In residential solar the benchmark you want doesn't exist — so the model builds your baseline first

Every other vertical in this library has a published media benchmark to argue with. This one does not. There is no credible published Google Ads cost per click, click-through rate, conversion rate or cost per lead for residential solar, no Meta cost per lead, no local search benchmark, no lead-to-sale rate, no website conversion rate and no speed-to-lead research. What exists instead is unusually good market data: a 205% year-on-year increase in homeowners actively working with an installer, an all-time high in customer enquiries, an average system price of $2.49 per watt, and a battery attachment rate of 38% nationally that hides a range from 71% in California to 53% in Texas.

Every number an installer is shown as a solar benchmark is either borrowed from another category or drawn from a vendor's own book of business, and the second kind is rarely labelled.

1. 01

   Name the missing set out loud, because somebody will fill it

   No published cost per click, click-through rate, conversion rate or cost per lead for solar. No Meta cost per lead. No local search benchmark. No lead-to-sale rate. No website conversion rate. No speed-to-lead study. Writing that list down is the most useful thing anybody can do for an installer being sold a media plan, because the plan will contain figures and they will have come from somewhere.

2. 02

   The proxy is wrong in a knowable direction

   Home and Home Improvement at $8.33 per click and $90.92 per lead is the closest defensible proxy, and it is a proxy. It pools same-week, low-consideration jobs with a purchase that is financed, permitted, roof-dependent and takes weeks to close. Used as a floor it is useful. Used as a target it makes a normal solar funnel look broken and invites the wrong correction.

3. 03

   Battery attachment is a state-level variable and the national average describes nobody

   38% nationally and falling from 41%, against 71% in California and 53% in Texas. A national creative and offer strategy sells the second product to the wrong half of the market, and a battery price rising 3.6% to $1,074 per kWh makes the attach decision a margin question as well as a messaging one.

4. 04

   The subsidy conversation has to be rewritten, not retired

   With the residential credit under Section 25D expired at the end of 2025, every page, script and advertisement built around a tax credit now says something untrue. The replacement conversation is payback period and bill displacement. It is a harder sell, it survives policy changes, and the content debt of not rewriting it is larger than most installers realise.

5. 05

   A baseline is cheap and nobody has one

   Enquiry to qualified, qualified to assessed, assessed to proposed, proposed to signed. Four rates, all measurable inside a month on an installer's own CRM, none of them published anywhere. An installer with those four numbers is better informed about their own business than any benchmark could make them.

No published solar CPC, CPL, lead-to-sale, site-conversion or speed-to-lead benchmark. What exists: record enquiry volume and a battery attachment rate that varies **40–50 points** between states (CA ~69%, HI 80%+ vs Southeast 10–20%).

The first deliverable is a measurement baseline rather than a campaign. Week one installs source capture on every enquiry, call tracking, and stage tracking from enquiry through to signed contract, plus battery attachment recorded per market rather than in aggregate. Proxy benchmarks are written into the plan with the word proxy attached and a date by which they are replaced. From month two the installer's own rates become the plan's inputs and the proxies are removed entirely. Content work runs in parallel to rewrite the subsidy conversation into payback and bill displacement, because the existing pages are actively wrong rather than merely dated.

1. 1

   Week 1

   Source and stage capture

   Every enquiry carries a source; four funnel stages tracked in the CRM from day one

   Owner: OmniFlow

2. 2

   Weeks 1–2

   Market segmentation

   Battery attachment, offer and creative separated by state rather than run nationally

   Owner: OmniFlow

3. 3

   Weeks 2–6

   Content rewrite

   Subsidy-era pages and scripts rebuilt around payback period and bill displacement

   Owner: OmniFlow

4. 4

   Month 2

   Proxy replacement

   Proxy benchmarks removed from the plan and replaced with the installer's own four rates

   Owner: OmniFlow

5. 5

   Week 12

   Review

   Four funnel rates, cost per signed watt, battery attach by market, no external benchmark in the report

   Owner: OmniFlow

The model's output at ninety days is a set of numbers that did not previously exist for this installer: enquiry to qualified, qualified to assessed, assessed to proposed, proposed to signed, cost per signed watt, and battery attachment per market. None of those have a published benchmark to compare against, which is the reason the model builds them rather than borrowing them. Where a proxy is used before the baseline exists, it appears with the word proxy attached and a replacement date, and it is removed from the reporting entirely from month two.

Modelled. Inputs: EnergySage, Marketplace Report #22 (data July – December 2025); Wood Mackenzie, US residential solar customer acquisition costs, March 2026; WordStream by LocaliQ, Google Ads Benchmarks 2026 (13,474 US search campaigns, April 2025 – March 2026). Modelled outputs are not a forecast or a guarantee of results. Cited: Wood Mackenzie / SEIA state storage-attachment data.

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest point is structural: the model is built on the absence of data, so its honesty depends entirely on the installer actually completing the instrumentation. If the baseline work slips, the proxy quietly becomes the target and everybody forgets it was borrowed from a different category, which is the exact failure this model exists to prevent. The second weakness is that the marketplace data comes from a panel of homeowners already shopping, so the enquiry-volume figures describe an active market rather than the whole one.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.