How We Rank Solar Installers

Updated October 29, 2026 · By the Solar Installer Guide team

Use our rankings to find the most active installers in your town, then check the license and compare quotes. The order comes from recorded grid connections, not a judgment of workmanship or customer service. Here is what we count, how we handle the records, where the method falls short and how we make money.

The short version

We rank installers by residential solar systems connected in your town during the latest twelve months of data.
License numbers group company name variants, while ZIP codes help us assign projects to towns.
Connection counts describe activity; they cannot tell you how well a company will handle your project.

We count connected residential solar systems

Our source is the public California DG Stats interconnected-projects data for PG&E. We use residential photovoltaic systems in Fresno, Madera and Kern counties. A system counts when its interconnection is approved, at the permission-to-operate stage.

The source records used here run from January 2020 through August 2026.
Rankings use September 2025 through August 2026, the latest twelve months in this dataset.
Trend columns use thirty-six months of records to give a longer view of activity.

A solar sales agreement and a live system are different stages of a project. We use the date PG&E approved the interconnection, so our count reflects systems that reached the permission-to-operate stage. It is not a count of sales appointments, proposals or contracts. This gives you a common milestone for comparing the installers listed in the utility records.

The ranking is a snapshot of a defined period. Older connections help supply the longer trend view, but they do not add to the latest twelve-month ranking total. Read the dates alongside the counts, especially if you are comparing our tables with another source using a different period. A change in the period can change the order even when the underlying records agree.

ZIP codes determine the town assignment

We map each ZIP code to the Census place it overlaps most, using the 2020 ZCTA-to-place file. This also assigns unincorporated places. Where that mapping does not supply a place, we use the mailing city on the project record.

Town counts follow a ZIP-based assignment rather than a check of each street address.
East Niles and Hillcrest share Bakersfield ZIP codes and are counted in Bakersfield.

A ZIP code can cross a town boundary, so this method is an approximation of local activity. We assign it to the Census place with the greatest overlap rather than splitting individual projects among places. Keep that boundary in mind when reading a town's total. The table follows the same geographic rule across installers rather than treating each company's service area as the town boundary.

The Bakersfield total includes East Niles and Hillcrest because they share Bakersfield ZIP codes. They do not get separate counts under this method. If a project cannot be assigned through the ZIP-to-place mapping, its mailing city supplies the fallback. These choices explain why our town totals can differ from a count based on street addresses or another set of boundaries.

License numbers keep company names together

We group records by California contractor license number, known as a CSLB license. Name variations filed under the same license count together. We display the most common name on those filings so spelling differences do not create separate entries.

Local connections and share of town show activity within the assigned town.
Region installs, battery share and leased or PPA share add context about recorded projects.
Active since reflects this dataset, which starts in January 2020, rather than a company's founding date.

The table brings together different views of the same recorded activity. Installs in town and share of town describe local connections. Region installs give a wider view. Battery share shows the mix of systems with storage, while leased or PPA share describes ownership arrangements. PPA means power purchase agreement. These columns describe past projects; none tells you which system or payment choice fits your household.

The CSLB license column links to the CSLB lookup so you can check the license yourself. A recorded license number is an identifier for grouping data, not our endorsement of a contractor. Likewise, active since means activity visible within records beginning in January 2020. It does not establish when a business opened. Counts measure activity rather than workmanship, so check the license and compare quotes.

Reported price medians use a narrower sample

Price figures are reported medians of Total System Cost per watt for customer-owned systems without a battery. We keep reported values between $1 and $8 per watt. A town needs at least 20 qualifying systems for its own median.

If a town has fewer than twenty qualifying systems, we show the regional figure instead.
Leased systems often have no reported price, so they do not provide the same price information.

The median is the middle reported value in the qualifying sample. It describes those records, not a proposal for a new project. The customer-owned, solar-only filter keeps the price comparison focused on that type of system. A system with a battery falls outside this price sample even though it can still count in the installer activity rankings and battery share.

The town threshold helps you see when a local sample is large enough for a town figure under our method. Where we use a regional median instead, read it as regional context. Neither figure prices your home. Roof conditions, equipment and the scope of work need a project-specific review by an installer. Ask for a written proposal rather than multiplying a reported median into an expected home price.

The records have limits you should know

Connections measure activity, not workmanship or customer service. The installer of record can be a subcontractor working for a sales company. New-home builder systems are included, so a large count does not necessarily describe projects for owners of existing homes.

Some installers work mostly on new homes; we add a note where we know that.
Public data arrives with a lag, so the latest visible connection is not necessarily an installer's latest work.
Missing lease prices limit what the reported price sample can tell you about the whole market.

Ask who would sell the project, who would perform the work and which license would cover it. The name in a utility record may belong to the installing subcontractor rather than the company a homeowner first spoke with. The records alone cannot explain that relationship for your proposed job. Use the company identity as a starting point for questions, then check the license and compare quotes.

Builder projects belong in the activity count because they are connected residential systems. Still, that work can differ from adding solar to an existing roof. Read any new-home note alongside the totals and ask about projects like yours. The same caution applies to battery and ownership shares: they describe the recorded mix, without proving how a company will manage your particular installation or later service needs.

Referral fees fund the site

Installers in our network pay a referral fee when a homeowner we introduce goes solar with them. Solar.com links may also earn a commission. Neither payment changes the order: rankings are set only by the data.

Network partners get a quote button so you can request an introduction through the site.
A quote button identifies a referral relationship, not a higher assessment of workmanship.

The ranking table and the referral feature serve different purposes. The table shows recorded activity; the button gives you a way to contact an installer in our network. Read the count and the project mix on their own terms. The presence of a button does not add connections to a company's record or change the order produced by those connections.

Solar Installer Guide is a comparison and referral service, not a contractor. We do not sell, price or install solar. An introduction gives you a way to begin a conversation about your home, and an installer must review the project to provide a proposal. You still need to check the license and compare quotes before deciding whom to hire.

Updates follow new data, and corrections are welcome

We update when new public data is published. The current ranking window ends in August 2026, so it cannot capture every more recent connection. Installers and readers can send corrections through the contact page.

Read the data period as part of every comparison. Publication lag means an installer can have recent activity that is not visible here yet. A ranking based on this dataset should not be read as a live project counter. The latest twelve-month window and the longer trend columns answer different questions, and both depend on the dates available in the public release.

If a name, license grouping or town assignment looks wrong, use the contact page to describe the issue. Installers and homeowners can both raise corrections. Pointing to the record or table entry helps explain what needs checking. The method above gives you the basis for reading the figures and identifying a possible mismatch, while the public source remains the basis of the comparison.

Frequently asked questions

Does most active mean best workmanship?

No. Most active means the most recorded grid connections during the ranking window. The records do not measure workmanship, customer service or how an installer will handle your project. Use the count to start your research, then check the license and compare quotes before choosing a company.

Why might the installer name differ from my sales company?

The utility's installer of record can be a subcontractor for the company that sold the project. We group records by CSLB license number and display the most common filing name. Ask who will perform your work and under which license, rather than assuming the sales name identifies the installing contractor.

Why does a town show a regional price median?

We use a town median only when at least 20 customer-owned systems without a battery pass the reported price filter. Otherwise, we use the regional figure. These are reported medians from a defined sample, not a price for your home. Only an installer can price your proposed project.

How can I report a possible error?

Use the contact page to identify the table entry and explain the possible error. Readers and installers can both send corrections. Remember that counts use ZIP-based town matching and the stated data window, so a different geographic boundary or more recent project can explain an apparent difference.

Get quotes from up to 3 installers that work in your town

Pick the installers you want to hear from. They review your roof and bill and send their own quotes. Free, and no obligation.

Get up to 3 quotes