I get asked about WiFi 6 in roughly half my callouts these days. Usually it's framed as a question, but it's really a hope: "If I buy the new router, will my broadband stop being rubbish?" The short answer is no — and the longer answer is the reason I'm writing this. WiFi 6 explained properly, by an engineer who fits it for a living, with the marketing scraped off and the actual physics left in.

WiFi 6 is the consumer label for 802.11ax, ratified in 2019 and now the default standard on basically every router sold in the UK. It's a real upgrade over 802.11ac (WiFi 5), but it's an upgrade in places most people aren't expecting. It is not, in any meaningful sense, "faster internet." Your line speed from Openreach, Virgin Media or CityFibre is your line speed. A new router can't conjure megabits out of the copper, and on FTTP the optical network terminal is already handing you everything the fibre is willing to give. What WiFi 6 changes is what happens after the router — how cleanly your radio environment handles a house full of devices all wanting to talk at once.

What 802.11ax actually does differently

WiFi 5 (802.11ac) was designed in an era where a household had a laptop, a phone, maybe a tablet and a smart TV. Five or six clients, all relatively chatty. WiFi 6 was designed for the household I walk into today: thirty-plus connected things, half of them tiny IoT sensors that wake up every few seconds to tell their cloud they're still alive. The standard added four headline features to cope with that reality, and three of them are genuinely useful.

OFDMA — sharing the airtime

OFDMA (Orthogonal Frequency-Division Multiple Access) is the big one. On WiFi 5, when your phone wanted to send a one-packet "I'm still here" message to the router, it took the whole channel for that one packet, then handed it back. Imagine a single-lane road where every car has to use the entire width of the road regardless of size. OFDMA chops the channel into smaller sub-channels (the spec calls them Resource Units) so multiple devices can transmit small bits of traffic at the same time. For a flat with three devices it's marginal. For a house with twenty-five doorbells, hubs, bulbs, plugs, thermostats and watches, it's the difference between everything working and everything stuttering at 7pm.

MU-MIMO upgraded — both directions now

MU-MIMO (Multi-User Multiple Input Multiple Output) existed in WiFi 5 but only on the downlink — router-to-device. WiFi 6 made it bidirectional, so your devices can also transmit simultaneously back to the router. On a Zoom call where four people in the house are all uploading video at the same time, that genuinely helps. On a single Netflix stream it does nothing.

BSS Color — telling the neighbours apart

This one matters enormously in London and almost nowhere else in the country. Walk down any street in Hackney or Battersea and your phone will see twenty to forty other WiFi networks. Before BSS Color, your router would politely back off whenever it heard any other 802.11 traffic on its channel, even if that traffic was your neighbour's router three flats away. BSS Color tags each network with a number; if the signal is coloured "not me," the router can keep transmitting through it. On a Saturday evening in a converted Victorian terrace, BSS Color is the closest thing to a magic wand I've seen ship in a consumer spec.

TWT — Target Wake Time

TWT lets battery-powered devices negotiate scheduled "you can sleep for the next X seconds" windows with the router. For your laptop it's irrelevant — it's plugged in. For your smart smoke alarm, soil moisture sensor or that battery-powered Hive thermostat, it can roughly double battery life. Worth knowing, not worth buying a new router for.

WiFi 6 vs WiFi 6E — the 6 GHz band

WiFi 6E is WiFi 6 plus access to the 6 GHz band, which Ofcom opened up for unlicensed use in the UK in 2020. This is the part of the spec that does feel like an upgrade in a busy environment, because 6 GHz is — for now — empty. Every legacy device you own (your 2017 laptop, your smart bulbs, your guest's phone running WiFi 5) is locked out by definition. Only WiFi 6E and WiFi 7 clients can join. That means no congestion from older traffic, far wider channels available (160 MHz is realistically usable), and very short range, which in dense London flats is a feature, not a bug — you and your neighbour stop interfering with each other.

The catch: 6 GHz signal attenuates faster through walls than 5 GHz, which attenuates faster than 2.4 GHz. So a WiFi 6E router in the hallway will reach further on its 2.4 GHz radio than on its 5 GHz radio than on its 6 GHz radio. If you want 6E coverage across a three-bedroom Edwardian semi, you need more than one access point. There's no way around the physics — see what actually blocks your WiFi signal for the gory detail.

And WiFi 7? Worth waiting for?

WiFi 7 (802.11be) is the next standard, ratified late 2024 and starting to appear on premium routers now. The headline feature is Multi-Link Operation — a single device can use 5 GHz and 6 GHz radios at the same time, picking whichever is cleanest packet-by-packet. There are also wider channels (320 MHz, theoretically) and higher-order modulation. In a dense London home with WiFi 7 clients, it'll matter. Today, most of your devices are still WiFi 5 or WiFi 6. I don't tell anyone to wait specifically for WiFi 7 unless they're planning a new build with structured cabling and a network refresh in two years' time anyway.

The bit nobody mentions: your line is still the bottleneck

This is the misconception I have to gently correct on at least one job a week. Buying a new router does not make your broadband faster. If you have a 67 Mbps FTTC line from Openreach, the absolute ceiling is 67 Mbps regardless of whether your router is WiFi 5, 6, 6E or 7. Even on a 900 Mbps full-fibre line, the modem-side of the router is the gating factor, not the WiFi side. The radio is only the bottleneck when the line itself is fast enough to overrun a previous-generation WiFi standard at close range — and even then, only on a single device close to the router.

Where WiFi 6 actually wins is at the aggregate level. Total throughput across all the devices in the house. Median latency at the busy hour. Consistency. Those are the numbers that change. Peak speed on a single iPhone two metres from the router barely moves. If you've never seen one of these distinctions on a real speed test, that's because the apps test peak speed on a single device — the worst possible metric for showing what WiFi 6 changes.

Channel widths, DFS and the bandwidth trade-off

Quick technical aside that matters in dense areas. WiFi 6 can use channel widths of 20, 40, 80 or 160 MHz on 5 GHz, and likewise on 6 GHz. Wider channels equal higher peak speeds — but only if the wide channel is clean. In a London flat with twenty neighbours, an 80 MHz channel is overlapping with several other networks; a 40 MHz channel is overlapping with fewer; a 20 MHz channel is often the only one that holds steady. Most routers default to 80 MHz on 5 GHz and "automatic" channel selection, which is fine in a detached house in the suburbs and often counterproductive in a converted Victorian. On 5 GHz I also enable DFS channels (the higher 5 GHz channel block, 52–144), which most neighbours' routers ignore — instant uncrowded spectrum, with the small overhead that the router has to listen for radar and may switch channels occasionally. For more on this kind of tuning, see WiFi in London apartments and conversion flats.

When WiFi 6 pays off: three scenarios from this year

Scenario one — the Wandsworth family with twelve hungry devices

I went out to a four-bedroom semi off Trinity Road in March. The brief was "WiFi keeps dying at dinner time." Two parents working from home, both on video calls most of the day. Three teenage kids, two of them serious gamers, the third a streamer with three monitors. Add four streaming TVs around the house, a Ring doorbell, a Tado heating system and the usual smart-bulb sprawl. The router was a WiFi 5 device from the ISP, sat in the hall on top of the master socket. At quiet times it tested 200 Mbps in the living room — perfectly fine. At 7pm it tested 12 Mbps and dropped packets every few seconds.

That household is the textbook WiFi 6 case. We swapped to a WiFi 6 ceiling-mounted access point on the ground floor, ran a Cat6 from the hall to the upstairs landing, and put a second WiFi 6 access point there. Both are powered by PoE from a small switch in the cupboard. Same 500 Mbps Virgin line, no change to the broadband at all. Saturday evening test: 380 Mbps anywhere in the house, no packet loss, no buffering on the streamer's three monitors. The difference was OFDMA and BSS Color doing their job in a high-density environment, plus the wired backhaul, plus two well-placed transmit points instead of one. A single mesh node added to the old WiFi 5 router would not have solved this — see mesh WiFi vs wired access points.

Scenario two — the single-occupant studio in Pimlico

Same week, near Tachbrook Street. One-bedroom flat. One occupant. A laptop, a phone, a Sonos, a smart TV. Existing WiFi 5 router from the ISP, sat on the bookshelf in the living room. The customer had read a tech blog and was about to spend a few hundred pounds on a WiFi 6E mesh system. I tested the existing kit: 380 Mbps on speedtest standing in the kitchen, 350 in the bedroom, no congestion, no contention. The line was a 500 Mbps full fibre. I told her to keep her money. The new router would be measurably faster on a controlled bench test and indistinguishable in real life. That's the honest version of the WiFi 6 question for half the flats in zone 1. If the existing kit works, the upgrade is cosmetic.

Scenario three — the architects' practice in Clerkenwell

Mid-sized architectural practice, mezzanine office on St John Street. Twenty-two staff, all on workstations with two 4K monitors each, doing live model collaboration over the LAN. The design floor has a row of plotters and a high-density of laptops on the studio benches during crits. They had a sensible business-grade WiFi 5 setup that had been fine for years and was now wobbling under load — the office had grown, the projects had grown, the BIM models had grown. The fix was three WiFi 6 access points across the open floor with a fourth WiFi 6E access point over the design bench area. Cat6 backhaul to a 24-port PoE switch. The 6E AP gave the senior architects' new MacBooks a clean 6 GHz channel with nothing else on it — measured ~900 Mbps to a desk-side server during peak. The WiFi 6 APs gave everyone else BSS Color and OFDMA on the 5 GHz band. Latency to the central file server dropped from 18ms median to under 3ms. The practice didn't have a slow internet line — they had a saturated radio environment, and WiFi 6/6E was the right tool. We also looked at the broadband side at the same time but the line itself was fine.

How to know which scenario is yours

You don't need a survey to make a sensible call on this. Run through the list:

  • How many devices are actively transmitting in the house at the busy hour? If it's under ten, WiFi 6 will give you almost nothing.
  • Is your line genuinely fast — say, 300 Mbps or more — and are you not seeing close to that within a few metres of the router? Then your bottleneck is the WiFi, and an upgrade might help. Or you might just need a wire.
  • Do you have a dead spot? A new router won't fix a dead spot. Walls fix dead spots, by being moved out of the way; access points fix dead spots, by being in the room with the dead spot. See WiFi installation.
  • Are you in a dense flat conversion with neighbours either side, above and below? Then BSS Color is genuinely worth having, and 6E (if you have any 6E-capable devices) is even better.
  • Do you have any 6E or WiFi 7 clients at all? If not, the 6 GHz radio on a 6E router is a future-proofing purchase, not a today purchase.

And the perennial question: should you trust the ISP-supplied router? In most cases the WiFi 6 hub from the major providers is fine on the radio side; it's just placed badly (in the hall, behind the front door, next to the gas meter) and configured for the average household, not for yours. Sometimes the better fix is a single well-placed access point on a Cat6 run from the existing router, not a wholesale replacement. The master socket almost never has to be the router location — it's just the cheapest place for the ISP to default to.

The honest engineer's checklist before buying a new router

  1. Run a speed test wired into the router with a known-good ethernet cable. That's your line speed, the absolute ceiling. If it's already disappointing, replacing WiFi won't help.
  2. Run the same speed test on WiFi, two metres from the router, on a recent phone or laptop. If that number is close to the wired number, your WiFi is not the bottleneck.
  3. Now do the same test in the worst room in the house. The drop-off tells you whether you have a coverage problem (more APs) or a contention problem (better standard, more APs).
  4. Count your active devices. If it's under fifteen, a new standard helps only at the margins.
  5. Check your immediate neighbours' WiFi count. In Soho or Shoreditch this number alone justifies WiFi 6 in the existing router.

And whatever you do, don't buy any kind of plug-in extender or repeater off the back of a Google search. They almost always make things worse — see why WiFi boosters and extenders almost always disappoint. The right answer is usually a wire and a properly-placed access point, not a second radio trying to repeat the first one's signal in mid-air.

Bottom line

WiFi 6 is real, it's good, and it matters most when your home or office has many devices, many neighbours, or both. WiFi 6E adds a genuinely-uncongested 6 GHz band that's worth having in dense London environments if you've got compatible clients. WiFi 7 is on the way but isn't urgent. None of these standards make your broadband line faster — that's set by Openreach, Virgin or whoever else is feeding you, and is fixed at the master socket. The right WiFi upgrade is the one that matches the house, the line and the devices in front of it. Anything else is shopping for the sake of shopping.

If you're in two minds, a proper coverage survey gives you the answer in an afternoon and saves the cost of the wrong router. I've covered what that involves on the heat-map survey article. Or just pick up the phone — no charge for an honest opinion.

Ring 020 3633 1131. Honest advice, freely given.