The most common way to waste money on a smart home is to buy a lot of it at once.
People decide to "do the smart home thing", order a bundle of lights and sensors and a hub, spend a weekend setting it up, and discover a month later that they use two of the twelve devices. The rest end up in a drawer.
Here is a better approach.
Start with an annoyance, not a product
The useful question is not "what smart home devices should I buy?" It is "what do I keep doing that irritates me?"
Some real examples:
- Getting up to turn off the lounge light after sitting down
- Never knowing whether the garage door is shut after leaving
- Walking down a dark hallway to the bathroom at 2am
- Forgetting to switch off a heater or towel rail
- Wondering if a parcel arrived
- Not knowing if the kids got home from school
Each of those has a cheap, specific solution. Each solves a real problem you will notice being solved every day. That is what makes people keep going, and it is what makes the second purchase an informed one.
Contrast that with buying a colour bulb multipack because it was on sale. Colour lighting is fun for a fortnight and then you leave it on warm white forever.
The first purchase
Buy three or four things, not twenty.
1. One controller. A smart speaker or display from whichever ecosystem suits your phones. This gives you voice control, and on most platforms it also serves as the hub that runs automations when you are out. An entry-level speaker is fine — you are testing the platform, not building a stereo.
2. Two or three devices that solve your actual annoyance. Smart bulbs for the lounge. A plug for the heater. A contact sensor on the garage door. Whatever matches the problem you identified.
3. Nothing else. Not yet.
Then live with it for a month. You will learn things you cannot learn from reviews: whether voice control fits how your household behaves, whether anyone else uses it, whether automations feel magical or intrusive, and which room you actually wish you had done first.
That month is the most valuable part of the process, and skipping it is what fills the drawer.
Choosing a platform without agonising
The platform decision used to be permanent. Matter has made it much less so — most new devices work with all the major platforms, and can be shared between several at once.
The short version:
- Everyone on iPhone? Apple Home. Reliable, private, weakest automations.
- Mixed household or Android? Google Home. Best voice assistant.
- Want the cheapest speakers and widest device support? Alexa.
- Enjoy tinkering and want real control? Home Assistant, but not as your first purchase.
Our full platform comparison goes deeper. For a first purchase, pick based on the phones in the house and move on — this decision is far less binding than it feels.
One buying rule: prefer devices with Matter support. It keeps your options open at effectively no cost.
What to buy first, by problem
Lighting you keep getting up to switch off. Smart bulbs if the fitting has one or two globes; a smart switch if there are several. In Australia, check the fitting is B22 bayonet or E27 screw before ordering — imported multipacks frequently only fit E27 or the US E26. Smart switches require a licensed electrician here, so budget accordingly.
Appliances you forget to turn off. A smart plug. Cheapest useful device in the category. Buy one with Australian approval markings and check the load rating against the appliance.
Not knowing if a door or garage is open. A battery contact sensor. Cheap, installs with adhesive, lasts a year or more on a coin cell.
Dark hallways at night. A motion sensor plus a smart bulb, with an automation that only triggers after dark and at low brightness. This is the automation people love most and it costs very little.
Wanting to see who is at the door. A video doorbell — but read our doorbell guide first, because the subscription decision matters more than the hardware.
The mistakes that cost the most
Buying a large bundle up front. Covered above. It is the big one.
Buying Z-Wave from overseas. Z-Wave uses different radio frequencies in different regions, and the Australia/New Zealand allocation differs from North America's. A US Z-Wave device will never work with an Australian hub. No settings change fixes it.
Ignoring subscription costs. Cameras especially. A cheap camera with a mandatory monthly plan costs far more over five years than an expensive one with local storage. Always price five years.
Buying mains-powered devices from grey importers. Anything that plugs into a wall socket should be approved for Australian sale. The savings are small and the risks — non-compliant plugs, insulation not rated for our supply, no local warranty — are not worth it. Low-voltage USB sensors are a much lower-stakes import.
Automating too aggressively too early. Lights that turn themselves off while someone is reading, or a house that decides everyone is out when someone is working quietly at home, turn the household against the whole idea. Start with automations that only ever add convenience, and add the assertive ones once you trust the sensors.
Forgetting other people live there. If your partner cannot turn on a light without your phone, you have built a worse house. Every automation needs an obvious manual override, and physical switches should keep working.
A realistic first-year path
Month 1: speaker plus two or three devices solving one annoyance.
Months 2–3: extend to the rooms you actually use. Usually lighting in the main living area and bedrooms.
Months 4–6: add sensors — motion, contact, temperature. Sensors are what turn remote control into automation, and they are cheap.
Months 6–12: the bigger-ticket items you now know you want. Cameras, doorbell, locks, climate control. By this point you know your platform, your network's limits and your household's tolerance.
Somewhere in there: upgrade your router. Most people hit their router's client limit around thirty devices and spend weeks blaming individual products. If devices start dropping off, read our Wi-Fi troubleshooting guide.
The point
A smart home that solves three real problems reliably is worth far more than one with forty devices and a household that has stopped using it.
Start small, solve something real, and let what you learn drive the next purchase.
