The Dish on the Pushpit: What Starlink for Boats Means for the Marina It Is Moored In

Walk any UK marina on a Saturday in August and count the dishes. Five years ago the roof of a 12-metre yacht carried a VHF whip, a radar dome and possibly a 4G aerial bought from a chandlery. Now it carries a Starlink terminal, usually a Mini strapped to the pushpit rail or a standard dish flat on the coachroof, and the owner will tell you it changed their season. They are right. Starlink for boats works. What a marina operator should take from it is harder, because several hundred berth-holders quietly buying their own connectivity says something about what the site was not providing, and it leaves the marina's own systems exactly where they were.

Does Starlink work on a boat in a UK marina?

Yes, with caveats that get worse the more crowded the basin. A Starlink terminal on open water is close to faultless: latency around 25 to 60 milliseconds, download speeds well into the tens of megabits, and coverage across the whole of UK and Irish waters including the far side of the Minch, where no terrestrial network has ever reached. For a boat on passage, at anchor in a loch, or on a swinging mooring off a village with one overloaded mast, it is the only thing that works properly, and we say that as a company that builds terrestrial networks for a living.

Inside a marina the picture changes. The dish needs a wide, clear view of the sky, roughly a hundred degrees of it, and a masted yacht in a packed basin does not have that. The mast itself, the boom, the backstay, a radar arch and a sprayhood all sit in the way, and so do the masts of the four boats on the neighbouring fingers, which are typically three metres away and fifteen metres tall. Starlink's own app will show the obstruction map, and on a mid-basin berth it is usually a mess. The terminal copes by reconnecting to a different satellite, which the owner experiences as a video call that freezes for two seconds every couple of minutes rather than a total failure. It is tolerable for streaming and irritating for work.

Then there is contention, which is the same problem marinas already have with public 4G and 5G, arriving by a different route. Starlink allocates capacity by geographic cell, and a cell covers a wide area rather than a single basin. On the Solent, in the Clyde, or along the north Norfolk coast in high season, every terminal afloat in that cell is sharing the same beam capacity, alongside every caravan, farm and rural home on the land side. Performance on a wet Tuesday in February is not the performance on the Friday of an August bank holiday, and the boats that need it most are all there at the same time.

What does Starlink for boats actually cost?

At time of writing the hardware sits at a few hundred pounds for a Mini or a standard kit, and the roaming plan most leisure boaters use runs at roughly £75 a month for unlimited data, with prioritised maritime plans several times that. The subscription is the part that matters, because it is the part that recurs. A berth-holder who keeps the plan live all year is spending around £900, which on many south coast berths is a meaningful fraction of the annual mooring fee.

Power is the second cost, and it is the easier one to overlook. The standard dish draws somewhere between 50 and 100 watts in normal use, the Mini roughly 20 to 40, and neither is free on a boat. On shore power that is pennies an hour and nobody notices until the marina moves to per-berth metering, at which point a dish left running through the winter becomes a line on the electricity bill. On batteries at anchor it is a genuine planning problem, which is why the Mini, with its native 12 to 48 volt input and no inverter required, has taken over from the standard dish among UK cruisers.

Multiply that across the estate and the number gets interesting from the operator's side. A 400-berth marina with fifteen per cent uptake has sixty subscriptions running at £75 a month, so around £54,000 a year of connectivity spend leaving the site. None of it touches the marina's P&L, and the marina carries the electricity draw and the aesthetic consequences without seeing any of the revenue.

Why does a dish struggle at the berth when it is fine at sea?

Because the things that make a marina a good place to keep a boat make it a bad place to see the sky. Boats are packed at three-metre centres, masts are tall, and the basin is often tucked under a harbour wall, a cliff or a row of apartments built specifically to overlook the water. A dish mounted low on a pushpit rail, which is where most of them end up because it is the only flat, reachable, lockable spot, is looking through the rigging of its own boat and its neighbours.

Rain is the other factor, and it is the one owners underestimate in the UK. Starlink's user links run in Ku band, which attenuates in heavy rain, and the effect compounds when the radome is wet. Nobody loses a session to drizzle, but a heavy autumn squall on the east coast will produce exactly the outage pattern boaters describe on the forums: fine all week, then dropping out for twenty minutes during the worst of it. Terrestrial cellular at 3.8 GHz, which is what a private marina network uses, is far less affected by rain over the few hundred metres between a pontoon and a mast on the harbour office roof.

For a boat that goes places, Starlink remains the right purchase. For a boat that spends 340 nights a year alongside, which describes the overwhelming majority of the UK fleet, it is an expensive way to get an imperfect signal.

What does Starlink not do for the marina?

Nothing at all. Every Starlink terminal in the basin belongs to a berth-holder and serves that boat. It carries no marina traffic. When the operator comes to connect the things that actually run the business, the satellite estate on the pontoons is of no help whatsoever.

The list of things needing connectivity on a modern marina is longer every year:

• Pontoon and perimeter CCTV, which needs continuous upload from fixed points along a finger, not from a boat that may leave on Friday.
• Smart shore power pedestals reporting per-berth consumption for billing, and increasingly managing load as electric boat charging arrives.
• Gate and barrier access control, ANPR on the car park and the boatyard, and fob readers at the head of each pontoon.
• Card terminals at the fuel berth and the visitor pontoon, which are always at the far end of the site where coverage is worst.
• Travel hoist and boatyard operations, lift scheduling, condition photographs at lift-out, and staff handhelds that need to work on the hardstanding as well as in the office.
• Berth occupancy sensing, so the duty team knows which visitor berths are genuinely free before a yacht calls up on channel 80.

Every one of those is a fixed or roaming device belonging to the marina, in a place where marina WiFi has historically stopped and where public mobile coverage is compromised by the same geography that made the harbour a harbour. Satellite terminals on the boats do not touch any of it.

What does a marina network do that satellite cannot?

It covers the whole estate with one signal the operator controls. A private 5G network on Ofcom's Shared Access spectrum in the 3.8 to 4.2 GHz band runs at power levels WiFi is not permitted to use, which is why a small number of cells, typically mounted on the harbour office, a lighting column at the end of a pontoon and the boatyard shed, can fill a basin that would need dozens of access points. A Low Power Shared Access licence costs £80 per 10 MHz channel per site per year, which is not the number that decides the business case either way.

What it buys is control. The marina owns the channel, so there is no interference from a berth-holder's home router or the café's guest WiFi. Devices handle handover between cells, so a staff handheld works walking from the office to the fuel berth to the hardstanding. Quality of service can be set per device class, so pontoon CCTV and the payment terminals get guaranteed capacity while berth-holder traffic uses what is left. And every one of those operational systems runs on one network rather than four overlapping ones.

The commercial consequence usually interests marina boards more than the technical one. Once the basin is covered, connectivity becomes a service the marina sells, priced like electricity rather than given away like the showers. A per-berth data service at £15 to £25 a month undercuts a Starlink subscription substantially for the boat that rarely leaves, and the revenue stays on site. Our experience is that uptake among liveaboards is close to universal and among weekend berth-holders considerably lower, so the honest model is a modest recurring line plus a strong retention argument, not a windfall.

Where do satellite and a marina network work together?

At the backhaul, and at the harbour entrance. The two technologies solve different halves of the problem and there is no reason to pretend otherwise.

For a remote harbour with no fibre, which describes a good number of west coast and island marinas, a Starlink business terminal on the office roof is a perfectly sound way to backhaul a private 5G network. One dish, one clear view of the sky from the highest building on site, feeding a cellular network that then distributes that capacity intelligently across 200 berths. That is a far better use of satellite capacity than 60 individual dishes each with their own obstruction problem, and it gives the operator a single connection to manage and upgrade. We have written before about hybrid satellite and terrestrial backhaul and the marina case is one of the cleanest examples of it.

For boats that cruise, Starlink stays aboard and stays subscribed. What changes is that it becomes the offshore tool rather than the everyday one, and a good number of owners will drop to a cheaper seasonal plan once the berth is properly covered. We would rather say that plainly than pretend a marina network competes with satellite on passage, because it does not and never will.

What we would do next

No marina needs to compete with Starlink. The decision in front of most UK operators is whether to keep running the business on a patchwork of office WiFi, 4G routers in weatherproof boxes and a single leased line into the harbour office, while the berth-holders solve their own half of the problem at £75 a month.

Three things are worth doing before any capital is committed. Count the dishes in the basin, because that number is the clearest available measure of unmet demand and it costs nothing to collect. List the operational systems planned for the next three years, particularly shore power metering and any electric charging, and ask what each one needs from a network. Then survey the site properly, because the answer for a 120-berth river marina under a tree line is not the answer for a 1,500-berth basin like Brighton, and the honest recommendation in some cases is that public 4G with well-sited external aerials is sufficient for now.

If you want that survey done on your basin, talk to us and we will tell you which of the three cases you are in.