How to Read a Solar Proposal: Every Line, Explained (2026)

A solar proposal is built to justify a price. That is not an accusation; it is simply what the document is for. The installer knows what they are recommending, and the proposal is the case for it. Your job is to understand what you are actually being asked to buy, because the language is dense by habit, and the things that matter most are rarely the ones in large type.

You do not need to become an engineer. You need to know what each number means, which ones are easy to inflate, and what to ask when something does not add up. This guide walks through a proposal section by section, and ends with the handful of questions that tell you most of what you need to know.

System size: is it built for your house?

The system size, given in kilowatts of DC, is the number that drives everything else: the production estimate, the price, the amount of equipment on your roof. Get it wrong and every figure downstream is wrong with it.

The sanity check is your own electricity use. Pull twelve months of bills, find your annual kilowatt-hours, and ask whether the system is sized to match. A rough rule: in California a kilowatt of panels produces somewhere around 1,300 to 1,600 kilowatt-hours a year, so dividing your annual usage by roughly 1,400 gives you a ballpark system size to hold the proposal against. Under the current rules this matters more than it used to. The state moved to a billing structure called NEM 3.0 in 2023, and under it the surplus an oversized system sends to the grid is credited at a small fraction of what you pay to buy power back, often around a nickel on the dollar. So oversizing now wastes money on production you will not be paid fairly for, while undersizing leaves savings on the table. Size up only for a real, known future load, an electric car, a heat pump, a new heir to the electric bill, and be skeptical of a system noticeably larger than your history without that reason. Some sales compensation rewards bigger systems, so the incentive to oversize is real even when the intent is not bad. (If you want the difference between the DC size and the AC size explained, our piece on system architecture covers it; for reading a proposal, the DC number is the one to check against your usage.)

Equipment: is it named, and is it real?

Every major component should appear in the proposal by brand and model number: the panels, the inverter or microinverters, the racking, and the battery if there is one. A proposal that says “high-efficiency panels” without a model is either unfinished or hiding something, and either way you should not sign it. Model numbers are not a formality; your utility verifies the exact models against safety listings when you apply to turn the system on, and a mismatch is a common reason approvals stall.

Once the equipment is named, you can confirm it is legitimate, though this is more reassurance than alarm. Major components carry a safety certification, UL 61730 on panels and UL 1741 on inverters, with California requiring the grid-support version, UL 1741-SA. In practice, essentially every panel and inverter sold for rooftop use in the United States is listed; unlisted major equipment is rare, barely available here, and something no installer would risk using, since it would fail inspection and expose them to real liability. So this is rarely a problem, not a chore to lose sleep over. It still costs nothing to confirm the listings appear on the spec sheets, and the one place worth a second look is an unfamiliar racking or minor hardware brand rather than the panels and inverter; the racking names you would expect to see are IronRidge, Unirac, and SnapNrack. If you ever want to check a specific model yourself, UL’s public database at productiq.ul.com will show it. The broader point stands regardless: your building department requires listed equipment and your utility checks it, so anything unlisted surfaces eventually, and unpermitted or non-code work can create headaches with inspectors, your utility, and potentially an insurance claim later.

The production estimate: is the math honest?

This is the number the entire financial case rests on, and the one most likely to be optimistic.

The production estimate tells you how many kilowatt-hours the system is expected to make in a year. Because every dollar of projected savings depends on it, check it against an independent source before you trust it. The free tool the industry itself uses is NREL’s PVWatts, at pvwatts.nrel.gov, run by a national laboratory. (Confirm that address; look-alike sites exist.) Enter your address, the system’s DC size, your roof’s tilt and the direction it faces, leave the system-loss figure at its default of fourteen percent, and it returns an independent estimate in a few minutes. If the proposal’s number sits more than ten to fifteen percent above what PVWatts gives for the same system, ask the installer to show you the specific inputs, the tilt, the direction, the shade study, that justify the gap. A detailed shade analysis from a real site visit, using professional tools like Aurora or HelioScope, is a legitimate reason a number might differ; a vague answer usually means the estimate is padded. A number well below PVWatts is worth a question too, since it may reflect real shading you should know about.

Then there is the assumption that quietly does more work than any other in the document: how fast your utility’s rates will rise. Every projection of lifetime savings has to assume a yearly escalation rate, and small differences compound into enormous ones over twenty-five years. Here is the fact that puts you in control of it: the state allows these savings estimates to assume rate increases of up to ten percent a year, an extraordinarily high ceiling, which is exactly how a proposal can show savings that look too good to be true. So the single most useful question you can ask is what escalation rate the proposal assumed, and then ask to see it re-run at three percent.

What is a defensible rate? The honest answer is that nobody knows the next twenty-five years. The recent past is a caution against both extremes. Over the last decade, California utility rates rose on the order of seven percent a year on average, according to the state’s own ratepayer advocate, and the stretch from 2022 to 2025 was steeper still, above ten percent a year. But rates do not only climb: PG&E cut residential rates more than once in 2025, and customers entered 2026 paying less than they had two years earlier. Treat the bands accordingly. Two percent understates your likely savings; three percent is conservative and well supported, close to the long-run national average; four to five percent is plausible; anything at six and above is aggressive, and eight or more is reaching for the worst years to make the pitch look better than it is.

Finally, ask whether the production estimate comes with a guarantee, a written commitment that the system will make at least a stated amount in its first year, with the installer obligated to investigate and remedy a shortfall. Many installers offer none, and a guarantee is only as good as the company still being in business to honor it. It is a sign of confidence, not a substitute for your own independent check.

The financial analysis: where the real decisions live

This is where the money is, and where the structure of the deal matters more than the headline number.

Your comparison tool is cost per watt. Take the cash price before any incentives, divide by the system size in watts, and you have a figure you can compare across proposals and against the current California market. Compute it on the cash price specifically, because, as you will see, a financed price can have a fee buried in it. A cost per watt well above the local market is not automatically wrong, but it is automatically a question, and the best way to know your local range is to get three competing bids.

How you pay changes everything, and each path has its own thing to watch.

Paying cash is the simplest case: the system is yours and so are the savings. One caution specific to 2026: the federal tax credit for homeowners who buy their own system, Section 25D, expired at the end of 2025. A salesperson or an older proposal still showing a thirty percent credit subtracted from the price of an owned system is misrepresenting the cost. Stop subtracting it.

A loan is where the most documented harm in this industry lives, and it hides in a single mechanism: the dealer fee. To offer a low advertised interest rate, the lender charges the installer a fee, and that fee is folded into the price you finance. The federal consumer-finance regulator found these fees often raise the loan cost by thirty percent or more above the cash price, and frequently are not separated out, which means a headline “1.99 percent” loan can cost you more over its life than a plain seven percent loan with no fee. The defense is simple and it works: ask for the cash price and the financed price, both in writing. The difference between them is the fee you are being charged to borrow. Check for prepayment penalties while you are at it.

A lease turns on its escalator, the clause that raises your payment a set percentage every year. Escalators usually run near one, two, or three percent, and the compounding is brutal over a long term: a payment that climbs three percent a year is far larger at year twenty-five than where it began, and can outrun what you would have paid the utility, especially in a year when utility rates fall. A zero-percent escalator costs more in year one but stays predictable. An escalator is a bet that utility rates keep rising fast, and 2025 was a reminder that they do not always.

A PPA is similar, except you are buying the power rather than renting the system, at a stated rate per kilowatt-hour that usually carries its own escalator. Confirm who owns the system, what the rate does over time, and what happens at the end of the term.

The number that actually matters is none of the ones in large type. It is the total you will pay over twenty-five years, every payment and every escalator included, set against what you would have paid the utility with no solar at all. Run that comparison at a conservative three percent escalation and again at whatever rate the salesperson used. If the proposal will not show you that honestly, that itself is your answer.

Warranties: what is actually protected, and by whom

A solar system carries several warranties from several parties, and they do not all mean what the proposal implies.

The panel typically carries two: a product warranty against defects and a separate performance warranty on output, now commonly twenty-five years on the product and up to thirty on performance from mainstream manufacturers. A ten- or twelve-year product warranty marks a budget panel. Confirm both are present.

The inverter warranty depends on the type. Enphase microinverters carry twenty-five years; SolarEdge runs twelve years, extendable to longer terms for a fee; a conventional string inverter is usually around ten to twelve. Most of these cover the part, not the labor to swap it, which becomes your cost if the installer is gone.

The workmanship warranty is the installer’s promise about their own labor, and the one to scrutinize hardest right now. It is only as good as the company behind it, and the last two years have been brutal: SunPower, Titan, Sunnova, and the financier Mosaic all failed. A long workmanship warranty from a shaky company is worth less than a short one from a stable installer. Check how long the company has actually been licensed, which you can see on its state record.

The roof penetration warranty covers leaks where the mounts pass through your roof, and for how long. Ask about it specifically, and confirm your roof has at least ten more years of life before you put a system on it, or you will pay to remove and reinstall the array when you re-roof.

The hard truth underneath all of these: if the installer goes out of business, the manufacturer warranties on the equipment generally survive, provided the manufacturer is also still in business and you have kept your serial numbers, but the installer’s own workmanship and roof warranties usually do not. We cover that situation in detail in our pieces on orphaned systems and installer bankruptcies.

Before you sign: the questions that reveal the most

These are short and fair, and the way an installer answers them tells you most of what you need to know.

Ask whether they use their own licensed crew or a subcontractor, since subcontracted installs have produced some of the worst outcomes when companies fail. Ask for their license number and verify it yourself at cslb.ca.gov; California solar work should be done under a C-10 electrical license or a C-46 solar license, with C-10 the stronger of the two, and the state record also shows how long they have really been in business. Ask who pulls the permit, because it should be the contractor, which keeps the liability theirs. Ask, plainly, about the company’s financial footing and whether they operate under a national dealer model. And ask to speak with a customer in your zip code whose system has run three or more years. A confident, specific answer to each is a good sign. Evasion is also an answer.

Two protections the state already gives you are worth knowing before you sign. A solar contractor cannot legally take a down payment of more than one thousand dollars or ten percent of the contract price, whichever is less. And you have the right to cancel, three business days for most buyers and five for homeowners aged sixty-five and older. (Confirm these against current law, which can change.) In the major utility territories the installer must also give you the state’s solar consumer guide and collect your signature before connecting the system; skipping it is a bad sign and is reportable.

What the proposal often leaves out

A finished installation is not yet a working system. Your utility has to grant Permission to Operate before you can legally turn it on, and that step is separate from the city’s inspection and outside your installer’s full control. Solar-only systems commonly take a month or two after install; battery systems take longer. The state’s solar trade group has documented utilities missing their own connection deadlines a large share of the time, so do not believe “next week,” and do not tie your final payment to a date the installer cannot guarantee.

A proposal also rarely dwells on the rate plan you will be moved to after going solar, usually a time-of-use plan on which the hour you use power matters as much as the amount, or on new fixed charges like the monthly base charge PG&E introduced in 2026 that you pay regardless of how much you produce. And it will rarely volunteer that under NEM 3.0 a solar-only system now exports most of its midday production for very little, which is why a battery, by storing that production for the expensive evening hours, has gone from a luxury to close to a necessity for strong economics in California. If a battery is on your proposal, ask to see the payback with and without it. If one is not, ask why.

What this guide is, and is not

This teaches you to read your proposal and recognize where proposals tend to mislead. It stops short of telling you whether the specific deal in front of you is a good one, because that depends on your roof, your usage, your rates, and the exact terms on the page. If you want a personalized read on the proposal you are holding, that is what the practice’s proposal review is for.

The Installer’s View is an independent solar advisory practice for California homeowners. We do not sell, install, or service solar equipment, and we are not paid by any manufacturer or lender named here. This article is general educational information, not legal, financial, or engineering advice. Figures, programs, warranty terms, and laws change; verify current details before making a decision.