There are some difficulties with this approach. For example, the parties will need to agree whether this is treated as a termination for convenience, most likely resulting in an early termination charge, or a termination for cause, where no termination charge is payable. Customers will naturally take the view that they should not have to pay an early termination charge (on the basis that termination in this scenario is far from "convenient"). However, the likelihood is that if the supplier has financed the costs of transition or the run services, it will wish to recover those costs from the customer even if the contract is terminated for this reason. Many suppliers will, however, agree that any termination charge in this scenario will be limited to recovery of unrecovered costs only, and will not, for example, include any element of lost profit.
For customers, this compromise position remains problematic, as they are still faced with a potentially large and prohibitive cost on termination. This is, of course, in addition to the costs the customer will have to incur in transitioning the services from the supplier (either back in-house or to a replacement supplier). These costs will often outweigh the savings that the customer would make by transitioning to a cheaper provider, with the result that any termination right in this scenario is often meaningless in practice. (In limited circumstances, where the services are highly commoditised, and the costs of switching to a replacement provider are relatively low, a right on the customer's part to terminate (without payment of an early termination charge) may provide useful leverage to ensure a meaningful renegotiation on price, informed by the benchmark results.)


  • Capped price adjustments. A more meaningful alternative (from the customer's standpoint) is for the benchmarker's report to be binding subject to a cap. This means that if the supplier's prices fail to meet the benchmark target, the supplier must adjust its prices to meet the benchmark target, subject to an overall cap on the amount by which it must reduce its pricing.


While not ideal for either party, this approach can sometimes be the only one that provides the customer with some form of binding price protection, while at the same time ensuring that the supplier is not locked into a potentially loss-making deal.

Price adjustments: forward-looking or retrospective?

A final issue that needs to be addressed in relation to any price adjustment is whether the adjustment is forward-looking only, or whether it will have retrospective effect.

While many customers will seek to insist upon some form of retrospective price adjustment (often to take effect from the start of the benchmark process), this approach causes several issues for suppliers, including in relation to revenue forecast and (potentially) revenue recognition.

Forward-looking only adjustments are however not without risk for the customer. The primary risk with this approach is that the supplier will seek to prolong any benchmark exercise with a view to delaying the point at which it is required to make any price adjustment.

To address the problems associated with these two alternatives, one compromise that is sometimes adopted is for the price adjustment to be forward-looking only with a back-stop date (perhaps aligned with the period of time a benchmark of that type would typically be expected to take) to which the charges will be retrospectively adjusted in the event that the benchmark exercise runs beyond that date.

Conclusion

Benchmarking remains an important form of price protection for customers in long-term outsourcing transactions, providing a mechanism to address price drift over the contract term. However, the benchmark process is not without its challenges: it can be complex, resource intensive and, ultimately, is an art, not a science.

Market practice continues to evolve in relation to the commercial aspects of the benchmark terms, as suppliers, customers and their advisors, learn lessons from earlier outsourcing deals. There is now a greater recognition within the industry that, in order to create a benchmark process that is workable in practice, it is important that the contract terms on benchmarking are fair, balanced, and reflects the legitimate concerns of both supplier and customer.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.