The whole residential electricity business runs on one multiplication. Not a model, not an estimate — a multiplication your meter performs continuously:

kWh = watts × hours ÷ 1,000

A kilowatt-hour is what happens when a 1,000-watt load runs for one hour. That’s it. Your bill is that number, totaled across every device in the house, multiplied by a rate. Everything else on the statement — riders, tiers, demand charges — is bookkeeping around that product; the product itself never changes.

Power vs energy: the confusion everyone makes once

Watts measure power — the rate of draw right now. Kilowatt-hours measure energy — power accumulated over time. The analogy that sticks: watts are your speedometer, kWh is your odometer. A 3,000 W dryer is “driving fast” while it runs, but what you’re billed for is the distance covered: 3,000 W × 0.75 h = 2.25 kWh per load.

This is why the two units can’t be traded: “my heater uses 1.5 kilowatts” says nothing about cost until you add the missing ingredient — for how long.

The monthly version, and worked examples

The estimator uses the same formula stretched over a billing cycle:

kWh/month = watts × qty × hours/day × days/month ÷ 1,000

Run it on real devices and the numbers start to explain bills:

device	math	kWh/mo
9 W LED bulb, 5 h/day	9 × 5 × 30 ÷ 1000	1.35
100 W TV, 4 h/day	100 × 4 × 30 ÷ 1000	12.0
1,000 W window AC, 8 h/day	1000 × 8 × 30 ÷ 1000	240
1,500 W space heater, 4 h/day	1500 × 4 × 30 ÷ 1000	180
3,000 W dryer, 45 min × 12 loads	3000 × 0.75 × 12 ÷ 1000	27.0
7,200 W EV charger, 2 h × 15 days	7200 × 2 × 15 ÷ 1000	216

Then the last line: kWh × effective rate = cost. At $0.16/kWh that AC is ≈$38/mo, the LED ≈$0.22/mo. Same formula, wildly different consequences — which is precisely why the heavy hitters are worth hunting (ranked here).

Reading a nameplate like a pro

Flip any device over and the label tells you its ceiling. Three label styles you’ll meet:

• “120 V ~ 1,500 W” — the honest one. That’s the max continuous draw; use it directly.
• “120 V ~ 12.5 A” — amps instead of watts. Multiply: 12.5 A × 120 V ≈ 1,500 W (resistive loads; for motors the true wattage runs a bit lower thanks to power factor, which makes the estimate conservative — you’ll slightly overstate cost, the safe direction).
• “Input: 100–240 V ~ 0.5 A” on a power brick — the brick’s ceiling across all voltages, not the device’s draw. A laptop “65 W” adapter usually feeds a machine idling at 15–25 W; the brick just has headroom. Use the device’s own spec when available.

And the label you won’t find: standby draw. It’s rarely printed — a device nameplated 400 W may still sip 2 W all night, which is the phantom-load story.

Where the arithmetic bends — three honest caveats

Nameplate watts are a ceiling. The label on the back is the most the device may draw, not what it draws on average. Thermostatted appliances cycle: a fridge nameplated ~400 W spends most of its day with the compressor off and averages more like 60–150 W. An oven element cycles to hold temperature. The estimator handles this two ways — fridge uses a duty-cycled average over 24 h; the water heater uses real element watts but only ~3 h/day of actual heating time.

Electronics have modes. A desktop PC idles near 65 W, works at ~200 W, games at 400–500 W. A game console pulls ~150 W in play and ~1 W in rest. Use the wattage for the mode you’re actually in, for the hours you’re actually in it — or accept the fudge factor knowingly.

Standby isn’t zero. “Off” often means “waiting”: TVs, cable boxes and consoles keep sipping 0.5–30 W around the clock. Individually small, collectively a real line item — the honest accounting is in phantom loads.

When the formula isn’t enough

Two cases where nameplate math genuinely fails:

• Anything with a compressor or heating element and a thermostat — measured consumption beats computed. A $25 plug-in energy meter (the Kill A Watt and friends) reads actual kWh over a day or a week and ends all guessing. For 240 V built-ins, a smart-panel monitor or your utility’s hourly-usage portal does the same job.
• Time-of-use rates. The kWh math is identical, but the rate changes by hour — 3× more expensive at 5 p.m. than at 2 a.m. on some plans. The bill-reading guide covers spotting and using those.

A seasonal example, end to end

Put it together on a real question — “why did my bill jump $60 in July?” A window AC (1,000 W) running 8 h/day adds 240 kWh ≈ $38 at $0.16. The fridge works harder against the warmer kitchen, maybe +15 kWh ≈ $2. The box fan on all night (75 W × 8 h) adds 18 kWh ≈ $3. And the rate itself may have stepped: summer tier pricing or a fuel-cost adjustment adding a cent pushes the same 900 kWh bill up another $9. Stack them: $38 + $2 + $3 + $9 ≈ $52 — close enough to $60 that the mystery is solved without a single guess, just the one formula applied four times.

The takeaway

Watts tell you how hard a device pushes; hours tell you how long it pushes; ÷1,000 turns it into the only unit your utility sells. Once that’s reflexive, a power bill stops being a surprise and starts being an audit — and every “should I unplug this?” question reduces to one multiplication you can do in your head.

Frequently asked questions

Is 1,000 watts the same as 1 kWh?

No — 1,000 watts is a rate (1 kW of power); 1 kWh is an amount of energy. A 1,000 W appliance running for one hour uses exactly 1 kWh. Run it for ten minutes and it uses 0.167 kWh. Power is the speed; kWh is the distance.

How many kWh does a house use per day?

US homes average roughly 29 kWh/day (~880/month, per EIA residential data), but the spread is huge: a mild-climate apartment might use 8–12, an all-electric house in a cold snap can blow past 100. HVAC is usually what separates low users from high ones.

The nameplate says 1,200 W but my meter says less — who's right?

Both. The nameplate is the maximum the device is allowed to draw; your meter measures what it actually draws. A fridge nameplated 400 W might average 80–150 W because the compressor cycles off. For billing math, average-over-time is the honest number.

Why do two 60 W-equivalent bulbs cost different amounts to run?

Because 'equivalent' describes light output, not power draw. A true 60 W incandescent burns 60 W; the LED that matches its brightness burns ~9 W. Same lumens, one-sixth the kWh — which is the entire reason the LED swap math on the estimator looks the way it does.

Can I just multiply volts × amps to get watts?

For resistive loads (heaters, incandescents) yes: a 15-amp space heater on a 120 V circuit is 1,800 W max — which is why most are capped at 1,500 W. For motors and electronics there's a power-factor wrinkle (volts × amps gives VA, slightly above real watts), but for estimating a bill it's a fine approximation.

Does a 240 V device use less energy than a 120 V one?

Not automatically. A 4,500 W water heater is 4,500 W on either voltage — 240 V just needs half the current to deliver it. What matters is watts × time. (The exception: EVs and dryers can charge/heat faster at 240 V, so they draw more power but finish sooner — roughly a wash on kWh.)