• 08/16/2011
    2:36 PM
  • Network Computing
  • News
  • Connect Directly
  • Rating: 
    0 votes
    Vote up!
    Vote down!

Data Center Pricing Schedules: Driving Positive Vendor Behavior

One of the most scrutinized and highly negotiated elements of an IT outsourcing (ITO) agreement for infrastructure is the resource pricing schedule. Like a tax table or any other financial schedule, the ITO rate structure has a significant impact on the behavior of those governed by it--in this case, the service provider.

2. Structure RFPs to solicit line-item pricing: The structure of the data center pricing schedule begins with the terms of the RFP, which leads to the nature of the resource quotes in the vendor response. This is then translated to the final pricing schedule structure in the contract. Even if your organization is predisposed to a holistic outsourcing approach inclusive of support services, facilities and hardware assets, there is merit in structuring the RFP such that it solicits discrete pricing for remote infrastructure management (RIM) services, facilities hosting and asset ownership/lease cost components. This level of detail allows for more effective cost management over time, including an improved forward-pricing discount negotiating position. It also allows clients to more aggressively negotiate facilities costs and lease rate components that have little impact on service quality, while more carefully negotiating RIM rates that may have a more direct impact on staffing levels and service quality. It also provides an ancillary benefit of making the outsourcing engagement more contract-friendly for offshore RIM providers partnering with onshore co-location providers--often resulting in very compelling hybrid proposals.

3. Make provisions for non-production resources: The prevailing practice of gold/silver/bronze service levels--typically associated with availability (for example, 99.95%, 99.9% and 99.5%, respectively)--is becoming antiquated. Virtualization technologies are creating a level of abstraction between logical instances and the underlying hardware, and hardware/OS platforms are becoming more standardized in content and automated in maintenance. As a result, it is increasingly difficult to isolate and price workloads by thinly stratified classes-of-service. Further, for 99.99% and higher availability requirements, special configurations are needed (automated failover, clustering), requiring separate and distinct pricing schedules and support service descriptions.

The more relevant issue to address is production vs. non-production support requirements. In a classic client-server development model, as much as 50% to 75% of server and associated resources are non-production ("sandbox," development and QA instances). These resource environments deserve individual consideration to specify an appropriate level of support services, and this typically results in substantially discounted resource support rates. The bottom line: Don’t force non-production resources into production support service levels and rates.

4. Anticipate requirements for platforms and services currently out of scope: Client organizations are rarely more exposed than when faced with the prospect of negotiating for new support services or new platform support mid-term in a multiyear agreement. This situation may result from an acquisition or a major architectural/technology change in direction. A particularly challenging task, it calls for planning for the unknown; however, there are at least two risk remediation strategies to consider. The first is to anticipate possible platform or technology additions and negotiate rates for these up-front. Even with unknown quantities, vendors will typically be willing to offer reasonable, not-to-exceed rates to win your initial business. The second is to establish a price arbitration or benchmarking clause that sets out specific procedures for pricing of new services with objective third-party oversight.

For "disruptive" technology advances that result in step-function decreases in the cost of services delivered, cost reduction benefit division between client and service provider can and should be negotiated in the contract. However, as always, the devil is in the details of the cost reduction measurement and valuation.

The other side of new service requirements involves project-driven elements. All resource-based variable usage contracts should also contain labor rate tables by job category and skill level, and include contract terms that specifically allow the use of alternative service providers for project activity. Additionally, project hours should be estimated and controlled by the client organization and should not require a change order (otherwise known as a change request or project change request); rather, they should simply represent another billable resource on the pricing schedule. Habitual use, or misuse, of the contract change order process for project work or off-pricing schedule equipment additions is a recurring theme in ITO contract cost overruns--death by a thousand cuts.

At the end of the day, standard cost-containment techniques such as resource usage audits, rate benchmarking, and other IT governance and vendor management practices are must-haves to maintain equitable relationships with service providers. While not an exhaustive list, the aforementioned four key data center pricing schedule and contract considerations will help ITO clients more appropriately define resources and management approaches--ultimately driving positive vendor behavior and optimizing the company’s data center utilization and pricing schedule.

About the Author
Colin Rankine is a senior associate at Pace Harmon, an outsourcing advisory services firm providing proven guidance to Fortune 500 and high-growth middle market organizations on complex outsourcing and strategic sourcing transactions, process optimization and vendor program governance. He has more than 20 years of IT industry experience with a focus on operational efficiency and service level quality, including 10 years at IBM in technical field positions and 10 years at Forrester Research, where he served as VP of the computing infrastructure group.

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.

Log in or Register to post comments