Time-of-use (TOU) rates have an awkward relationship with rooftop solar: your panels produce the most at midday, but the expensive hours are the early evening — typically somewhere around 4–9 p.m., after solar output has fallen off a cliff. If you do nothing, you sell (or self-use) cheap midday power and buy expensive evening power, and your bill savings underperform what the installer’s proposal promised.
The good news is that the fix is mostly behavioral, and it’s the same discipline whether or not you own a battery: move every flexible load into the cheap windows, and shrink what’s left in the peak. We run a 40-panel roof on two Tesla inverters at our own house, and watching the daily production curve against the rate windows is what turned TOU from a penalty into an advantage for us — the midday production hump and the evening price spike are two different problems, and you manage them separately.
First, Learn Your Three Numbers
Every TOU plan reduces to three numbers, and you can’t optimize until you’ve written them down from your actual tariff (not a blog’s example):
- Peak rate and hours — what you pay during the expensive window.
- Off-peak rate and hours — what you pay in the cheap window, often overnight.
- Export credit — what the utility pays or credits for solar you send to the grid, which may differ by time of day and is often well below the retail rate under newer net-billing rules.
The relationships between these numbers set your strategy. A wide peak/off-peak spread rewards aggressive load shifting. A weak export credit rewards self-consuming midday solar instead of exporting it. As a labeled example: with a $0.40/kWh peak, $0.18 off-peak, and $0.08 export credit, every kilowatt-hour of laundry you move from 6 p.m. to noon saves you $0.32 — you avoid the peak purchase and give up only the small export credit.
Also check whether your utility offers multiple TOU plans (many have an EV-specific plan with very cheap overnight rates). Solar households frequently save more by switching plans than by any behavioral change.
The Load-Shifting Playbook
Here’s the core of it — which loads can move, where to move them, and why. The savings mechanics assume the example rates above; your spread will differ.
| Load | Flexibility | Move it to | Why it works |
|---|---|---|---|
| EV charging | High — it’s a scheduled task, not a lifestyle | Midday if home (self-consume solar), otherwise the overnight off-peak window | Largest single shiftable load in most homes; every kWh moved off peak captures the full rate spread |
| Dishwasher, laundry | High — delay-start buttons already exist | Midday solar window | Pure behavioral change, zero hardware cost |
| Heat pump water heater | Medium — needs a schedule or timer | Heat the tank hard at midday, coast through the evening | The tank itself is a thermal battery; hot water made from solar at noon is still hot at 8 p.m. |
| Air conditioning | Medium — pre-cooling works | Pre-cool the house 1–3 °F below your setpoint in the hour or two before peak, then raise the setpoint during peak | The house’s thermal mass rides through the expensive window; a smart thermostat automates this |
| Pool pump | High | Midday | Big, dumb, perfectly schedulable load |
| Cooking, evening lighting, TV | Low — this is just life | Can’t move meaningfully | This is the residual peak load you either buy from the grid or cover with a battery |
Work the table top-down. In most homes, EV charging plus water heating plus laundry covers the large majority of shiftable energy, and everything below it is fine-tuning. The honest ceiling: an ordinary household can’t shift dinner, and the residual evening load is what determines whether a battery makes sense.
Where a Battery Fits (and When It Doesn’t)
A home battery is, financially speaking, a machine for buying electricity at midday prices and using it at peak prices. Whether it earns its cost comes down to arithmetic you can do on one sheet of paper:
- Estimate your residual peak consumption — what’s left in the expensive window after you’ve done the load shifting above. For many homes that’s a handful of kilowatt-hours per evening.
- Multiply by the spread between your peak rate and what the stored solar would otherwise have earned as export credit. That’s your daily arbitrage value; scale it to a year.
- Compare against the battery’s installed cost (after the federal tax credit, which currently applies to standalone and solar-paired storage — verify current terms before counting it).
With a big rate spread and a weak export credit, the math can be genuinely good. With a narrow spread or generous net metering, it’s often poor, and load shifting alone captures most of the available savings for free. Batteries also carry non-financial value — backup power during outages — and it’s fine to buy one for that reason; just don’t let a proposal blur the two justifications together.
If you already own a battery, make sure its operating mode matches your rates: most systems (Tesla, Enphase, and others) have a time-based control mode where you enter your utility’s peak windows and the battery charges from solar at midday and discharges through the peak automatically. A battery left in pure backup mode on a TOU plan is leaving money on the table every single evening.
Review It Twice a Year
TOU optimization isn’t set-and-forget, but it’s close. Two things drift:
Seasons. Solar production, sunset time, and AC load all move through the year, and many utilities shift peak windows or prices seasonally. The pre-cooling strategy that’s essential in July is irrelevant in January, and winter’s shorter days may mean less midday surplus to work with.
Tariffs. Utilities revise TOU windows and export credit rules with little fanfare. A plan that was optimal at installation can quietly become second-best.
A twice-yearly check — spring and fall — is enough: confirm your rate plan is still the best fit, confirm your battery schedule and appliance timers match the current peak windows, and glance at a month of bills to see whether evening grid purchases are creeping up. Whatever monitoring app your system uses will show consumption against production by hour; the pattern you want to see is imports pinned near zero during the peak window and the big loads clustered under the midday production curve.
None of this requires new gadgets or heroic effort. Learn your three numbers, move the movable loads, size any battery against the residual — and the evening price spike stops being a tax on your solar investment and becomes the thing your system is quietly engineered around.