In short: Narrowboat WiFi fails for reasons no dongle can fix: a steel hull is a Faraday cage, canal-side masts are sparse, and marina WiFi rarely reaches past the first pontoon. An inland marina that deploys one managed wireless network across its basin can deliver working internet to every mooring, including liveaboards who depend on it, and bill for it the same way it bills for electricity.
Key Takeaways
- A narrowboat is a steel box that blocks its own signal — the hull and cabin sides attenuate mobile signal so severely that connectivity has to be caught outside the boat and brought in, which is why dongles on the galley shelf disappoint and roof-mounted antennas become essential.
- Inland marinas sit in exactly the places mobile operators skip — rural junctions, valley pounds and deep cuttings are marginal business cases for public networks, so the marina that wants reliable coverage for its moorers generally has to provide it.
- Connectivity is the next berth-side utility — around 15,000 people live afloat on UK waterways and remote work has followed them aboard; a marina selling managed internet per mooring converts its biggest complaint into a recurring revenue line alongside electricity and pump-outs.
In a nutshell
Why is narrowboat WiFi so hard to get right?
Ask anyone who lives or works aboard on the UK canal network what their biggest daily frustration is, and the answer is rarely water, diesel or even winter condensation. It is signal. The boater's forums are full of the same questions, asked year after year: which network works best on the Grand Union, whether a booster is worth the money, how to take a Teams call from a marina berth without walking to the car park.
The problem has three layers, and it is worth being precise about them, because each layer defeats a different popular remedy.
Firstly, the boat itself. A narrowboat is a welded steel tube, and steel is very good at keeping radio waves out. A phone or 4G dongle sitting inside the cabin might see one bar where there are four bars on the towpath. No amount of shopping around between networks fixes this; the fix is an external antenna, mounted on the roof or a pole, with the router inside. Most boaters discover this after buying at least one product that promised otherwise.
Secondly, the location. The canal network was engineered to avoid gradients, which means it runs through cuttings, around contours and along valley floors, precisely the terrain that public mobile networks serve worst. An inland marina is typically a basin dug beside a rural stretch of canal, several miles from the nearest town, with a customer base of a few hundred moorers. For a national operator planning macro coverage, that is not a site that justifies a mast. Coverage at the marina is whatever spills over from infrastructure planned for somewhere else.
Thirdly, the marina's own WiFi. Most inland marinas have broadband to the office and a handful of access points on the facilities block. The signal reaches the chandlery, the laundry and perhaps the first row of moorings. It does not reach a steel boat on the far side of a 300-berth basin, and adding more consumer-grade access points along the pontoons trades one problem for another: outdoor cabling, corroding hardware, and a mesh that degrades every time the marina fills up for a bank holiday.
Who actually needs the connection?
It is tempting to file boat internet under leisure, something for streaming on a summer weekend. The reality on the inland network is more interesting, and more commercially significant for the marina.
The Canal & River Trust licenses roughly 35,000 boats across some 2,000 miles of canals and rivers, and estimates suggest around 15,000 people now live afloat, a population that has grown steadily since the pandemic. A significant share of those liveaboards hold winter moorings or permanent berths in marinas; Aquavista alone operates more than 30 inland marinas, and residential and long-term leisure moorings are the backbone of the inland marina business model.
For these customers the connection is not a nicety. Remote work has moved aboard: the boater taking video calls from the dinette is now common place. Children aboard need connectivity for schoolwork. Boat systems increasingly phone home, with battery monitors, bilge alarms, solar controllers and heating that owners expect to check from a phone. And the marina itself is quietly becoming a data operation: berth-side electricity metering, CCTV over the basin and car park, gate access control, and charge points as electric propulsion creeps onto the inland network.
There is also the seasonal dimension. A marina that is sleepy in February hosts hundreds of visiting boats in August, each arriving with phones, tablets and expectations. The network has to handle the peak without falling over, which is exactly the load profile that a thin scatter of access points handles worst.
What do boaters do today, and why does it fall short?
The current state of narrowboat internet is a study in individual improvisation. A typical marina pontoon carries 4G routers with roof antennas, signal boosters of varying legality and effectiveness, unlimited SIM plans from whichever operator seemed least bad locally, and increasingly a Starlink dish on the semi-trad stern or cabin roof.
Each of these works to a degree, and we should be honest about that: a well-installed 4G router with a decent external antenna, on the right network for that stretch of canal, gives many boaters a serviceable connection. Starlink gives an excellent one, at a price, and with a power draw of some 30 to 75 watts that a boat on shore power can shrug off but a boat on batteries has to plan around.
The shortcomings are structural rather than individual. Every boat pays separately, typically £15 to £40 a month plus £100 to £600 of hardware, for a connection whose quality the boater cannot control. When the local mast is congested or the marina sits in a coverage hole, no amount of spend on board fixes it. The marina, meanwhile, captures none of this value and all of the complaints: poor connectivity is now a standard reason given for leaving one marina for another, and berth reviews mention WiFi with the same frequency as showers and parking.
What does a proper fix look like?
The pattern we deploy for coastal marinas applies directly inland, with some differences worth noting. One private cellular network, run from a small mast or building-mounted antennas at the marina, covers the entire basin: every mooring, the hardstanding, the wharf, the car park and the entrance. Boats connect either directly, where a phone or router supports the band, or through a small outdoor unit on the boat that catches the private signal outside the steel and serves WiFi inside, which resolves the Faraday-cage problem cleanly.
Why cellular rather than more WiFi? Fundamentally, because the physics and the economics both point that way across a site this shape. A basin of ten acres with several hundred metal boats is a hostile radio environment for WiFi, which was designed for rooms; it is a routine one for cellular, which was designed for exactly this kind of open, obstructed, many-client coverage. One radio point replaces dozens of pontoon access points, along with the armoured cable runs, the corroded PoE injectors and the annual failure rate that comes with mounting consumer electronics a metre above canal water.
The same network then carries the marina's own operations. Berth-side metering stops being a quarterly walk with a clipboard. CCTV covers the basin and the boatyard without trenching cable along every pontoon, which matters at inland sites where security expectations are rising as boat values do. Charge points for electric boats and cars report over the same layer. None of this requires a second infrastructure project; it is capacity on the network the marina already built for its moorers.
We should also be candid about what private 5G does not fix. It does not help the boater once they cast off and head up the flight; out on the cut they are back on public coverage or satellite, and a marina should not pretend otherwise. What it fixes is the place where boats spend most of their time, which for marina-based and liveaboard craft is the mooring.
How does the marina make money from it?
The commercial logic mirrors electricity. Marinas do not give power away; they meter it per berth and the moorer pays for what they use. Connectivity can work the same way: a basic tier included in the mooring fee, paid tiers for liveaboards and remote workers who need guaranteed bandwidth, and day passes for visiting boats in season.
The arithmetic is straightforward to sketch. A 250-berth inland marina with 60 liveaboard or long-term berths, selling a £20 per month working-aboard tier to half of the liveaboards and seasonal passes over the summer, generates a five-figure annual revenue line from customers who are currently spending that money, with more friction and worse results, on individual SIMs and dishes. Against that sits the cost of a managed network sized for a marina rather than a container port, which is precisely the gap in the market we set out to serve; the payback period is typically a few seasons rather than a decade.
There is a softer return too, and marina operators tell us it matters more than the direct revenue: retention. Moorings are a competitive market on popular stretches of the network, and a marina that can honestly advertise working internet at every berth has a differentiator that costs its neighbours customers.
Where should an inland marina start?
Not with hardware. The first step is a coverage and demand survey: what public signal actually reaches the basin (measured, not assumed), how many moorers are liveaboard or long-term, what the office and yard need from the same network, and where a mast or antenna can go within the marina's planning envelope. Inland sites often sit near conservation areas and the survey should treat that constraint seriously from the outset rather than discovering it at the planning stage.
From there the build is modest by telecoms standards: shared-access spectrum licensed from Ofcom for the site, one or two radio points, a core that can run locally or in the cloud, and boat-side units for moorers who take a paid tier. The marina ends up owning a utility, not running a telecoms company; the management, monitoring and upgrades sit with the network provider.
Ultimately however, the argument is the one every marina already understands from its own history. Electricity on the pontoons was once a luxury; then it was a differentiator; now no moorer would consider a berth without it. Connectivity is travelling the same road, and on the inland network, where the boats are steel and the masts are far away, the marina that provides it will be the marina that keeps its moorers. We think the bollard was just the start.