Cable assemblies, meaning the electrical conductor and associated connectors at one or both ends, are components with a bipolar life. On one side, they're simple copper links and easy to check out for basic continuity; all it takes is an ohmmeter function on a multimeter. On the other side, we expect more and more of these interconnects when handling power and especially when conveying data. After all, who would have predicted years ago that a thin, twisted-pair cable would be handling megabit-per-second data rates?
That level of performance is what we routinely get from a properly designed and constructed cable rated for Cat5 or Cat6 standards, for example. Even coaxial cables have their imperatives, with rating to hundreds of megaHertz and tens of gigaHertz. Yet when a system's performance is not what it should be, we usually don't think about the cable, since it was marked and marketed to fulfill the requisite specifications.
Reality is that counterfeit, substandard, or poorly installed cables are out there, shocking as it may seem. After all, why bother to produce fake ICs when it is so much easier to make or take an inadequate cable and stamp a certification on it? Even better, the substandard assembly will work somewhat, and link problems will usually be blamed on excessive external noise, inadequate noise-margin, timing skew and jitter, marginal line driver/receivers, and many other possibilities.
That's why I like to look through the publications such as Cabling Installation & Maintenance, which target the folks in the field (literally, as well as figuratively). These people are the ones who have to deal with the consequences of designs that worked in prototype and even pilot installations, yet somehow are having issues during or after wider deployment; these problems must be often be worked on under adverse circumstances and in challenging locales. By seeing how field personnel handle these problems, you can often find solutions which work more effectively