Wi-Fi 7 for Offices: What It Actually Changes About Your Cabling

Wi-Fi 7 access points are showing up in every vendor quote right now, and the pitch is always about the wireless side: wider 320 MHz channels in the 6 GHz band, higher-density modulation, and the ability for a client to use two bands at once. All of that is real. What the brochure tends to skip is that most of the gain lands on the wired side of the access point, which is where a lot of offices are about to hit a wall.

We get called into Chicagoland offices after the APs are already on the ceiling and the wireless somehow feels no faster than the gear it replaced. Nine times out of ten the radios are fine. The cable, the switch port, or the power budget behind them is not. Here is what actually has to change.

The short version

• Wi-Fi 7 APs can move more traffic than a 1 Gbps switch port can carry, so the uplink becomes the bottleneck.
• Enterprise APs increasingly want a multi-gig port (2.5G, 5G or 10G) rather than plain gigabit.
• They also draw more power, which pushes you toward 802.3bt (PoE++) switches instead of 802.3at.
• Cat6A is the sensible cable to pull for AP runs. Cat5e will physically link, but it caps what you can do later.
• Your comm room and switch power budget usually need attention before the ceiling does.

Why gigabit to the AP stops being enough

A single gigabit Ethernet port carries about 1 Gbps, full stop. Wi-Fi 6 access points could occasionally brush up against that in a busy conference room, but in practice most offices never noticed. Wi-Fi 7 changes the math. With 320 MHz channels available in 6 GHz and multi-link operation letting a client run on two bands simultaneously, the aggregate throughput an AP can push in real client conditions can exceed what one gigabit uplink will let through.

That does not mean every AP needs 10 gig. It means the port behind the AP is now a design decision instead of an afterthought. In a typical office with normal user density, a 2.5 Gbps uplink covers a Wi-Fi 7 AP comfortably. In high-density areas such as training rooms, auditoriums, cafeterias or open floors with a hundred-plus devices, that is where you want 5G or 10G capability, or at least the cable to support it when you need it.

What to ask your vendor

• What port speed does this specific AP model have? Some Wi-Fi 7 models still ship with a single 1G port. Some have 2.5G. Some have 10G.
• Does it support link aggregation across two ports, and do you actually want to burn two cable runs and two switch ports per AP?
• What is the PoE class, and what does the AP do when it only gets 802.3at power? Many will boot but disable a radio or drop transmit power.

Power: the part that quietly breaks the budget

Wi-Fi 7 APs are power hungry compared to what they replace. Where an older AP might have been happy on 802.3af (about 15W at the device) and most Wi-Fi 6 gear ran fine on 802.3at PoE+ (about 25W), enterprise Wi-Fi 7 units commonly land in a higher class and want 802.3bt, sometimes called PoE++.

Two things bite people here. First, an AP that is underpowered often does not fail loudly. It comes up, associates clients, and silently runs in a reduced mode with a radio disabled or transmit power turned down, so the coverage survey you paid for no longer matches reality. Second, per-port capability is not the same as switch power budget. A 48-port switch advertising 802.3bt on every port almost never has the internal power supply to deliver maximum power on all 48 at once. You have to add up the actual draw of everything on that switch, cameras and phones included, and leave headroom.

This is exactly the same trap we wrote about in our post on PoE. The switch spec sheet has two numbers, and the one that matters is the total power budget, not the per-port maximum.

What cable to pull

For new access point runs, Cat6A is the right default. It supports 10GBASE-T across a full 100-meter channel, it handles 2.5G and 5G without drama, and it has better heat and alien-crosstalk behavior in the bundles that PoE++ power levels create. Pulling Cat5e to a ceiling AP today is choosing to redo the run in a few years.

Cat6 sits in an awkward middle. It will do 10G, but only over shorter distances, so whether it works depends on your specific run lengths. If you are already in the ceiling and already paying for labor, the incremental cost of Cat6A over Cat6 is small compared to the cost of pulling cable twice. Labor is the expensive part of any structured cabling job, not the box of cable.

Bundle heat is a real consideration

Higher PoE power levels mean more heat inside large cable bundles, and heat raises insertion loss, which is exactly what kills marginal links. The practical answers are unglamorous: don’t overstuff bundles, keep them out of hot plenum spaces where you can, use cable rated for the power levels you’re actually delivering, and certify the finished runs so you know what you have instead of guessing.

The part almost nobody plans for: the closet

Adding multi-gig, higher-power switches to support Wi-Fi 7 tends to expose whatever was already marginal in the comm room. We routinely find:

• Not enough uplink capacity from the closet back to the core. If every closet is now aggregating multi-gig APs, a single 1G uplink to the core is a hard ceiling on the whole floor.
• No spare rack space or power for the new switch, so it ends up stacked on a shelf with a power strip.
• Cooling that was borderline with the old gear and is not borderline anymore. PoE++ switches throw off real heat.
• No documentation, so nobody knows which port feeds which AP when something goes wrong at 4pm on a Friday.

In multi-floor buildings, this is usually where a fiber optic installation conversation starts. Copper risers between floors run out of headroom quickly once every closet is pushing multi-gig traffic; fiber backbone between the closets and the core is the straightforward fix.

Do you actually need Wi-Fi 7 right now?

Being blunt about it: for a lot of Midwest offices, the honest answer is not yet, and there is no shame in that. If your users are on Teams calls, email, and browser-based line-of-business apps, Wi-Fi 6 or 6E is not what is slowing them down. Wi-Fi 7 earns its keep where density is genuinely high, where you have modern client devices that can actually use 6 GHz, or where you are running bandwidth-heavy work such as video production or large file transfers over wireless.

What we would not do is skip the cabling. If you are already opening ceilings, already doing a buildout, or already replacing switches, pull Cat6A and put in switches with multi-gig and 802.3bt capability, even if the APs you install today are Wi-Fi 6. The cable and the closet are the expensive, disruptive parts. The AP on the end of the run is the cheap, easy part to swap later.

A practical sequence for an upgrade

• Audit what you have: cable type and length on each AP run, switch port speeds, PoE standard, and the switch’s total power budget.
• Decide density: which spaces genuinely need multi-gig uplinks and which are fine at 1G for now.
• Fix the closet and the backbone first: uplinks, power, cooling, rack space.
• Pull or replace cable where needed, and certify the runs with a tester so you have a report, not an opinion.
• Then hang APs, and run a post-install survey to confirm the coverage you designed is the coverage you got.

That order matters. Every project we get called in to rescue did it backwards.

Bottom line

Wi-Fi 7 is less a wireless upgrade than a wired one. The radios are the easy part. The switch port, the PoE budget, the cable in the ceiling and the backbone out of the comm room are what determine whether you see any of the performance you paid for. Get those right and the wireless takes care of itself. Get them wrong and you have expensive access points running at the speed of your weakest link.

If you are weighing a Wi-Fi 7 refresh, a new office buildout, or a network installation project anywhere in Chicagoland or the Midwest, we’re happy to walk the space and tell you honestly what needs to change and what doesn’t.