Executive Summary
After decades of successful growth, cable MSO operators are today facing a crucial 
challeng—how to continue to grow profitably, and at the same time efficiently compete with other broadband vendors, predominantly using FTTx/PON. For many cable MSOs, the way forward is migration to FTTx—with or without extending the lifetime of their DOCSIS-based infrastructure as an intermediate step.

Summary Bullets 
—Cable MSO networks are fully feeling the strain of the unprecedented service demand generated by COVID-19 pandemic impact on work, education, and media consumption habits. Most are investing into upgrades of their hybrid fiber-coax plant, but many are looking at complete overhaul of their infrastructure.
—Migration to full fiber infrastructure is the final goal of practically all cable MSOs, and many are building FTTx infrastructure in parallel with their HFC plant or choosing a "cap and grow" migration strategy. The final goal of this transformation will in any case be purely FTTx infrastructure; operators need to consider their migration options carefully and examine ways to optimize their networks deployment to ensure smooth and cost-efficient transition.

Cable MSO infrastructures are one of the main foundations of broadband service offerings in many countries globally. Thanks to physical characteristics of coaxial cable, MSO operators have extended the usefulness of their networks, historically associated with cable TV distribution, to broadband connectivity services. Coaxial cable infrastructure with the latest upgrades to DOCSIS 3.1 has transformed into a hybrid fiber-coax plant (HFC) capable of both distributing video and providing broadband services. 
However, changing customer behavior, increasing demand for high bandwidth services, and the growth of cloud traffic—further exacerbated by the COVID-19 pandemic—have put several deficiencies of HFC into spotlight: 
—Scaling and overall capacity: Most cable networks today have been built with a relatively stable bandwidth demand in mind. The changing usage patterns and increased demand associated with COVID-19 have quickly shown that HFC infrastructure needs constant upgrades to cope with increasing demand. Due to highly concentrated nature of legacy HFC infrastructures, these upgrades are costly and relatively complicated.
—Upload/download disbalance: Although coaxial cable is significantly better medium for broadband than copper twisted pair, it still has capacity constraints. Due to this, and tendency of residential users in the past to use broadband mostly with download-heavy applications, cable data transfer (standardized under DOCSIS specifications) is highly asymmetrical. DOCSIS 3.1, for example, provides maximum of 10 Gbps download and 1 Gbps upload per connection, but in practice customer services are provided with download speeds over 20 times faster than upload. These upload constraints become quickly evident in two-way video communication, video creation and uploads, game streaming, or the use of productivity cloud-based applications.
—Sunset of cable TV: Cable operators' investment has traditionally been equally aimed at increasing their data transfer capabilities, but also in maintaining and improving their TV distribution capabilities. However, in last several years, consumer viewing habits have changed—most residential users consume video content through streaming platform, effectively putting more strain on data part of the network; at the same time, the number of cable TV subscriptions keeps dwindling. Cable operators thus are looking for ways to scale their data transfer capabilities, and at the same time de-emphasize investment associated with video distribution.
—Beyond residential use cases: Cable MSOs have usually relied on residential customers for most of their revenue. The onset of highly symmetrical faster-than-1 Gbps services in FTTx combined with SD-WAN has shown that commercial broadband can be successfully used for SME and business connectivity. Cable operators today are mostly forgoing this opportunity because their services are highly asymmetrical and would impede upload-heavy applications.

This set of challenges has shaped the ways cable MSOs transform their networks today. Cable MSOs and their suppliers have developed and implemented several technologies that allow cable operators to make some improvement and at least partially address the challenges. However, most cable MSOs need to change faster, and the cable industry's transformation is accelerating. Within this context, MSOs need to consider the following: 
—Both major migration paths lead to FTTx: A broad consensus among cable MSOs is that their networks will—in time—become full-fiber FTTx. The latest cable architectures, outlined under DOCSIS 4.0 are designed with a view that they will be a temporary, and very likely the last stage of traditional cable access technology. Upcoming DOCSIS 4.0 is still asymmetrical (10 Gbps/6 Gbps) and is inferior to currently mainstream FTTx technology XGS-PON, which provides symmetrical 10 Gbps connections. Some operators have thus chosen to move directly to FTTx from their current DOCSIS 3.1 cable networks, while others view DOCSIS 4.0 as a transitional stage. In essence, both major migration paths envision full fiber infrastructures replacing HFC in the next several years.
—Passive infrastructure is the top concern: As with greenfield FTTx, the main challenge in cable MSOs fiber transformation is the investment into passive network plant, in particular the optical distribution network (ODN). According to networking equipment vendors, the investment into ODN can make up the majority of overall FTTx deployment investment. Cable MSOs can, however, significantly lower this part of their network spending, by utilizing existing network ducts to the degree possible in combination with pre-connectorized ODN elements that allow MSOs to lower OPEX and accelerate deployments.
—Market dynamics play a big role: As with other parts of the marketplace, the wider broadband market dynamics can serve as a major moderating factor in cable MSO transformation plans. The paramount among these factors is FTTx competition, which is increasing significantly in practically all major cable markets except for the United States, and access to public and private financial support, which is overwhelmingly directed at fiber infrastructures in markets like the UK, the EU, or India. Finally, the vendor landscape of FTTx suppliers is wider and more vibrant than dedicated cable solutions market, enabling faster innovation, and much higher shipment volumes which, in turn, drive lower prices and faster price decreases.
These trends thus outline the future transformation paths that most cable MSOs need to undertake. In both major migration paths, the future MSO infrastructures will need to be rebuilt as FTTx infrastructures. The transformation will either happen through complete overbuilds, with operators deploying FTTx in parallel with the existing HFC, and then migrating customers to FTTx, or through "cap and grow" deployments, where operators deploy FTTx in newly covered and highly competitive areas, while upgrading transport and moving fiber closer to customers in areas still served by cable. In both cases, MSOs need to seek ways to perform these upgrades efficiently, taking into account the technological advancements in deployment techniques in solutions, and expanded opportunities for monetization that FTTx brings them.