We install UniFi. That doesn’t make it right for everyone, and pretending otherwise would waste your money.

Every page on this wiki is about how UniFi works. This one is about when it shouldn’t be what you buy.

We’d rather tell you that in the first conversation than have you find out in the third year. Knowing when not to sell you something is part of the job — arguably the part that makes the rest of the advice worth anything.

Here are the honest cases.

1. A single-AP home, where a decent consumer router is genuinely enough

If you’re in an apartment, a condo, or a compact single-storey house — one access point covers it, you have no cameras and no plans for any, and nothing about your current Wi-Fi frustrates you — a good consumer router is a real answer. Not a compromise. An answer.

UniFi’s advantages are things you’d never use in that scenario: multiple access points roaming as one network, VLANs, granular firewalling, an app that manages cameras and doors alongside Wi-Fi. In one room, with one AP and no cameras, that’s a management layer over nothing.

Where the line honestly is. It’s worth a look if: you have dead zones a second AP would fix, more than one storey, wired cameras in the plan, a home office you want segmented off, or a Wi-Fi setup you already resent. Any of those and UniFi starts earning its keep. None of them, and it doesn’t.

2. Anything that has to be certified life-safety or fire-alarm

This one isn’t a preference. It’s a boundary, and it’s firm.

If your project needs:

  • fire alarm, or integration with a fire alarm panel,
  • certified life-safety monitoring,
  • a monitored intrusion system with a certified central station,
  • anything an insurer or an inspector requires a certificate for,

that is a licensed alarm contractor’s work. It is not ours, and UniFi is not the product for it.

UniFi Protect is video surveillance. UniFi Access is door entry. Neither is a certified life-safety system, neither is sold as one, and no amount of good design turns one into the other.

Fire and life-safety go to the appropriate licensed trade, and the requirements are set by local code and the Authority Having Jurisdiction. That confirmation is not ours to give, and nothing on this wiki is code, legal or compliance advice.

This overlaps with real projects constantly — a building often wants cameras and a fire panel. The answer isn’t that we can’t help; it’s that the licensed trade does their part and we do ours, and the boundary between the two is stated in writing rather than blurred. If someone quotes you UniFi as your fire or life-safety system, that’s the moment to stop and ask a licensed contractor.

Related, and in the same category: egress. A fail-secure strike stays locked when power fails, which is why AVNFi pairs every one with mechanical egress — a push bar or pad that works with no power and no network. Same caveat applies: local code and the AHJ govern, and must be confirmed for your specific door. See What Works With What.

3. You want it fully cloud-managed, with no local hardware

Some people genuinely want the cloud model: nothing in a cupboard, no box to maintain, no drive to fail, someone else’s data centre holding the footage, and a monthly fee that covers all of it.

That is mostly not UniFi, and it’s worth being precise about which half:

  • The controller doesn’t have to be local. Ubiquiti sells Official UniFi Hosting, a paid monthly subscription that runs your UniFi Network controller on its infrastructure, and an independent gateway (the UXG line) to pair with it. If your objection is “I don’t want a brain in the cupboard”, that has an answer, and we can build it.
  • Video records to a drive you own, in a box you own, in a room you own. That part is not negotiable, and it’s the part people usually mean.
  • That drive is your responsibility. You buy it, you size it, and when it eventually fails, it’s yours to replace.

For most people the local model is the selling point: no monthly creep, footage that stays in the building. But it’s only a selling point if you want the hardware. If your honest preference is “I want a service, not equipment”, that’s a legitimate preference, and for cameras specifically, a cloud-native platform serves it better than we can. Buying UniFi and then resenting the box in the cupboard helps nobody.

4. Environments needing gear we can’t support

We’re certified UniFi specialists. That’s the scope, and it has edges:

  • Industrial control, SCADA, process networks with their own protocols and their own contractors.
  • Enterprise networks with an incumbent vendor and a support contract — where the answer needs to be Cisco or Meraki because that’s what the client’s team runs, is trained on, and is contracted for.
  • Specialist AV, broadcast, or low-latency audio infrastructure.
  • Anything requiring certifications or a trade licence we don’t hold.

The honest version: if we can’t support it properly, we shouldn’t sell it to you. A system installed by someone who can’t maintain it is worse than no system.

Worth separating out: if you have a working third-party network and want UniFi cameras or door access on it, that’s usually not this case — it’s a normal, supported design, with two named caveats about whose network it is. See What Works With What, Rule 6.

5. Very small camera counts, where a plug-in camera is honestly fine

One camera on a garage. One on a nursery. A doorbell and nothing else.

For that, a plug-in consumer camera is a perfectly reasonable purchase, and we’ll say so.

Here’s the arithmetic, because it’s the point. A UniFi camera system means a recorder (or a gateway with a drive bay), a PoE switch, a surveillance-rated drive, and cable pulled to each camera. That infrastructure is most of the cost, and it barely changes between two cameras and eight. Spread across eight cameras it’s excellent value. Spread across one, it’s a lot of money for one camera.

Where the maths flips:

  • 1–2 cameras, no other UniFi gear, no plans → a plug-in camera is honest value. Buy one.
  • Four or more, or you already have a UniFi gateway with a bay, or you know more are coming → UniFi wins, and the gap widens with every camera.
  • Somewhere between → it’s a genuine judgement call, and it depends on your building rather than on a rule. Ask.

What you give up with the plug-in camera — and it’s real: cloud subscriptions to keep the footage, the vendor’s servers holding it, Wi-Fi dependence, wall-socket dependence, and no possibility of it becoming part of a system later. Which may be fine for a nursery, and is not fine for a perimeter.

And note the same trade inside UniFi’s own range: the G6 Instant is the plug-in one — real 4K, two-minute install, 6 metres of night vision, indoor only, wall socket. Excellent for the spot you couldn’t cable. Not a system.

6. Nobody will ever administer it

The quietest failure on this list, and the one we see coming most often.

UniFi is a managed system. That’s its strength — and it assumes someone, occasionally, will manage it. Not constantly, not a full-time job, but somebody who will:

  • apply firmware updates,
  • notice when a camera has been offline for three weeks,
  • notice when a drive is failing before it fails,
  • add and remove door access cards when staff change,
  • open the app once in a while.

If that person doesn’t exist, the system decays quietly. It won’t announce it. The failure mode is discovering, on the one day the footage mattered, that a camera dropped off in March.

Three ways this ends well:

  1. Someone on-site owns it. Doesn’t need to be technical — needs to be reliable, and to open the app.
  2. AVNFi maintains it under an arrangement, so the answer to “who’s watching this” is a name.
  3. Deliberately smaller scope. Fewer things to look after, honestly specified, rather than a big system nobody tends.

What doesn’t end well: a large system, no owner, no maintenance, and everyone assuming it’s fine because nothing has beeped. If nobody will ever look at it, buy less of it. That’s not us talking you down — it’s the difference between a system that works in year three and one that only looked good on install day.

7. When the honest answer is “keep what you have”

The most common wrong sale in this industry isn’t the wrong product. It’s a product the customer didn’t need at all.

If you have UAP-AC-Lite, AC-Pro or AC-LR access points, a UDM-Pro, a UNVR, or G4 or G5 cameras — those are current products in Ubiquiti’s catalogue as of July 2026. Not obsolete. Not end-of-life. Not “due.” Working equipment doing its job.

“No longer sold” is not “no longer works,” and the gap between those two sentences is where a lot of unnecessary money changes hands.

There’s a short, specific list of gear that genuinely does need replacing — USG gateways, the UniFi Video platform, Cloud Key Gen1, the 802.11n access points. It’s short on purpose, and we can show you exactly why each item is on it.

Everything else: if it works, keep it. The full, sourced version — including a correction to some widely-circulated blog claims about the UAP-AC-Pro — is here: I Have Old Ubiquiti Gear. Do I Need to Upgrade?

Why this page exists

Because the alternative is worse.

A specialist who thinks their product is right for everyone isn’t a specialist — they’re a salesperson with a favourite. UniFi is genuinely excellent at a specific set of things: multi-AP networks that roam properly, camera systems with no subscription and local footage, door access that lives in the same app, and all of it manageable by a competent person on their phone. When those are what you need, it’s very hard to beat.

When they’re not, saying so costs us one sale and earns something worth considerably more.

So if you’re reading this and thinking “that’s me” — good. That’s the page working. And if you’re not sure which side of one of these lines you’re on, that’s the most useful question you could bring us.

Related

The only way to know which of these applies to you is to look at your building and ask what you actually want it to do. If the answer turns out to be “not this,” we’ll tell you — and that conversation is free. AVNFi.