How Many Solar Panels Do You Need? A Homeowner's Sizing Guide
"How many solar panels do I need?" It's the first question almost every homeowner asks, and it's a good one. The honest answer is that there is no single number — your neighbor two doors down with an identical roof may need four more panels than you do. But the math behind the number is surprisingly simple, and you can get to a solid estimate with three inputs you already have or can look up in five minutes.
This guide walks through the formula step by step, shows you where the numbers come from, and points out the real-world constraints — roof space, shading, and local sunshine — that turn a textbook answer into a system you can actually install. When you're done, you'll know roughly how many panels your home needs. And if you'd rather skip the arithmetic, our free solar savings calculator does every step for you.
The core formula: usage ÷ output ÷ sun-hours
Every panel-count estimate reduces to one idea. You need enough panels so that, over a full year, they generate about as much electricity as you use. Put as an equation:
Number of panels = (annual kWh usage) ÷ (panel wattage × peak sun hours × 365) × (derate factor)
Broken into plain English, the three things that decide your panel count are:
- How much electricity you use — your annual kWh, from your bills.
- How much each panel produces where you live — its wattage times your local peak sun hours.
- Real-world losses — wiring, inverter, soiling, and temperature losses, usually captured in an ~0.86 derate.
A worked example
Let's run the numbers for a common scenario: a household using 12,000 kWh per year, installing 400-watt panels, in a region with about 5 peak sun hours per day.
| Step | Calculation | Result |
|---|---|---|
| Daily energy needed | 12,000 kWh ÷ 365 days | 32.9 kWh/day |
| Output per panel per day | 400 W × 5 sun-hours | 2.0 kWh/day |
| Panels (ideal, no losses) | 32.9 ÷ 2.0 | ~16.4 → 17 panels |
| Apply 0.86 derate | 2.0 × 0.86 = 1.72 kWh/panel/day; 32.9 ÷ 1.72 | ~19 panels |
In a perfect world, ~17 panels cover the load. In the real world — where panels aren't perfectly clean, inverters aren't 100% efficient, and hot rooftops reduce output — you land closer to 19–20 panels. That gap between the tidy textbook answer and the installed answer is exactly why installers and calculators apply a derate. Plan for the real-world number, not the ideal one.
If you'd rather not track the derate yourself, our Solar Savings Calculator builds it in and adjusts for your exact address.
Where to get your kWh (and why annual, not monthly)
The most important input is also the one homeowners most often get wrong: they grab one month's usage and call it a day. Don't. Your December bill and your July bill can differ by two or three times if you run air conditioning or electric heat. A single month hides the seasonal swing that defines your real annual load.
The right way: add up the "kWh used" line from your last 12 months of bills. If your utility provides it, a 12-month usage summary or interval data export is even better. This gives you true annual consumption — the number the formula actually needs.
For reference, the U.S. Energy Information Administration (EIA) puts the average American household at roughly 10,000–11,000 kWh per year (about 863 kWh/month in 2024 data, and 10,791 kWh/year in EIA's commonly cited figure).[1] But "average" is a wide band: homes in mild climates like Hawaii use under 500 kWh/month, while air-conditioned homes in the South often exceed 1,100 kWh/month. Your own bill is the only number that matters for your system.[1]
If you don't have a full year of bills handy — maybe you just moved in — our electricity usage calculator estimates annual kWh from your appliances, home size, and location. It's a reasonable starting point, but real billing data will always beat an estimate.
Always size from annual usage, not a single month. A January-only number will under-build your system; a July-only number will over-build it. Twelve months smooths out both.
Panel wattage matters: 350W vs 400W vs 450W
Wattage is simply how much power a single panel produces under test conditions. In 2025–2026, most residential panels fall in the 400–450W range, with some premium models reaching higher.[2] The higher the wattage, the fewer panels you need to hit the same annual output — and the less roof space the system occupies.
Using our 12,000 kWh/year, 5 sun-hour example, here's how wattage alone changes the count (with the 0.86 derate applied):
| Panel wattage | Annual output per panel | Panels needed (12,000 kWh) |
|---|---|---|
| 350 W | ~550 kWh/yr | ~22 panels |
| 400 W | ~628 kWh/yr | ~19 panels |
| 450 W | ~707 kWh/yr | ~17 panels |
The takeaway: stepping from 350W to 450W panels drops the requirement from ~22 to ~17 panels — a meaningful difference if your roof is tight. Higher-wattage panels usually cost a bit more per unit but save on racking, labor, and roof real estate. If roof space is not a constraint, standard 400W panels are the cost-effective default for most homes. For a deeper look at what panels cost and how wattage affects total project price, see our 2026 solar panel cost guide.
Sun-hours by region: why location changes everything
The same 400W panel in Arizona and in Massachusetts does not produce the same energy. "Peak sun hours" measures the equivalent number of hours per day that the sun delivers its full, noontime intensity; it bundles together latitude, cloud cover, and climate. It's the variable that explains why a sunny-state home needs fewer panels than a cloudy-state home using the identical amount of electricity.
The values below come from NREL's PVWatts model, which draws on the National Solar Radiation Database. These are statewide annual averages — your specific roof, shading, and tilt will shift the real number, so treat them as ranges, not gospel.[3]
| State (representative) | Avg. peak sun hours/day | Effect on panel count |
|---|---|---|
| Arizona | ~6.5 | Fewer panels needed |
| California | ~5.6 | Below-average count |
| Texas | ~5.3 | Below-average count |
| Florida | ~5.3 | Below-average count |
| Massachusetts | ~4.2 | Above-average count |
| New York | ~4.0 | Higher count needed |
| U.S. average | ~5.0 | Baseline |
Figures reflect statewide averages from NREL PVWatts; actual site values vary with roof orientation, shading, and local weather.[3]
To make the regional effect concrete: a home using 10,800 kWh/year with 400W panels needs roughly 23 panels in sunny Arizona but about 35 panels in cloudier New York — same house, same usage, different sun. That's why a national "average panel count" is only a ballpark; your address is what decides it. Texas homeowners can see how this plays out specifically in our Texas solar guide.
Roof constraints: when you can't fit the "ideal" count
The formula tells you how many panels you need; your roof decides how many you can fit. Three constraints come up constantly:
- Usable south-facing area. In the Northern Hemisphere, south-facing roof planes capture the most annual energy. A typical panel occupies about 17–20 square feet, so a 20-panel system needs roughly 340–400 sq ft of clear, unshaded south-facing roof.
- Shading. Chimneys, vent pipes, dormers, neighboring trees, and buildings all steal production. Even partial shading on one panel in a string can drag down the whole row, so installers leave buffer space and use optimizers or microinverters to limit the damage.
- Orientation and pitch. East and west faces still work — usually at 80–90% of south-facing output — and west-facing arrays are popular because they produce more during the expensive late-afternoon peak. North faces are the weakest and typically avoided.
If the ideal count won't fit, you have three realistic options: choose higher-wattage panels to pack more output into the same space, accept a system that covers most (not all) of your usage and keeps buying a little grid power, or consider a ground mount or community solar if your roof is truly maxed out. The point is that "roof reality" is the step where the textbook number gets negotiated down — and a good calculator that asks about roof area and orientation will show you the collision before you get a quote.
Battery changes the math
Adding home battery storage doesn't change how many panels you need to cover your usage — but it changes why you size the way you do. Without a battery, the goal is usually to zero out your annual grid draw and let net metering bank your midday excess for nighttime use. With a battery, you may deliberately oversize the array so the battery fills every evening from cheap midday sun, reducing your reliance on the grid during high-rate evening peaks or outages.
Common design targets when a battery is in the mix:
- Self-consumption focus: size the array ~10–20% larger than bare usage so the battery charges fully even on cloudy days.
- Backup focus: the panel count is driven by essential loads (fridge, well pump, lights) you want to ride through an outage, not your whole home.
- Reduced-credit net metering: in states like California (NEM 3.0), exported energy is worth far less than retail, so oversizing to self-consume more of your own solar becomes the smarter play.[4]
If storage is part of your plan, read our solar battery storage guide for how array size, battery capacity, and your rate plan interact — and use our free solar battery savings calculator to model the combination for your address.
The typical homeowner answer: 15–25 panels
After all the variables, what's the number most people land on? Industry guidance — including EnergySage's analysis of hundreds of thousands of quotes — puts a typical U.S. home in the range of roughly 15–25 panels to offset most or all of its electricity use.[2] EnergySage specifically cites about 15–22 panels for an average-usage home on 400–450W modules.[2]
Why such a wide band? Because it stretches from a frugal 900 kWh/month condo in a sunny state (fewer panels) to a 1,500 kWh/month all-electric home in a cloudy state (more panels). The average quoted system size on the EnergySage marketplace sits around 11–12 kW, which at 400W panels is roughly 28 panels — but that skews toward larger homes and higher-usage households.[2]
For most mainstream single-family homes, 15–25 panels is the honest "you're probably somewhere in here" answer. Your actual spot in that range is set by the three levers above: usage, wattage, and sun-hours. To pressure-test whether that system is worth it for your bill, our "is solar worth it in 2026?" breakdown runs the payback math.
Oversizing vs right-sizing: net metering and export limits
There's a temptation to max out the roof "just in case." Sometimes that's smart; sometimes it's wasted money. The deciding factor is your net metering policy.
- Full-retail net metering: excess daytime production is credited at (or near) the retail rate, so a smaller system can still zero out your annual bill. Oversizing mainly helps if you plan to add load (EV, heat pump) later.
- Reduced-credit or no-net-metering: exported energy is worth a fraction of retail (or nothing). Here, oversizing to self-consume more of your own solar — paired with a battery — often pays off because you avoid buying expensive grid power you can't cheaply export.
- Export limits: some utilities cap how much you can send to the grid, effectively limiting system size. Building past that cap without a battery means throwing away production.
The right-size principle: build to cover your true annual usage plus a small buffer for future load, then let your net metering rules tell you whether going bigger makes financial sense. Our Net Metering 101 guide maps the rules state by state so you can see which camp your utility falls into.
When to just use the calculator instead of guessing
The formula in this article is real and useful for intuition — but it assumes you know your derate, your exact roof orientation, your local shading, and your utility's rules. Hand-calculating all of that is error-prone, and small mistakes swing the answer by several panels. That's the moment to stop guessing and let a tool do it.
Our free solar savings calculator takes your address, your annual usage (or pulls an estimate), your roof details, and current incentive data, then returns a panel count, system size, and savings projection in under a minute. It bakes in the NREL sun-hour data for your location and the standard system losses, so you skip the rounding errors. It's also the fastest way to see how adding a battery or switching panel wattage changes the answer.
Use the formula to understand why the number is what it is. Use the calculator to get the number you'd actually install.
Bottom line: usage + roof + sun-hours decide it
If you remember one sentence, make it this: how many solar panels you need is decided by your annual usage, your roof's usable space and orientation, and your local sun-hours — and a free calculator does the math for you.
The three levers recap:
- Usage sets the target (pull 12 months of bills, or estimate with our electricity usage calculator).
- Sun-hours set the local multiplier (sunny states need fewer panels; cloudy states need more).
- Roof sets the ceiling (south-facing, unshaded area limits how many you can fit).
Most U.S. homeowners end up needing somewhere in the 15–25 panel range, but your home's exact spot depends entirely on those three factors. Don't overthink the arithmetic — use our free solar savings calculator to turn your own numbers into a precise panel count and savings estimate in under a minute.
Frequently asked questions
How accurate is the kWh-from-bills method versus a calculator?
Pulling 12 months of usage from your bills gives you the single most accurate input — your actual consumption. A calculator then applies your local sun-hours, roof constraints, and system losses automatically, so it removes the manual rounding you do by hand. If you don't have a full year of bills, our electricity usage calculator estimates your kWh from your appliances and home size, which is a reasonable starting point but less precise than real billing data.
Can I install fewer panels now and add more later?
Yes, but it usually costs more per watt to expand later. Inverters, racking, and permitting are cheaper when installed once. A common compromise is to right-size for today and leave physical roof space (and inverter headroom) for a future addition — especially if you plan to buy an EV or heat pump. Talk to your installer about designing the array and inverter for modest expansion before they finalize the layout.
What if my best roof area doesn't face south?
South is ideal in the Northern Hemisphere, but east and west faces still work well — typically at 80–90% of south-facing output. West-facing arrays are popular because they produce more during the expensive late-afternoon peak. North-facing roofs are the weakest and usually avoided. If no single face has enough space, a mix of east and west rows is a practical solution; a calculator that accepts orientation will reflect the loss automatically.
Do I need more panels if I buy an EV or switch to a heat pump?
Almost always, yes. An EV can add 2,000–4,000 kWh per year, and an electric heat pump can add several thousand more depending on what it replaces. Size your system for expected future usage, not just today's bill, or you'll under-build and leave savings on the table. If you're not ready to commit, design for today and reserve roof and inverter capacity for later, as described above.
Does net metering change how many panels I need?
Net metering changes the value of each panel, not the count you need to cover your usage. Under full-retail net metering, excess daytime production offsets nighttime grid use at full value, so a smaller system can still zero out your bill. Under reduced-credit schemes (such as California's NEM 3.0), exported energy is worth far less, which makes oversizing and adding a battery more attractive. See our net metering guide for state-by-state detail.
Sources
- U.S. Energy Information Administration (EIA) — average residential electricity consumption and monthly usage, 2024 data and FAQ (≈10,000–11,000 kWh/yr; 863 kWh/month in 2024). eia.gov
- EnergySage — "How many solar panels do I need?" (2025 guide): typical home 15–22 panels; average marketplace system ~11–12 kW; 400+W panels predominant. energysage.com
- NREL PVWatts / National Solar Radiation Database (NSRDB) — average daily peak sun hours by state (AZ ~6.5, CA ~5.6, TX ~5.3, FL ~5.3, MA ~4.2, NY ~4.0; U.S. avg ~5.0). nrel.gov
- California Public Utilities Commission — Net Energy Metering (NEM 3.0) reduced export compensation structure. cpuc.ca.gov