Are benchmarking clauses in outsourcing contracts overused?

Stuart Harris, a partner and Data Management Center of Excellence leader at TPI, makes a compelling case for this viewpoint in a recent IT Business Edge interview.

"The use of benchmarking as a regular, annual tool to try to chisel 2.1 percent or 1.6 percent from a service provider does not do justice to managing that relationship," Harris says.

Rather, benchmarking should only be employed when there is a significant discrepancy -- agreed upon in the contract terms -- between the price a client is paying for services and the fair market price.

Fair market price can be hard to nail down, due to a dearth of support data as well as other factors. So the threshold for discrepancy should be set no lower than 10 percent, Harris recommends.

If an outsourcing provider does not want to make a price change in this instance, Harris suggests, then the client should be given an opportunity to exit the deal at little or no cost.

This CIO.com article details some of what Harris calls "cuteness" in his interview -- efforts by both service providers and outsourcing clients to derail the benchmarking process, ultimately rendering it useless.

It also identifies several specific areas that characterize successful benchmarking initiatives. For example, the reference group should be as similar to the client's deal as possible, in terms of size, scope, geographic distribution and other criteria.