Why a roofing permit is a better solar lead than a solar permit
In San Francisco, roofing permits outnumber solar permits roughly six to one, and unlike a solar permit, a roofing permit means no installer has been chosen yet. The precursor pool is the larger, uncontested half of the market.
The obvious record is the contested one. The lead worth having is the record that comes immediately before it.
A solar permit means you already lost
By the time a solar photovoltaic permit is filed with a city, the homeowner has chosen an installer, signed a contract, and in most cases paid a deposit. The permit is the paperwork at the end of a sales process, not the start of one.
That does not make it worthless, it is a competitive-displacement lead, and it tells you exactly who is winning work in your territory. But it is the hardest possible conversation to open, and every other installer reading permit data is looking at the same row.
The pool one step upstream is six times larger
Roof replacement is the single most reliable precursor to a solar install. The structure has been assessed, the envelope is sound, and the owner has just demonstrated both the willingness and the means to spend on their property. Critically, no solar installer has been engaged.
The gap is not marginal. Over a twelve-month window in San Francisco, the city's own permit dataset shows roofing permits outnumbering solar permits by roughly six to one.
Roofing permits against solar permits filed in San Francisco over twelve months
Source: DataSF, Building Permits dataset i98e-djp9
Absence is the signal, not presence
The useful query is not "who filed a roofing permit", it is "who filed a roofing permit and has no solar on record at the same property". That requires resolving records to a property rather than treating each filing as an isolated row, which is why property identity matters more than it first appears.
Once records are grouped by parcel, a set of absence signals becomes available that no single filing contains: a roof replaced with no array, a service panel upgraded with no load added, storage installed with no generation in front of it.
- ·Roof replaced in the last 18 months, no solar at the parcel, the structure is sound and the spend appetite is proven.
- ·Main panel or service upgraded to 200A or 400A, no solar, the most common physical blocker has just been removed at someone else's expense.
- ·Battery storage permitted with no generation, the economics only close properly with an array in front of it.
- ·EV charger installed, a load increase that changes the payback maths, at a household already electrifying.
The same logic runs through every trade
This is not a solar-specific trick. A solar permit on a roof with no replacement on record is a roofing lead, because the array will have to come off and go back on within the decade. A service upgrade with no mechanical work is a heat-pump lead. A deed transfer precedes almost everything.
In each case the precursor pool is larger than the direct pool and is not yet being worked, because the trade that benefits is not the trade that filed the permit.
How far back should a roofing permit count as a solar precursor?
Roughly eighteen months. Beyond that the owner has usually moved on to other spending, and the urgency that made the roof worth replacing has passed. Inside six months is strongest.
Does this work outside San Francisco?
The ratio varies by city and by roof stock, but the direction does not. Any metro where solar adoption is below saturation will have more roofing permits than solar permits, because every house has a roof and only some have arrays.