Most homes leak money through predictable, boring holes. Not dramatic failures — small behavioral ones: a heat pump that fights the sun, a water heater working at 3 a.m. for nobody, vampire loads that never sleep. Utility bills flatten these into one monthly number, which is exactly why they survive.
Here are the five patterns we see most often when EnergyIQ analyzes a household — and why catching them by hand is nearly impossible.
1. The Phantom Schedule
Your thermostat and water heater run on a schedule you set years ago, for a household that no longer exists. Kids moved out, someone started working from home, weekend routine flipped. The schedule keeps executing the old life. Cost: typically 8–15% of heating and cooling spend. Nobody re-audits these settings, because nobody remembers they exist.
2. Peak-Hour Drift
In time-of-use tariff regions, running the dishwasher at 6:40 p.m. instead of 9:00 p.m. can triple the cost of that cycle. Multiply by laundry, EV charging, pre-cooling. Humans underestimate time-of-use multipliers by an order of magnitude; the tariff table is boring on purpose.
3. The Baseline Creep
Every home has a "everything is off" floor. When that floor rises 40W at a time — an old set-top box here, a forgotten dehumidifier there — it adds up silently. A 60W creep is roughly half a kilowatt-hour every day, ~$2–3 per month, every month, forever. It never appears on a bill as a line item.
4. Weather-Blind Heating
The classic: two sunny autumn afternoons, and the heating runs the same both days. Solar gain is free heating — unless your system ignores it. Weather-blind control wastes the exact days when the house could have mostly heated itself.
5. The Appliance That Got Old
Appliances degrade gradually. A fridge with a failing door seal, a heat pump low on refrigerant, an HVAC filter nobody changed: each consumes 10–30% more than its healthy baseline. The only human-visible symptom is... a slightly higher bill, which gets blamed on rates.
Why This Is an AI Problem
Look at the shape of the list: none of these are visible in a single day's data. Each one is a pattern deviation — a schedule that doesn't match occupancy, a baseline that drifted, consumption that ignores weather and tariffs. Detecting pattern deviations across months of interval data, then explaining them in one clear sentence with a dollar figure attached, is a data problem, not a willpower problem.
That is the division of labor that actually works:
- You decide what comfort you want and what changes you'll accept.
- The model watches the meter data, cross-references weather and tariffs, and surfaces only the anomalies worth your attention, ranked by dollars.
Willpower-based energy saving fails because it asks you to be an always-on analyst. Pattern-based saving works because the boring part runs in the background.
The Payoff Math
Across the five patterns above, a typical home recovers 10–20% of its energy spend when they're caught and fixed — often with zero equipment purchases, purely from schedule and habit corrections. The catch has always been the labor cost of the audit. That is precisely the cost AI removes.
EnergyIQ turns smart-meter and connected-device data into ranked, dollar-quantified savings actions for your home — no spreadsheet required.
