A discussion is provided of Quality of Service (QoS) and changes needed when of QoS must provide end-to-end support when voice is converted to IP packets in the enterprise and must reach more portions of the data network. IT organizations therefore have to improve support and manage converged infrastructures, especially because telephone users expect to have dial tone, call completion, and consistent, high-quality sound on their voice calls. Because end users are less technical, network administrators have to listen to the experiences of the end user and find possible fixes. For instance, if the caller and the called party seem to have talk-over during their calls, the problem is usually a latency condition. The ping utility is provided with all TCP/IP-ready PCS and workstations, and can be used to test end-to-end latency. The user or the administrator can set the ping packet size and determine transit times to other nodes. More specific information can also be gathered with the widely available tracert (traceroute) TCP/IP application. Also discussed are the good and acceptable ranges of latency, jitter, and packet loss. In order to avoid any of these problems from the outset, policies and procedures can be established for all phases of a VoIP implementation. Such policies are described and include tasks required for planning and design, implementation, documentation, and management and monitoring.