I have occasionally been asked, "Why bother doing that when we can just call the cabling company?" My answer is pretty simple: variety, experience, and troubleshooting knowledge. I enjoy keeping my hands-on skills sharp, and sometimes a straightforward cabling job can be completed while waiting for an outside contractor would take days. On one installation, we needed to pull and terminate three Ethernet cables. They were short, straightforward runs through existing cable trays, roughly 20 feet each. The alternative was calling a cabling contractor and potentially waiting two or three days. In that particular situation, having the tools and experience meant we could complete the work ourselves and keep the project moving. That doesn't mean every cabling job should be DIY—it means you should understand enough about the physical layer to know what the job actually involves.
There is also a troubleshooting benefit that is easy to overlook. When you have physically installed and terminated Ethernet cable, you develop a better appreciation for the things that can go wrong. Poor terminations, excessive cable bends, damaged cable, improper strain relief, incorrect pair arrangement, bad connectors, and questionable installation practices can all create problems that eventually show up as network symptoms. A network engineer who has only ever seen the network from a management interface may look at a switch port and see "up" or "down." Someone with hands-on cabling experience is more likely to start asking what happened between the switch and the device. That physical-layer knowledge can be extremely useful when troubleshooting intermittent connectivity, link-speed problems, packet errors, or a connection that mysteriously works when somebody wiggles the cable.
Of course, there is a big difference between having basic cabling skills and experimenting on a production network. If you have never terminated an Ethernet cable before, don't make your first attempt on the company's critical uplink. Buy some inexpensive cable, connectors, and a basic crimping tool and practice somewhere that a bad termination won't bring down half the office. Testing your finished cables is just as important as making them. A cable that clicks into the RJ45 jack is not automatically a good cable. Depending on the situation, proper cable testing can verify continuity, wire mapping, length, and other characteristics that may not be obvious simply by looking at the connector. The goal isn't to replace professional cabling contractors; it is to understand the work well enough to recognize good and bad installations.
I also learned early that having the right tools can make a huge difference. On one construction project, I arrived onsite with my own tools, a hard hat, and the required safety credentials while some of the other consultants had been told to stay out of certain areas. That experience reinforced something I still believe today: certifications and theoretical knowledge are valuable, but they don't replace practical experience. You don't need to become a professional cable installer to benefit from understanding Ethernet cabling. Even learning how twisted-pair Ethernet is constructed, how connectors are terminated, how cable categories differ, and how to test a completed run gives you another layer of knowledge when you're troubleshooting a network.
Networking has always been a combination of software, hardware, protocols, and physical infrastructure, so I think network engineers benefit from understanding all four. You don't have to terminate 120 pairs of cable or climb an 80-foot tower to prove anything. Sometimes it is simply about being willing to get out from behind the keyboard and understand what is physically connecting everything together. The next time someone says, "It's probably the network," take a look at the network—but don't forget to look at the cable connecting the device to it. After all, packets may be digital, but they still need somewhere to travel.
