Chapter 21 - Performance-Based Contracting and Service Level Agreements

The QASP is the guide that will be followed by both agency and supplier as the engagement is managed. It provides the methodology for monitoring performance against standards for required work. It provides for scheduling, observing, and documenting supplier performance against standards; accepting service; determining causes for deficiencies; and calculating payment due (formulas). Similar to the QASP is the supplier's quality control plan (QCP). The QCP will be developed by the supplier and will be submitted as part of the proposal for evaluation by the agency. After award, the QCP will be the plan the supplier is to follow during the performance of the contract. These two documents, the QASP and the QCP, are the control documents for the engagement.

Selecting the most appropriate surveillance method for the effort involved is important. Agencies should take into consideration task criticality, task lot size, surveillance period, performance requirements and standards, availability of quality assurance data, surveillance value in relation to task cost/criticality, and available resources. Careful selection of appropriate surveillance methods enables the agency to determine the amount of resources and associated costs needed to perform the surveillance task.

Surveillance results may be used as the basis for contract actions, including payment deductions, if provided in the contract or SOW. Acceptable surveillance methods include:

• 100 percent inspection:  This method, where performance is inspected/evaluated at each deliverable occurrence, is too expensive to be used in most cases. It is usually most appropriate for:

  • infrequent tasks
  • for small quantity but highly important products or services, or
  • when there are written deliverables and stringent requirements such as tasks required by law, safety or security.

• Random sampling::  This is usually the most appropriate method for recurring tasks. With random sampling, services are sampled to determine if the level of performance is acceptable. Random sampling works best:

  • frequent tasks
  • when the number of instances of the services being performed is very large and a statistically valid sample can be obtained.
  • for large quantity, repetitive activities with objective and measurable quality attributes.

• Periodic inspection:  This method, sometimes called "planned sampling” uses a comprehensive evaluation of selected outputs. Inspections may be daily, weekly, monthly, quarterly or unscheduled. Sample results are applicable only for the specific work inspected.

• Direct observation:  This method can be performed periodically or through inspection(s).

• Management information systems:  This method evaluates outputs through the use of management information reports.

• User survey:  The user survey method combines elements of validated user complaints and random sampling. Random surveys are conducted to solicit user satisfaction. It is appropriate for high quantity activities.

• Validated user/customer complaints:  This method is highly applicable to services provided in quantity and where quality is highly subjective. It relies on system or contract users to identify supplier performance deficiencies where complaints are then investigated and validated.

• Progress or status meetings:  Scheduled reoccurring meetings with contract users and suppliers are conducted to discuss progress made, problems encountered, problems resolved and/or plans for the next reporting period.

• Supplier progress reports:  The agency conducts analyses on regularly occurring progress reports delivered by the supplier.

• Performance reporting:  The agency evaluates performance or other required metrics for a specific time period.