Quote audit · the canonical guide

How to read a solar quote.

Updated May 16, 2026. Sources verified the same day. We rerun this guide every quarter against the IRS, EIA, LBNL, and NREL reference pages.

A residential solar proposal is a deceptively short document for the size of the financial commitment it represents. Six pages, two pricing columns, a glossy production chart, a payback figure that lands as if it were a quote from a Bloomberg terminal. Most of it is window dressing. Five numbers on the page actually decide whether the math holds up — and one number isn't on the page at all.

Here's where to find each one, what it's really telling you, where it can mislead, and how to test it before the signature page. Everything below is sourced — see the footnotes at the bottom, all verified against the primary source on May 16, 2026.

1. The system size, in kilowatts DC

Usually labeled something like "System Size: 8.4 kW DC" on the first page. Kilowatts DCis the panel-side rating — the sum of every panel's nameplate watts, divided by 1,000. A proposal might also list a kilowatts ACfigure, which is the inverter-side maximum after conversion losses. Both numbers matter, but the cost-per-watt comparison anyone hands you (yours, the salesperson's, ours) is computed off the DC number.

Where it misleads:system size is a function of your roof and your usage, not the installer's preference. Oversized systems pad commission and may push generation past what your utility's net-metering policy actually credits at retail rate. Undersized systems leave bill savings on the table. Berkeley Lab's 2023 sample¹ shows the national median residential install landed at 7.4 kW — useful only as a reference point. The right answer for your home depends on what your meter actually pulls each year.

2. The cost per watt — cash basis

Take the cash price (we'll get to the financed price separately) and divide by the DC system size in watts. An 8.4 kW DC system at $31,800 cash comes out to $3.79/W. That's the single cleanest cost comparator anyone can hand you. Berkeley Lab's Tracking the Sun 2024 report¹ documents 2023 state-level median residential installed prices ranging from $3.20/W to $5.20/W — a $2/W band that varies by state, system size, contractor, and market conditions.

Where it misleads:a $/W figure outside the band isn't automatically wrong, but it earns a question. Below $3.20 — verify the cash price isn't omitting permitting, inspection, or interconnection fees. Above $5.20 — earn the price (steep roof, complex install, premium equipment, high-cost state) or negotiate.

3. The annual production estimate

Stated in kilowatt-hours per year. Something like "Estimated annual production: 10,900 kWh."Two useful tests. First, divide by your system's DC kW to get kilowatt-hours per installed kilowatt — call it production per kW. For 10,900 kWh on an 8.4 kW system, that's 1,298 kWh/kW. Second, compare against NREL's PVWatts²for your ZIP, roof azimuth, and tilt — that's the canonical estimator. For most US residential locations PVWatts lands between 1,000 and 1,600 kWh/kW annually at its default 14% system-losses assumption.

Where it misleads:if a proposal's production lands more than 10–15% above PVWatts for your specific site, the salesperson has assumed either ideal orientation, low shading, or lower-than-default system losses. Each is testable. Note also that PVWatts cannot see your particular trees, chimneys, or future neighbors — the inflated estimate could be honest modeling that's still wrong on your roof.

4. The payback claim

Usually shown as a single confident number: "Payback: 7.2 years." It's the year your accumulated bill savings cross your net system cost. Underneath that one number sit three assumptions, each of which moves the headline year significantly: your annual production estimate (covered above), your current utility rate (look at your last bill), and an assumed annual utility-rate escalator — typically 3–5% compounded for 25 years.

Where it misleads: recent EIA data³ shows U.S. residential rates rose 7.4% year over year through February 2026 — higher than most proposal escalators — but the 10-year trend is closer to 2–3%. Both directions of error are real. If you rerun the payback math with a conservative escalator (2.5%) and an aggressive one (5%) and the headline year shifts by more than two, the original number was a forecast resting on a guess.

5. The tax credit assumption

Usually buried under the net cost figure: "After 30% federal tax credit: $22,260." Read this line as carefully as the cash price. The federal Residential Clean Energy Credit (§25D) ended for systems placed in service after December 31, 2025 per the IRS Form 5695 (2025) instructions⁴. A 2025 federal law change set that placed-in-service deadline; installations completed in 2026 or later no longer qualify for the residential credit.

Where it misleads: "placed in service" means installation is fully complete — not the day you signed, not the day a deposit cleared, not the day a permit was pulled. A contract signed in 2025 with installation slipping into 2026 doesn't qualify. If your proposal still bakes in the 30% on a 2026 install, ask which section of the Internal Revenue Code authorizes it. The credit is also non-refundable — you need income tax owed to claim it. A retired homeowner with $4,000 of annual tax liability cannot use a $9,540 credit in a single year (carryforward helps, but slows the math).

The sixth number — the one that's not on the page

If you're being shown a financed price — total of all loan payments — alongside the cash price, the gap between them is the sixth number, and it's the most quietly informative figure in the document. Berkeley Lab's own language¹ says loan-financed solar prices "likely include dealer fees, adding anywhere from 5–50% to the total up-front price paid by the customer." The financed price isn't just the cash price plus loan interest. It often containsa dealer fee baked into the cash quote, and the loan's headline APR is the mechanism that funds it.

How to test it:ask for the cash-only price separately, divide each by your DC system size, and compare $/W. If the financed price comes out 15–35% higher per watt, you're inside LBNL's documented dealer-fee range. Past 50% and you're past the documented upper bound — ask the loan provider directly for a breakdown of fees versus interest.

Run the audit in ten minutes

The solar quote calculator runs every threshold above against your specific numbers — LBNL state median for $/W, PVWatts range for production, the §25D placed-in-service check for the tax credit, and the dealer-fee gap if you've got a financed price. It takes about 10 minutes start to finish if you have your proposal in front of you.

What to ask the installer before signing

Three questions, each grounded in the numbers above. Read the answers carefully — they're where the math becomes a contract.

1. What's the cash-only price? Not financed monthly. Not after-credit. The number you would write on a check tomorrow, with no loan. Then ask for an itemization: equipment, labor, permitting, interconnection, sales tax. The line items reveal whether the installer is even configured to quote on a cash basis — many financed-volume sellers aren't.
2. What annual utility-rate escalator did the payback model assume — and where did that number come from? Acceptable answers: the installer's spreadsheet default, your utility's recent published rate cases, EIA data for your state. Unacceptable: a shrug.
3. What's the planned date of full installation completion, and what contractual penalty applies if it slips into 2026? The §25D cutoff is the placed-in-service date, not the contract date. If installation slips past Dec 31, 2025, the 30% credit disappears for owned systems. Get the date and the penalty in writing.

What this guide doesn't cover

Plenty. Roof condition and age. Future shading from trees you haven't planted yet. Inverter clipping under DC oversizing. Net-metering policy in your specific utility territory (it changes the payback by years; we'll have a dedicated guide on it). Battery economics versus battery resilience value (different math; also coming). Lease versus PPA versus loan versus cash as a comparative framework. Sales tactics worth walking out on.

Each of those becomes its own guide in this section over the next few weeks. What the page above covers is the numerical core — the five-plus-one numbers that decide whether the proposal's math is honest before any of the qualitative judgment calls come in.

1. 1. Berkeley Lab, "Tracking the Sun, 2024 Edition" (Executive Summary, August 2024; data through year-end 2023). State-level median residential installed price in 2023 ranged from $3.20/W to $5.20/W; loan dealer-fee gap in financed prices documented as 5–50% of the upfront price paid. Verified 2026-05-16. emp.lbl.gov · Tracking the Sun 2024 (PDF) ↩
2. 2. NREL PVWatts (v8) — annual production estimator using NSRDB irradiance data with default 14% system losses. Typical US residential output lands at 1,000–1,600 kWh/kW. Verified 2026-05-16. pvwatts.nrel.gov ↩
3. 3. EIA Electric Power Monthly, Table 5.6.A — Average Price of Electricity to Ultimate Customers by End-Use Sector. February 2026 residential average: 17.65 ¢/kWh, +7.4% year over year vs February 2025. 10-year trend: ~2–3%/yr. Verified 2026-05-16. eia.gov/electricity/monthly ↩
4. 4. IRS Form 5695 (2025) instructions, verbatim: "You can't claim residential clean energy credits for expenditures made after December 31, 2025." The One Big Beautiful Bill Act, signed July 4, 2025, terminated §25D for systems placed in service after that date. Verified 2026-05-16. irs.gov/instructions/i5695 ↩

Next: Run the quote calculator on your proposal's numbers — every threshold cited above is encoded as a verdict.