27 of 87 solar software products publish a price

Method

On 28 September 2026 we loaded the pricing surface of every product in the dataset — 231 URLs across 100 products — and recorded only what each page itself stated. Roundups, review sites and third-party summaries were not used as a source for any price.

The control, and why it matters

A “no price published” result is worthless unless the method is known to return prices when they exist. So the pass opened with a control: two products whose records said they publish prices, and one that said it does not.

Because the method demonstrably returns prices where prices exist, every subsequent “not published” is evidence of absence on the page rather than a fetch failure.

Seven kinds of non-answer, kept separate

Collapsing these is how a directory publishes false facts. Only the first is evidence about a product; the rest are facts about the vendor’s website.

• 200, no price on the page — the vendor does not publish a price
• HTTP 404 — the URL is stale. Says nothing about pricing
• HTTP 403 — a bot block. Says nothing about pricing
• DNS, no A record — a broken domain for users too
• TLS chain failure — a site defect
• JavaScript shell — our capture method’s limit, not the product’s
• Login wall — capability cannot be publicly verified at all

Finding 1 — there is no single unit to compare on

12 distinct pricing models are in use across this market. Normalising them to a “starting price” column would misrepresent most of it.

A per-seat subscription, a per-project fee, a per-plan-set deliverable and a flat monthly fee plus a per-closed-deal success fee cannot be reduced to one number without inventing an assumption about the buyer’s volume. We refuse to invent it.

Finding 2 — who publishes, and what

Prices are reproduced exactly as each vendor writes them, in the vendor’s own currency and unit. Nothing is converted — a converted figure would be a number we invented, and exchange rates move.

Finding 3 — some vendors contradict themselves

Pricing pages that disagree with themselves were found during the same pass. These are recorded neutrally, as observed, with the date.

• Arka 360 MAJOR INCONSISTENCY: Basic is simultaneously $199, $159 and INR 11,500; Premium $358, $199 and INR 16,500. Enterprise reads 'Includes 1 user'. A Lottie-animation pricing table from another product renders on the page. Footer (c)2025, webinar dated Jan 2025. Both Allen TX and Bengaluru addresses; no legal entity named.
• DNV SolarFarmer Strongest head-to-head claim in the dataset: 'standard deviation decreased from 3.2% using PVsyst to 2.3% when using SolarFarmer'. Also 'yield accuracy within 0.1% of site measurements'. Internal inconsistency: 2.4GW-years vs 3.6 GWp validation.
• Enerflo Claims '150+ native integrations', names 7. US-only by explicit policy. Self-contradicts on one page: '$16B+' vs '$15 Billion' transactions.
• PV*SOL premium Richest integration record in the dataset. Documents REMOVED integrations (OpenStreetMap, Bing direct, Blender) — the change schema needs a 'removed' state. Release notes to 22 Sep 2026. Two different DB counts on one page.
• Scoop Solar Claims 'a proven 30 to 50X return on investment' with no methodology — untestable. Two different toll-free numbers on one page; app identifiers across four brand names.
• SurgePV RECUSED: publisher's own product. CONTRADICTION: prices in $ but carry +18% GST, an Indian tax, while the entity is described elsewhere as a UK company. Page states 'Last updated: March 1, 2026'.

Finding 4 — free is a floor, not a funnel

10 products price at zero, and a further set offer a permanent free tier. Several are published by hardware vendors and distributors, where the software is paid for in equipment margin rather than subscription. See which products are genuinely free →

Limitations

• This is a snapshot of one day. Vendors change pricing without notice, and several already had between our two passes.
• 10 products could not be read at all and are excluded from every percentage rather than counted as non-disclosing.
• “Publishes a price” means a figure appears on a public page. It does not mean the figure is what a buyer will actually pay.
• Products are counted once, by their primary category. Several serve more than one.
• We did not contact any vendor. This measures what is public, not what is obtainable.