Harnessing Contact Center Data for Actionable ResultsHarnessing Contact Center Data for Actionable Results
Sorting through the sources of information -- what’s needed and why -- can be confusing
July 9, 2019
Contact centers have always had a plethora of data, but in the last few years the amount and sources of have grown dramatically. Still, when it comes to supporting an integrated contact center, understanding what data to use and getting robust measurements can be tricky.
What Contact Centers Expect
In the past, contact centers have emphasized historical and real-time information about inbound and outbound calls -- the quantity; Dialed Number Identification Service; the customer’s automatic number identification; time in queue; status (answered, abandoned, overflowed, or went to voicemail); which agent answered; call duration; time spent in after-call work). Agent-specific data, such as login and logout and all the various agent activities, like call codes and work modes, also has been important; this data is captured by the ACD or contact routing application.
Contact centers also have placed immense focus on agent activity and adherence to schedule, which are important for meeting access or service-level goals. Agents go into various modes throughout the day. For example, they might be active working on inbound/outbound contacts, doing after-contact work, available to take calls, in training, at lunch or on break, receiving coaching, doing research, and other modes specified in the contact routing and workforce management (WFM) application. The WFM application monitors these to ensure agents are in the right mode at the scheduled time to meet customer demands. The WFM application also enhances forecasting and scheduling capabilities, to various levels -- monthly, daily, interval -- and uses metrics such as volume and average handling time to determine staffing requirements.
In addition, the contact center’s quality team usually measures and reviews agent quality scores. To do so, quality analysts review calls, along with the screen utilization during the calls, to see how agents handle customer interactions. They assess foundational skills, such as proper greeting, identity verification, and correct information, along with soft skills. The contact center logs this information in a quality monitoring application.
If the contact routing solution is an integrated system, it provides statistics on the other customer contacts -- chat, SMS, email, fax, video, and social. Integrated means one application does all the routing of the various contacts and data collection. However, I’ve seen many sites handle chat using an application outside the voice routing solution, along with social on another application platform, and agents usually have to be assigned to only one or the other -- voice, social, or chat contacts. This also means the contact center has more sources for collecting chat and social data. An integrated system keeps all contacts within one application for routing, historical, and real-time information. It simplifies support, reporting, and agent interfaces to the various platforms.
However, even when supporting integrated routing, not all vendors provide comparably robust and detailed statistics to those available for voice calls today. As I’ve gleaned through the many discussions I’ve had with vendors, they’re beginning to understand why better data on these newer contact types is required.
Click below to continue to next page: What Contact Centers Need
What Contact Centers Need
For example, contact centers supporting chat need two types of chat interaction time:
Total chat engagement time -- measured from the time the agent initiates conversation with the customer to chat completion. If a single agent is chatting with two or more customers, overall chat handle time may be inflated, thus throwing off forecasting time for average chat active time (ACAT). Additionally, if agents are handling more than one chat or email at a time, it may be perceived as “double” counting time. This can also be true when agents are handling email and fax contacts.
ACAT -- this is a measure of the time an agent is active on the chat. An agent is either active on one chat or another, but not more than one, so there’s no inflation associated with this measure. While an agent may be engaged in more than one chat, time isn’t counted in the absence of active engagement. ACAT will provide better data for forecasting and a better measurement of chat handle time for agents then the total chat engagement time measure. Furthermore, an inactive chat is considered suspended time; calling it “hold” time would be confusing because in a voice call when agents put callers on hold, they’re usually doing research or calling for assistance, so voice hold time is considered part of handle time.
Now some chat vendors are providing detailed data as above, yet not all. When looking at an integrated routing solution, be sure to ask vendors granular questions about the statistics they can and cannot capture with their systems, as well as level of reporting and data structure they provide.
Many contact centers also have an outbound campaign management system for voice, emails, and/or SMS. The data included in those systems is also plentiful, so be sure to list all the data and metrics you would like to get from these systems.
I’ve mentioned four sources of contact center data, and provided a few examples of the type of data available from each. To recap, they’re:
Contact routing applications
Outbound campaign management systems for voice, emails and SMS (isn’t integrated)
WFM
Quality management
Organizations also may be able to gather data from a slew of other applications, such as:
Web reports and analytics for customer behavior and data
IVR reports and analytics for customer behavior and data
Desktop analytics
Customer surveys (CSAT, NPS)
Learning management system
Knowledge management system and related usage information
Customer relationship management system
Each has its own database, data structure, and sometimes rather cryptic manuals describing the data… if they have any information at all about what some of the data fields mean and how they’re calculated. If the vendor doesn’t provide a data dictionary or manual for use during product evaluations, you should hesitate to investigate it any further.
With so many data sources for the contact center and the business, it’s important to determine contact center and business goals, target improvements, and possible actions for the enhancement so as not to be overwhelmed. Then determine how to measure for your needs, and from what data sources.
If you’re looking at current reports that you haven’t defined yourself, it’s important to step back and ask:
Is the information adding value?
Does the report present actionable information?
What purpose and at what level -- by executives, IT and business managers, supervisors, agents -- is the information being utilized?
This was just a small dip into the contact center world of data sources, which reaches far beyond just the contact center. Knowing all the different applications that provide data, how each data element with that application is measured, what it is used for or can be used, is key for the contact center manager, business analysts, and the workforce management team to understand. It can be overwhelming, yet with the right strategy and people, all the data sources and measurements available can be harnessed and used to make actionable decisions or measure results of actions taken.
"SCTC Perspectives" is written by members of the Society of Communications Technology Consultants, an international organization of independent information and communications technology professionals serving clients in all business sectors and government worldwide.
