Most people don’t overspend on CCTV because they’re careless. They overspend because the conversation gets hijacked by features.
Look, if an installer is pitching you “AI smart detection,” “color night vision,” and “ultra-mega zoom” before they’ve even asked what you’re trying to protect… you’re already on the slippery part of the slope.
One line you should keep in your head the whole time: you’re buying coverage outcomes, not camera specs.
Hot take: More cameras doesn’t mean more security
I’ve seen eight-camera systems that still couldn’t show a face at the front door. Meanwhile, a tight four-camera layout, proper angles, correct lens choice, clean lighting, caught everything that mattered.
The goal isn’t just to get CCTV installed.
It’s to record useful video.
Start with goals, not gear (the part salespeople skip)
Grab a rough floor plan or just sketch your property. Then ask yourself three blunt questions:
– What do I actually want this footage for, deterrence, evidence, live monitoring, or all three?
– Where do incidents realistically happen (not where you imagine they happen)?
– When do I need visibility, overnight, business hours, weekends, deliveries?
Now translate that into zones:
High-priority zones usually include entrances, cash/stock areas, vehicle approaches, gates, and anywhere people “pause” (because faces show best when someone slows down). Hallways and wide open yards? Often lower value than people think.
Allow for blind spots on purpose. Total coverage everywhere is expensive and often pointless.
Coverage planning: angles beat resolution
A 4K camera pointed at the wrong place gives you 4K of disappointment.
When you map camera roles, think like this:
Deterrence camera: obvious placement, wide view, “I’m being watched” energy.
Evidence camera: tight framing on faces, plates, hands, entry actions.
Live monitoring camera: stable, low-latency, minimal false alerts.
Overlap can help, but overlap can also become an excuse to sell you extra units. In my experience, installers love overlap because it’s an easy upsell that’s hard to argue against.
The must-haves vs the “sounds cool” list
Here’s the thing: most modern cameras are “good enough.” The difference between a smart purchase and an overpriced one is deciding what you’ll actually use.
Must-haves (for most installs)
– PoE (Power over Ethernet) cameras and switches where possible
Fewer failure points. Cleaner installs. Easier troubleshooting.
– A real NVR if you want fast playback and local control
Cloud is fine, but cloud-only systems can quietly bleed you with monthly fees.
– Proper lens selection (this matters more than brand for many sites)
– IP weather rating appropriate to exposure (IP66 is common for outdoors; don’t buy IP67/68 just to feel safe)
Nice-to-haves (only if your site demands it)
– AI person/vehicle classification (helpful, but not magic)
– Color night vision (works great in some lighting, looks awful in others)
– PTZ (pan-tilt-zoom) cameras
Opinion: PTZ is frequently a toy unless someone is actively controlling it or you’ve planned presets carefully.
One more: don’t pay extra for “mixed power setups” unless there’s a real constraint. Hybrid power tends to create messy installs and messier support calls later.
Recording & storage: choose a retention window like an adult
Storage is where budgets quietly blow up.
A simple approach that usually lands well:
– Local NVR for continuous recording
– Cloud backup for critical clips/events only (optional, but smart for theft/fire resilience)
Retention is not a status symbol. It’s a risk decision.
Many sites do fine with 30, 90 days. Longer retention makes sense if you’ve got slow discovery risk (tenant disputes, inventory shrinkage patterns, legal requirements). Otherwise you’re paying to keep video no one will ever watch.
A specific benchmark: video storage needs scale fast with resolution and frame rate, and vendors love pushing “max settings everywhere.” The UK’s Information Commissioner’s Office (ICO) generally expects retention to be justified and not excessive for CCTV systems, which is a useful mindset even outside the UK. Source: ICO guidance on CCTV/data retention practices (ico.org.uk).
Also: insist on encryption in transit and at rest if cloud is involved. If your installer looks confused when you ask, that’s a warning.
Camera features that matter in real life (not in brochures)
Night vision
Don’t chase “longest IR distance” numbers. Check the distance you need to identify someone, not the distance you can vaguely detect movement.
If you need identification at 10, 15 meters, design for that. If the camera is trying to light up 40 meters of darkness, it’ll often wash out the closer subjects (faces go white, details vanish).
Glare reduction (WDR)
If you’ve got headlights, glass doors, or a bright background, WDR is not optional. Bad WDR turns plates into glowing rectangles.
Focal length
This is where installs win or lose.
Wide lens: covers more, but details shrink.
Tighter lens: captures details, but you’ll miss adjacent activity.
I’ve watched people buy wide lenses for entrances because “I want to see everything,” then complain the footage can’t identify anyone. Of course it can’t. The subject is 3% of the frame.
Weather rating and mounting
Outdoor cameras need proper sealing, yes. But the mount is just as important. A slightly vibrating soffit mount will ruin image clarity at night (and yes, wind can do that).
Installation reality check: terrain, access, cable paths
Some sites are easy. Others look easy until you’re crawling through insulation trying to route cable without drilling into something expensive.
So do a quick pre-install assessment:
– Where can you safely run cable without making it visible or vulnerable?
– Do you have stable mounting points (brick vs fascia vs poles)?
– What does rainwater do on your property? Drip lines matter. Flooding matters.
– Do you need lifts or special access gear?
Terrain and environment aren’t “nice-to-consider.” They decide labor cost.
If the installer hasn’t walked the site and is still giving a confident number, you’re not getting a quote. You’re getting a guess.
One-line truth: Labor is where CCTV budgets go to die.
Quotes: demand itemization or treat the number as fiction
Ask for a quote that separates:
Equipment, cabling, labor, mounts, network gear, configuration, testing, training, and ongoing fees.
And get clarity on what triggers extra charges:
– Longer cable runs than expected?
– Core drilling?
– Patch repairs and painting?
– Lift rental?
– After-hours work?
Push for a fixed price for a defined scope. If they won’t do it, at least insist on fixed labor rates and a written change-order process.
Now, this won’t apply to everyone, but if you’re in a high-cost area or a complex building, I’d rather see a slightly higher quote with tight scope than a cheap quote full of “allowances” and “T&M as needed.”
Warranties and support (where cheap systems get expensive)
Some warranties are basically marketing. You want specifics:
– Replacement timelines
– What counts as “install-related” failure vs hardware failure
– Whether firmware updates are supported or ignored
– On-site labor coverage (the part that hurts)
Service plans can be worth it, but don’t buy a gold-plated contract by default. Ask how often you’re realistically going to need on-site support and what a single visit costs without a plan.
Transferability matters too. Selling a property with a transferable service agreement is cleaner than you’d think.
Security, privacy, and future-proofing: don’t skip this part
If your system ends up on the internet, and many do, security isn’t a bonus feature. It’s the baseline.
Minimum standard I’d accept:
– Unique user accounts (no shared “admin/admin” nonsense)
– Least-privilege access (staff shouldn’t have admin rights)
– MFA for remote access where supported
– VPN for off-site viewing if possible
– Automatic firmware updates or a documented update process
– Secure backups and separated encryption keys
Also ask where footage can be accessed from. Some cloud platforms allow broad sharing by default. That’s convenient right up until it isn’t.
Budgeting and negotiation that actually works
Set three numbers before you talk to anyone:
1) Max upfront hardware + install
2) Max monthly ongoing cost
3) Cushion for surprises (usually 10, 15%)
Then negotiate like someone buying a system, not a dream.
Good tactics that don’t turn the conversation weird:
– Ask for a “base coverage” package first, then add optional layers
– Propose a phased install (core areas now, expansion later)
– Request alternative camera models that meet the same goal
– Require a written scope, timeline, and handover checklist
If you feel pressured into extras, pause and ask: What problem does this solve on my property, specifically?
If the answer is vague, the feature is probably fluff.
The deal you want (and the one you don’t)
A solid CCTV deal feels boring on paper:
Clear coverage plan. Correct lenses. Sensible retention. Itemized quote. Defined support.
A bad deal is exciting:
Shiny features. Big promises. “Future-proof AI.” Mystery fees. Unclear scope.
Buy boring. Installers who do clean, disciplined work will respect that, and the ones who don’t are doing you a favor by revealing it early.