IPDR allows the collection of data from the CMTS in streams. Not only can you pinpoint – with high accuracy – the utilization of every cable modem on your network, but you can also identify what interface, node, and service group this traffic comes from, how much bandwidth each interface utilized, establish the maximum throughput, and discover possible congestion points in the interfaces, nodes, and service groups.
You can apply this data to every department of a modern cable operator — from engineering and planning to marketing and sales.
On the engineering and planning side:
- Understand where network congestion occurs. Gain visibility through notifications and key performance indicators of congestion points and traffic patterns that can lead to congestion.
- Predict congestion points to make smarter investments. Analyze past subscriber growth and subscriber bandwidth utilization trends to predict saturation of interfaces and nodes. This allows you to base investment decisions such as regional upgrades or equipment replacements on real data rather than best guesses.
- Create and implement fair usage policies. Identify peak usage periods, congestion points, and heavy users. You can then enforce business policies — such as temporarily throttling down heavy users during peak times — to guarantee fair usage of the network.
- Learn what actually works. Identify bandwidth utilization patterns on nodes post-node split to understand how and in what scenarios node splits are effective.
- Define new processes based on real data. Create new metrics and priorities for node splits, upgrades, and policies based on subscriber concentration, subscriber type, and percentage of premium subscribers on the node.
- Pinpoint heavy users. Isolate key, repetitive network offenders that are costing you more than the revenue being provided.
For marketing and sales teams:
- Deliver the right product to the right customer. Identify subscribers who should upgrade plans according to bandwidth utilization history, traffic hours, traffic type (Internet, voice, VoD), and subscriber type (business or residential).
- Understand what new products you should be offering. Identify when a customer should be offered different media access or a different package. A small or medium-sized business using too much bandwidth under its existing plan may have different needs than a residential subscriber. For example, a small hotel using cable Internet to distribute Internet access to guests may consume a large percentage of the bandwidth available in the node during peak hours. This indicates that the operator should consider offering a new service, such as fiber, to remove the strain caused in the node and reduce the impact on quality of experience (QoE) for other subscribers.
- Locate premium customers. Identify geographic areas and nodes that are saturated at specific times of days, and the percentage of key business and premium customers impacted by the congestion. These may be areas where the operator needs to act fast to shift resources and make quick investments to retain premium customers by providing high QoE.
- Help subscribers stay on track. Provide customers with notifications and alerts that make them aware of their consumption habits and help reduce excessive traffic.
- Create user portals for self-service. Drive users to portals that allow subscribers to track usage and also provide the opportunity to increase revenue by offering plan upgrades and/or quota increases to targeted customers.
As you can see, there is a world beyond usage-based billing for IPDR. The main challenge for cable providers is to store, aggregate, compute, and normalize information that can easily accumulate in terabytes of data and billions of rows of a database.
Avoid a complicated analytics problem with a solution that makes sense of big data by automatically filtering IPDR statistics and tracks trends to provide valuable information on network status, performance, and bottlenecks. Learn more about how Incognito Software Systems can help you with bandwidth management with an IPDR-based solution here.