By Sramana Mitra and guest author Shaloo Shalini
SM: Very interesting! Tell me more about how this private cloud at TeleTech built. Can you talk about which vendors are supporting this initiative? This is a fascinating case study, and to me it sounds like a large-scale infrastructure. Are you running this from a central data center or multiple data centers with redundancy management?
CK: Of course there is redundancy. Our core data center infrastructure serving North America has two core GigaPOPs, as we would call them. They are geographically diverse. It is a mix of technology. At the APD level, we use both Avaya and Cisco. We also use the Aspect eWorkforce management tool. We have one of the largest implementation of Aspect deployed. In addition, we have our own call monitoring application called IQ360, which records calls – both desk and voice calls, and it also performs analytics. There is call evaluation tied to it as well. We have an agent performance management tool. I could take you through this entire process from the time a candidate gets in, or even before a job gets posted, to the time a candidate is selected through a screening process. From the evaluation process to the hiring process, from the training process –how we train new employees, bringing them online and getting them connected to our client’s connection, all the way through performance management and coaching of an associate. We leverage our GigaPOPs infrastructure in order to support and facilitate all this and more around the globe.
SM: What about support for North American clients?
CK: For North American clients, it would primarily hub into our two GigaPOPs here in the United States because we have a large Australian company that we serve. We have a point of presence there, and we anchor in Italy to serve Europe, which I mentioned earlier. Not only does this infrastructure support our brick and mortar sites, but we have nearly 5,000 at-home agents who use this infrastructure as well. What I am saying is that we can take a call anywhere on the globe. We could route it to Reno, Ohio, in the U.S., or we can take it to the Philippines or to your house. and it really doesn’t matter because the technology we have deployed enables us to do that.
SM: I see. Is Latin America also served through the North American GigaPOPs?
CK: Yes. Let me be clear about what I am saying here. The client base that we serve is primarily a North American base. I would say that 95% of the clients we serve are large global multinationals. They are based primarily in North America. What we do from an architecture standpoint is connect from their data center to our data center, and then we distribute calls anywhere in the world. In a typical program we have a multi-country delivery model. We may use on-shore resources for Spanish, but we may use Central and South American resources to fulfill that as well. For English, some clients may be routed to North American–based agents and others may be routed to different English-speaking countries around the world.
SM: What about knowledge bases? How are you managing your knowledge bases?
CK: It is a mix again. Often we use a client’s knowledge base, and we also have our own knowledge base tool called Responsive that is deployed for several of our clients. This is a hosted offering we make available for clients. With our knowledge base tool we help clients build their content, and build our content in our knowledge base, too. So, it is a mix. But everything is delivered through a virtualized infrastructure – GigaPOPs.
SM: In other words, even while you are using your client’s knowledge base tools, those are deployed through cloud technologies such that your agents can plug in to instances of those knowledge bases, is that correct?
CK: Yes. Everything is such that the connection happens through GigaPOPs and fulfillment is across that network. It looks the same regardless of whether that is the client’s CRM application or whether that is the client’s knowledge base or if it is fulfillment of some other application.
SM: How does security work in this environment where TeleTech’s agents are plugging in to clients’ CRM systems which contain, I presume, hypersensitive data?
CK: Actually, I think it is one of our key benefits; I say this because we don’t have servers, images, and data centers on site. When you go into one of our sites, our MDF and IDF look pretty small because there are not a lot of recording devices there, and not a lot of servers or databases. Our data is in our centralized data infrastructure. It is highly secure. It looks like the standard data center with controlled access. It is part of what enabled us to be PCI compliant, SAS-70 compliant, and HIPAA compliant because we have only a few points where we ever store this data and from which our agents access data.
SM: You are talking about storing data, but your agents are accessing, touching, and changing systems and data on the clients CRM systems, right?
CK: Correct! But it is an agent’s desktop, right? That is viewing a client system. The client’s systems data, that gets updated at the client database.
SM: Right. I guess the question I am asking is, What are the security policies of managing that transition? Are you managing security in the cloud, and how is it being managed?
CK: There is a dedicated network between our client and our data center. So, in effect, it becomes a private network. There are, of course, strict security processes to control that private network. Only specific agents have access, on a program level, to the specific client systems that they will be working on. It is rights managed as to what systems they actually can access. The actual database, that we are not managing at all. We don’t own the customers’ CRM database, which is in their data center. Does that make sense?
SM: Okay, got it.