VoIP Phones
in South Africa

Every enterprise telephony feature included. No hardware. No maintenance.

20-person South African business. Current PSTN vs CallVault VoIP.

Switching to VoIP is simpler than you think. Your numbers stay the same.

Everything that needs to happen before your first call on the new system.

Deploying a VoIP system for a South African business is a more involved process than simply ordering phones and plugging them in. A successful deployment requires careful pre-deployment planning, network readiness assessment, systematic configuration of the call routing logic, thorough testing before go-live, and proper staff training. Businesses that skip steps in this process typically end up with call quality issues, routing problems or staff who cannot use the system effectively -- all of which are avoidable with proper preparation.

Network readiness is the first prerequisite for VoIP deployment. VoIP calls are extremely sensitive to network impairments that would be imperceptible or tolerable for data applications. Packet loss above 1% causes audible crackling and dropped syllables. Jitter (variation in packet arrival time) above 30ms causes voice to sound choppy or robotic. Latency above 150ms causes participants to talk over each other because they cannot hear the other person quickly enough. Before deploying VoIP, Digitalx runs a network assessment using dedicated VoIP testing tools that measure the three critical parameters on the actual network path between your office and ICTGlobe's SIP servers. If the assessment identifies issues -- typically caused by network misconfiguration, inadequate QoS, or shared bandwidth contention -- we resolve them before the VoIP system is deployed.

Call flow design is the process of mapping out how inbound calls to each of your phone numbers should be handled. A simple office might have: all calls to the main number ringing the reception desk for the first 20 seconds, then extending to an overflow ring group of three additional staff, then going to voicemail if still unanswered after 30 seconds. A more complex call flow might include: an auto-attendant menu that routes different departments based on caller selection, after-hours routing to a voicemail-only mode or an emergency contact number, VIP caller identification that routes certain numbers directly to specific extensions bypassing the queue, and call recording rules that apply to certain departments but not others. We document the call flow in a diagram before configuration begins and obtain your approval before implementing it in the CallVault platform.

User acceptance testing before go-live is a step that many VoIP deployments skip and most regret. We configure the complete system on a staging basis, then conduct structured testing: inbound call routing testing (does the auto-attendant work correctly? do ring groups and fallbacks behave as designed?), outbound call testing (can users call landlines, mobiles and international numbers?), transfer testing (blind transfer, attended transfer, transfer to voicemail), conference call testing, softphone app testing on iOS and Android across different network types (WiFi, 4G, 3G), and call recording verification. Only when every test scenario passes does the system go live.

Staff training is the most commonly underinvested element of VoIP deployments. A technically perfect system that staff do not know how to use effectively delivers poor results. We conduct a two-hour group training session covering every feature staff will use in their daily work, followed by quick-reference cards covering the most common call handling scenarios. We also provide individual coaching for staff with specific roles (reception staff who handle high call volumes, managers who need the analytics dashboard, the administrator who will manage users and extensions). The goal is that every staff member can confidently handle calls on day one after go-live.

Why VoIP call quality issues happen and how we prevent them.

The most common concern SA businesses raise about VoIP is call quality. Stories circulate about businesses that switched to VoIP and experienced poor quality, dropped calls and frustrated staff and clients. These experiences are real but they are almost always the result of incorrect deployment -- specifically, deploying VoIP on a network that was not assessed for suitability, without proper QoS configuration, or on a consumer internet connection that does not provide the consistency that voice traffic requires.

Quality of Service (QoS) configuration is the network mechanism that prevents VoIP call quality from being impacted by other network activity. Without QoS, a large file download or a backup job running during business hours can steal bandwidth from VoIP calls, causing packet loss and the associated call quality degradation. QoS assigns different priority levels to different types of traffic: VoIP packets receive the highest priority and are processed before any other traffic type, regardless of other network activity. This means that even if your bandwidth is momentarily fully utilised by a large download, VoIP traffic is guaranteed its required bandwidth and packet delivery timing. Digitalx configures QoS at both the router level and the switch level as a standard component of every VoIP deployment.

Codec selection has a significant impact on VoIP call quality and bandwidth usage. A codec is the compression algorithm that encodes and decodes the voice audio for transmission over IP. The G.711 codec (the most commonly used for landline-quality calls) requires approximately 87Kbps of bandwidth per concurrent call. The G.729 codec provides similar perceived quality at approximately 25Kbps per call -- important for businesses with lower-bandwidth connections or a large number of concurrent calls. ICTGlobe's SIP platform supports both codecs and selects automatically based on the negotiation between the endpoint and the platform. For businesses with consistent bandwidth quality, G.711 is preferred because it avoids the processing overhead of compression. For businesses on LTE or lower-bandwidth connections, G.729 may be configured to reduce bandwidth requirements.

Monitoring and diagnostics are the tools that enable rapid identification and resolution of call quality issues after deployment. CallVault provides a call quality monitoring dashboard that displays per-call Mean Opinion Scores (MOS) -- a standardised measure of perceived voice quality rated on a scale of 1-5, with 4.0+ considered high quality and below 3.5 indicating noticeable degradation. When a specific call or group of calls shows poor MOS scores, the diagnostic data identifies whether the degradation occurred in the network path to the SIP server (suggesting an internet connectivity issue), within the local network (suggesting QoS misconfiguration or network congestion), or at the endpoint itself (suggesting a phone configuration or hardware issue). This diagnostic capability enables us to identify and resolve quality issues with precision rather than guesswork.