Outsourcing metrics are measurements of performance used by companies to manage their outsourcing operations. Also referred to as key performance indicators, or KPI’s, metrics enable companies to enforce service level agreements (SLAs), compare vendors, and maintain day-to-day visibility into remote operations.

Many companies fail to recognize the value of metrics, looking only at end-of-the-day profits and losses. This may be due to a lack of management resources, systems sophistication or business process knowledge. In any case, it is dangerous because it can lead to backlog, utilization and performance issues, which can further lead to even worse issues, such as service penalties, government non-compliance and litigation. In short, you need metrics.

Metrics in Action

Metrics are more than just data points. They constitute actionable information that should play a role in company decision-making on a daily basis. As an illustration, call center metrics, such as average speed to answer (ASA), abandon rate, and average handle time (AHT) are metrics that require routine analysis and feedback to operations. If your forecasted call volume is inaccurate, you may end up with more calls than the scheduled agents can accomodate, or idle agents waiting for calls to come in. Forecasted call volume will also dictate ASA and abandon rate. Inaccurate AHT forecasts will have much the same effect. For example, in the case of callers irate about the length of time to answer, AHT may be impacted by the time needed to complain!

Examples of common call center metrics include: 

• Percent occupancy
• Number of calls offered
• Number of calls answered
• Percent calls answered
• Average speed to answer
• Average talk time
• Total talk time
• Average call waiting time
• Number of calls waiting
• Average hold time
• Number of calls held
• Average transfer time
• Number ofr calls transferred
• Number of abandoned calls
• Average abandon delay
• Percent abandoned calls
• Average ACD time
• Average ACD abadon rate
• Average after call work time
• Average break time
• Average staff time

These call metrics are just examples. They do not include outbound calling metrics (e.g. leads received, net callable leads, total contacts, etc.), or special campaign metrics, such as incentive compensation-based upsell and save-the-sale campaigns.

Examples of BPO metrics include:

• Number of FTEs (especially if you pay per FTE)
• Turnaround time or cycle time
• Accuracy rate
• Error, defect or failure rate
• Percent quality
• Percent rework

Several of the metrics above overlap. Detailed BPO metrics really depend on whether you are entering data, scanning documents, processing forms, etc. Much of what needs to be measured is volume, speed and quality, so it is not hard to identify the essential metrics.

Data Collection and Reporting

Metrics data is typically captured by an operations system, such as a PBX system in the case of a call center. Even highly-manual BPO operations such as data entry and transcription are logged in some manner or another. The more sophisticated and standards-based the approach, the better. Companies should request metrics as soon as an engagement starts and preferably in both summarized and non-summaraized (i.e. raw data) formats.

Depending on the service provided, metrics should be made available on a daily basis. If the vendor has a Web reporting system, the metrics should be made available online with the option to download them in Excel, CSV or XML file format. Vendor’s that don’t have this kind of reporting tool should extract their data and put it on a secure FTP or Internet file system location where it can be downloaded. More sophisticated metrics provisioning can be developed if both parties have the technology. For example, service transaction data can be delivered on a near-real time basis via application program interfaces (API’s) or Web services. While overkill for many situations, this kind of coupling enables true synchronicity between the outsourcer and the vendor.

Reporting of metrics can involve simple spreadsheets and macros, or more sophisticated business intelligence and data visualization applications. There are several good commercial packages for collecting, storing, aggregating and visualizing metrics. Those with strong charting capabilities, such as streaming Flash-based charts, are ideal for executive reporting.