Ubiquiti UniFi is excellent kit, and the management interface is the best in its class. Which is precisely why the faults are frustrating: the dashboard shows you a great deal, and almost none of it tells you what to actually do. We are called out to UniFi networks most weeks. The hardware is rarely the problem. It is access points in the wrong places, roaming never tuned, devices on the wrong network, and warnings nobody has been given the context to interpret. Here is what we find most, and what fixing it properly involves.
The 30-second version
- The kit is rarely the problem. Almost every UniFi call-out is design or configuration, not hardware.
- Sticky devices cause most complaints. Your laptop decides when to change access point, and it decides badly unless the network is tuned.
- More access points usually makes it worse. APs too close together compete rather than co-operate.
- Dashboard warnings are symptoms, not causes. A low experience score tells you something is wrong, not what.
- We are certified on Ubiquiti and TP-Link and cover Merseyside and the North West with our own engineers.
On this page
- The signal looks strong, but it still feels slow
- Devices drop off, or cling to the wrong access point
- The internet works, but some devices cannot communicate
- Printers and casting that vanish between networks
- We added more access points and made it worse
- The dashboard is full of warnings nobody can interpret
- Throughput quietly capped by a security setting
- Access points that will not adopt
- Updates that install themselves overnight
- Where UniFi networks actually give businesses trouble
- Frequently asked questions
The signal looks strong, but it still feels slow
Full bars, a healthy speed test, and yet browsing stalls, video buffers and Teams calls break up. This is the complaint we hear most, and the reason it confuses people is that every number they can see looks fine.
A speed test measures one device, standing still, for a few seconds. It tells you almost nothing about how a building full of people share the air. The usual causes are:
- Channel congestion, with your access points and every neighbouring network competing for the same airspace
- Interference from other equipment, or from channel widths set too wide for a busy environment, which trades range and stability for a headline speed nobody achieves
- Access point placement chosen for where a cable already ran rather than where people actually work
- Transmit power left on automatic, so access points shout over each other instead of covering their own area
How we fix it
We go into the UniFi controller and look at what it is actually reporting: channel utilisation, interference, retries and how airtime is being spent. Then we review access point placement against the building and optimise channel widths, transmit power and roaming settings. The goal is consistent performance everywhere people work, not a strong symbol in the corner of the screen.
Devices drop off, or cling to the wrong access point
Somebody walks from the office to the meeting room and the call drops. A laptop in the warehouse is still connected to the access point by reception. Speeds are fine for some people and dreadful for others sitting a few metres away.
This is the single biggest cause of "the Wi-Fi is rubbish" in a building that has perfectly good coverage. Your device chooses when to move between access points, and it will hold a weak connection long past the point of usefulness rather than switching to the strong one overhead.
The fix that often makes it worse
The usual response is to turn up minimum RSSI, which tells an access point to drop a device once its signal falls below a threshold. Used carefully it works. Set too aggressively it stops being a nudge and becomes an eviction, and people start getting thrown off in corridors, stairwells and anywhere coverage is naturally thinner. The complaint changes from slow Wi-Fi to Wi-Fi that keeps disconnecting, which feels worse.
How we fix it
We tune roaming, minimum RSSI, transmit power and coverage together rather than in isolation, because changing one without the others is what causes the eviction problem. We also check whether more access points are genuinely needed or whether the existing ones simply need configuring properly. It is frequently the latter.
The internet works, but some devices cannot communicate
Printers unreachable. CCTV offline in the app while plainly recording. VoIP phones registered but with no audio one way. A payment terminal that works until it does not. Everyone has internet, so the broadband is obviously fine, which makes this feel like a mystery.
It is almost always one of four things:
- VLAN configuration, with devices physically connected but sitting on a network that cannot reach what they need
- Firewall rules on the gateway blocking traffic between networks, either because a rule is missing or because one is doing more than intended
- DHCP problems, most often a scope too small for the number of devices, or something given a static address inside the range being handed out automatically
- Switch port profiles, where a port has been left on the wrong profile or native network, which puts whatever is plugged into it somewhere unexpected
How we fix it
We troubleshoot the complete UniFi network rather than the Wi-Fi in isolation: gateway, switches, VLANs, DHCP, firewall rules, port profiles and access points. The question is always where the traffic is being stopped, and the answer is rarely where the symptom appears.
Three things worth checking in your controller today
- Look at how many access points are running over a wireless uplink rather than a cable. Each one roughly halves the throughput available to everything connected to it.
- Check whether your guest network is genuinely isolated or just a second password onto the same network. If a guest device can see your printer, it is the latter.
- Check when automatic updates are scheduled, and whether anybody would know if one had failed.
Printers and casting that vanish between networks
This one deserves its own section because it appears immediately after somebody does the right thing.
You split the network into VLANs, which is good practice: guests separated from staff, cameras separated from payment equipment. The next morning nobody can find the printer, and the meeting room screen has disappeared from everyone's laptop.
The reason is that the discovery traffic devices use to announce themselves is deliberately confined to its own network. Once the laptops are on one and the printer is on another, the printer is still perfectly reachable, but nothing is telling the laptops it exists. The route is open. The signpost is missing.
The fix is to allow that discovery traffic between the specific networks that need it, rather than either abandoning the segmentation or opening everything back up. It is a small setting, and it is the most common thing left undone after a VLAN project.
We added more access points and made it worse
The instinct when Wi-Fi is patchy is to add another access point, and it is the change most likely to make things worse.
Access points share a limited number of usable channels. Put two too close together, leave both on automatic at full power, and they stop covering the building and start competing for the same airtime. Devices then sit in a zone where two access points look equally good and flip between them, which produces exactly the dropouts the extra access point was meant to solve.
We have improved plenty of networks by reducing transmit power, moving units and occasionally removing one entirely. Coverage is not capacity, and more Wi-Fi is not better Wi-Fi.
How we fix it
We review the whole RF layout and tune it as one system: access point positions, transmit power, channel selection, channel width and roaming behaviour. Sometimes that means adding an access point. Reasonably often it means the opposite.
UniFi installed but not working as well as it should?
We troubleshoot, optimise and manage Ubiquiti UniFi networks, from poor Wi-Fi and roaming problems to VLANs, switches, gateways and internet connectivity. We look at the whole network and fix the underlying problem, not just the symptom.
The dashboard is full of warnings nobody can interpret
UniFi surfaces more information than almost anything else in its class, which is a genuine strength and an occasional trap. High retries, a poor experience score, WAN failovers, packet loss, devices marked as having a bad connection. All real signals, none of which tell you what to do about them.
Two things are worth understanding:
- The experience score is a composite. It can be dragged down by one device at the edge of coverage, or by equipment old enough that it will never score well. A low number is a prompt to look, not a verdict on the network.
- Retries and packet loss are symptoms with several possible causes. They can point to interference, to a device too far from an access point, to a failing cable, to a switch port negotiating badly, or to a problem with the internet connection itself. The dashboard cannot tell you which.
How we fix it
We work out what is behind the alert. That means establishing whether the issue is wireless, switching, cabling, the gateway or the internet connection, and then fixing the cause. The alternative, which we see a lot, is a cycle of rebooting equipment that resolves the symptom for a few days and teaches you nothing.
Throughput quietly capped by a security setting
If you have a fast connection and are nowhere near it, check the gateway before you blame the line.
UniFi gateways include intrusion detection and prevention. They are worth having, but they inspect traffic as it passes, and on several models that inspection reduces the maximum throughput considerably below the figure on the box. Turn the feature on, and a connection that was delivering its full speed can quietly settle well below it, with nothing in the interface presenting that as a warning.
This is not a reason to disable security. It is a reason to know the trade-off exists, to size the gateway for the line you actually have, and to check this before assuming the fault is with your provider.
Access points that will not adopt
An access point showing as Disconnected, stuck on Adopting, or reporting Managed by Other is usually a discovery problem rather than a faulty unit.
UniFi devices find their controller automatically when they are on the same network. Introduce VLANs, move the controller somewhere sensible, or put management on its own network, and that automatic discovery stops. The access points carry on broadcasting Wi-Fi using the last configuration they received, which is why the Wi-Fi appears fine while the controller insists everything is offline.
The fix is to give them a way to find the controller across networks. It takes minutes once you know that is the problem, and it causes days of confusion when you do not.
Updates that install themselves overnight
UniFi can apply firmware updates automatically, and by default it may do so outside working hours. Most of the time this is fine and genuinely useful.
Occasionally it is not. An update can reboot access points, briefly take a site offline, or introduce a change in behaviour on a device that was working perfectly the previous day. If nobody is watching, the first anybody knows is a Monday morning where something has changed and there is no obvious reason.
Our view is that updates should happen, on a schedule you have chosen, with somebody who notices if one fails. That is a management question rather than a technical one, and it is the difference between a network you hope is fine and one you know is fine.
Where UniFi networks actually give businesses trouble
Ranked by how often we are called out to each, across offices, warehouses, practices and hospitality venues:
| What people report | How often we see it | What it usually turns out to be |
|---|---|---|
| Strong signal, slow performance | Very common | Channel congestion and access points left on automatic settings at full transmit power |
| Devices stuck on a distant AP | Very common | Roaming and minimum RSSI never tuned, so the device decides and decides badly |
| Devices cannot reach each other | Very common | VLANs, firewall rules or a switch port on the wrong profile |
| Printers and casting disappear | Common | Multicast traffic not passed between networks once VLANs were introduced |
| Worse after adding access points | Common | New APs placed too close, competing with the ones already there |
| Dashboard warnings nobody understands | Common | Symptoms surfaced without a cause: retries, packet loss, WAN failovers |
| Nowhere near the expected speed | Occasional | Intrusion detection enabled on the gateway, capping throughput |
| Access point will not adopt | Occasional | Controller on another subnet with no discovery method in place |
| Outage nobody caused | Occasional | Automatic firmware updates applied overnight |
Tell us what your network is doing
If any of this sounds like your building, send us the details. We will tell you honestly whether it is a configuration fix, a design problem or a genuine hardware limitation, and we will tell you when the answer is cheaper than you expect, which it frequently is.