Before you look at a single spec sheet or product listing, answer these questions. The utility company profits from your confusion. Walk through this before you spend anything.
This is the cheapest part of the entire process. It costs nothing but your time, and it could save you thousands.
That’s the most common starting point, and it’s a legitimate one. You’re writing a check to a government-protected monopoly every month. They have zero competition. They raise their rates every year. You’d rather keep more of that money.
But “reduce my bill” means different things at different scales. Cut it by 20%? 50%? Eliminate it? Each of those is a different system and a different investment. A modest setup can meaningfully knock down your summer bills. Getting to 50% reduction takes more hardware. Getting to zero takes a serious investment and some lifestyle adjustment.
This is a different goal entirely, and it’s one the utility company definitely doesn’t advertise. When the grid fails — and it does fail — a battery backup system keeps your lights on, your food cold, and your family comfortable. They can’t do anything about it. You can.
Think about what you’d need during an outage: refrigerator, freezer, furnace fan, internet, phone charging, a few lights. That’s probably 500—800 watts continuous. Most families discover this is a pretty modest load once they actually think about it.
If you’ve lived through a multi-day outage — ice storm, hurricane, grid failure — you already know exactly why this matters. That experience is what drives more people to solar than any sales pitch ever could.
If your utility uses time-of-use (TOU) pricing — and many do — peak hours cost significantly more. A battery system lets you charge when power is cheap and discharge when it’s expensive. You pocket the difference. The monopoly loses that revenue.
Check your rate situation to see if you’re on a TOU rate and what the spread looks like. Once you have the battery for backup, capturing TOU savings is essentially free money — you just set a charging schedule and let it run.
Both are legitimate. A purely financial analysis judges the system by payback period and bill reduction. An independence analysis judges it by how you feel during the next outage — and how you feel about not depending on a monopoly that has zero accountability to you.
Most people want both. That’s fine. A system that saves you money and keeps your family comfortable during a grid failure is a good system, even if it doesn’t fully optimize for either objective alone.
Most people come to solar after something happens — a big rate hike, a prolonged outage, a neighbor who’s now energy independent. That trigger is valid. Just don’t let it narrow your thinking.
If a rate hike triggered you, backup power might end up being the thing you’re most grateful for. If an outage triggered you, the bill savings over time might be the thing that actually changes your finances. The full picture is usually better than the trigger alone.
This is one of the most important decisions you’ll make, and it drives everything else.
Most people don’t need to power the whole house. They need to power the circuits that matter — refrigerator, freezer, furnace, internet, a few lights. A 10-circuit transfer switch lets you put your essential circuits on solar and leave the rest on grid power.
The advantage: you don’t need a massive system. You need a system just big enough for the things that count. A 3—5 kW system covers most essentials comfortably. Powering the whole house might require 10+ kW and a battery bank that costs more than a car.
Walk to your breaker panel. Read the labels. Think about what you’d want running if the grid went down. That list is your starting point.
Large appliances — electric dryers, ovens, well pumps, mini-splits, EV chargers — run on 240 volts. A standard 120V inverter won’t touch them.
If you need 240V coverage — especially a well pump or medical device — design for it from the start. See the inverters page for what that actually costs and what your options are. Retrofitting 240V capability later is expensive and labor-intensive.
Here’s where people often underestimate their system needs. A gaming PC that’s always on can draw 150 watts around the clock — 3.6 kWh per day from one device. A home server, a second fridge, an aquarium heater — they add up fast.
Always-on loads inflate your daily consumption and drain your battery overnight while you sleep. Find them before you size anything. A Kill-A-Watt meter gives you accurate measurements on any device. Plug it in and see what you’re actually running.
It probably is. Every day you want to offset as much of your bill as possible. During an outage, you’d be satisfied with just the essentials covered.
Your system handles both. A well-designed transfer switch lets you run more circuits in normal operation and shed to essentials during an outage. Plan for that flexibility — it’s built into the design, not bolted on after.
Solar production is not consistent across the year. Summer output can be five to ten times what winter delivers. If you want year-round impact, design for the worst months. If you want to crush your summer bill and accept less in winter, you can build a smaller, more efficient system.
Neither approach is wrong. They just lead to different system sizes and different ROI profiles. Be honest about which one you’re building.
Production in northern climates drops dramatically in winter. December averages 1—2 peak sun hours in many markets. June averages 5—6. Your panels keep producing in winter — just a fraction of what they do in summer.
The realistic answer for most homeowners: design for the productive months, use the battery for TOU arbitrage and outage backup in winter, and don’t expect full bill offset year-round. Check PVWatts for location-specific production data.
Overnight coverage means battery storage. Batteries are the most expensive component in most systems. The math is direct: estimate your overnight load, multiply by hours of coverage, add a safety margin.
Example: 500 watts overnight for 10 hours equals 5 kWh minimum. Add 20—30% safety margin and you’re at 6—6.5 kWh. That’s your starting point for battery sizing.
If your utility has TOU rates, programming your inverter to charge cheap and discharge during expensive hours improves your returns without generating more solar power. You don’t have to set this up immediately — but make sure the inverter you choose supports programmable charging schedules. Retrofit options are limited.
An automatic transfer switch detects a grid failure and switches your circuits to battery power in seconds — fast enough you often don’t notice. A manual transfer switch requires you to walk to the panel and flip it.
Some people want to watch production and consumption numbers like a hawk. Others want to set it and forget it. Both are fine.
At minimum, you should be able to see your battery state of charge at a glance. You should always know roughly how full your batteries are. Beyond that, your monitoring preference guides equipment selection more than it changes system performance.
If you travel or just want to confirm everything’s working from your couch, remote monitoring is genuinely useful. Most modern inverters support it. Even if you’re a set-and-forget person, you want to know if something stops charging. Most smart inverters push an alert when something goes wrong.
Roof mounting is the most common approach — it’s out of the way and uses space you weren’t using anyway. Ground mounting is easier to install, easier to service, and easier to adjust.
Some people do both — start on a shed or ground rack, add roof panels later. Your roof condition, orientation, and your comfort working at height all factor in. There is no universally right answer.
South-facing, unshaded is ideal. East- and west-facing panels produce roughly 80% of south-facing output — still very viable. Partial shade is the biggest performance killer: a shadow across part of a panel can reduce output for an entire string.
The best diagnostic is free: go outside on a sunny day and watch where shadows fall at different times of day. Do it in both summer and winter if you can.
Your equipment needs a home: protected from weather, reasonably temperature-controlled, accessible for maintenance. Garage, basement, utility room, or a covered area near the panel array.
Batteries matter here: LiFePO4 batteries don’t charge well below freezing. An unheated garage in a cold climate may need a heated enclosure or indoor placement in January.
Distance affects wire gauge and cost. A 20-foot run is easy. A 100-foot run requires serious wire sizing analysis to control voltage drop. Decide where panels go and where equipment lives at the same time — don’t optimize one without considering the other.
Your transfer switch connects to your main breaker panel. Check it now. Is it accessible? Does it have open slots? If the panel is full or buried in a finished wall, you may need a subpanel — that’s additional cost that belongs in the budget from the start.
There’s a real system at almost every budget level. A few hundred dollars gets you a portable setup to learn with. A few thousand gets you a system that covers essential circuits. $5,000—$9,000 gets you something that makes a serious dent in your bill and provides solid backup power.
See the System 1 (200W starter build) and System 2 (3.2kW permitted system) as concrete examples of what different budgets actually buy.
Some things are easy to add later: more batteries, additional panels. Some are expensive to change: the inverter, the transfer switch, wire gauge, mounting infrastructure.
If expansion is in your plan, invest in the right infrastructure now. Get the 10-circuit transfer switch even if you’re only using 6. Run wire sized for your eventual system. The marginal cost upfront is nothing compared to the cost of retrofitting later.
A fixed number is a real constraint — design within it. A flexible number with a clear ROI threshold is a different kind of constraint. Know which one you’re working with and be honest about it.
Buying all at once is usually more efficient per component. Building incrementally lets you learn before committing the full investment. Either works — as long as an incremental build uses infrastructure sized for the eventual system.
You can do this solo. But a partner — even someone who just holds panels or reads a multimeter — makes physical work faster and safer. Explaining your wiring plan out loud to another person is one of the best ways to catch mistakes before they cost you.
If you’re working solo, the DIY solar forums fill the same role. Post your plan. Get feedback. Catch issues early.
Be honest with yourself:
In every case: DC circuits from solar panels are live whenever the sun is shining. You can’t switch them off at the source. Respect that.
Owners have full latitude. Renters: your options are narrower but real. A portable panel and battery station works on a patio or balcony — you get hands-on experience and real backup capability without permanent installation. Some renters negotiate with their landlord about non-penetrating systems. Worth asking.
If you’re planning to sell in a year or two, a permanent rooftop installation is harder to justify. If you’re staying five-plus years, the payback math works strongly in your favor. The longer you stay, the more value you extract from the upfront investment.
If your tenure is uncertain, design for portability. Ground-mounted panels, modular battery systems, plug-in connections where possible. You trade some efficiency for the ability to take it with you.
You know what you want to accomplish, what you need to power, and what constraints you’re working within. Now let’s figure out what you’re actually using.
Know Your Numbers — because the utility monopoly benefits from your ignorance. Knowing your real consumption is how you stop overpaying and start designing a system that actually works.
DATA SOURCED FROM: Section-01-Define-Goals source document (primary). Production estimates reference NREL PVWatts methodology. System cost benchmarks derived from source material. Rate structure guidance based on typical utility TOU pricing structures.