Best financed emissions software in 2026: 9 PCAF tools compared

Financed emissions software divides into two things that get sold under one label: platforms that apply the PCAF Standard to a book and produce the reporting, and datasets that supply the emissions figure per counterparty that the calculation needs. The arithmetic is set by the Standard and every serious vendor follows it. What separates the nine tools here is what each one gives you for a holding that has not disclosed, because that is where a book's data quality is decided, and for most institutions it is the private and unlisted book.

Vendor list updated 23 September 2026. Verification status from our benchmark, evidence links last checked 19 August 2026.

How should you compare financed emissions software?

Four questions, in this order.

• What class of data goes in for a private holding? The PCAF data quality score (DQ, 1 to 5, one is best) is decided by the input, not the engine. A sector factor on outstanding amount is DQ 5. The company's own revenue with a sector factor is DQ 4. Production or energy data, or the company's own reported figure, is DQ 3, 2 or 1. A dataset can move a private holding from DQ 5 to DQ 4 on its own; anything higher needs something the company itself produced.
• Can you see how each figure was produced? Reported or modelled, and from which document. Your assurance provider will sample holdings and ask for the link back to the source.
• What happens where nothing exists? A blank row, a modelled figure presented as coverage, or a sector average shown as a sector average.
• Has the calculation been verified by a named third party against a standard? PCAF accreditation is not that: PCAF describes it as access to its database and support in applying the Standard, and neither PCAF nor its accredited partners describe any review of software or calculations.

One more: which edition of the Standard the vendor implements. Part A, Financed Emissions, is in its Third Edition, dated December 2025, which broadened the attribution denominator for listed equity and corporate bonds and recommends an inflation adjustment on the DQ 4 and DQ 5 factors. A good deal of vendor material still describes the Second Edition.

How much of a book sits at DQ 4 and DQ 5?

No credible market-wide figure exists. Two published studies are worth more than an invented percentage. MSCI examined its own private capital data and reported that of more than 58,000 portfolio companies in private capital funds, 1,312 had Scope 1 and 2 emissions reporting, a disclosure rate of 2.2% as at the second quarter of 2023. The Oxford Sustainable Finance Group examined 28 institutions that had disclosed PCAF-based financed emissions and found that none had compiled estimates entirely from directly reported company-level data, and that half used DQ 5; that study is dated January 2023. Both point the same way: the constraint is disclosure, not calculation, so the work that improves a book's score is data work.

The 9 best financed emissions software tools for PCAF reporting in 2026

Ranked by fit for a book with a private and unlisted tail: what each gives you for a holding that has not disclosed, whether the source of each figure is visible, and whether the calculation is verified. Platforms and datasets are ranked together because a buyer has to choose across both.

1. DitchCarbon

DitchCarbon holds verified emissions data for over 2 million organisations, built on primary emissions data wherever it exists, matched to a book by entity resolution against DUNS, LEI and ISIN identifiers, with parent and subsidiary relationships resolved so the figure attaches to the counterparty rather than the group above it. Where a private holding has published an inventory, in an annual report, a filing, a CDP response or a sustainability statement, the figure is that reported inventory with the source document behind it, and it lands high on the hierarchy because of what it is. Each holding records which method produced its figure, so the reported and modelled shares of the book are a filter rather than a reconstruction. Where a holding has not disclosed, a request goes out prepopulated with what can be found about it, and the sector average holds the line, shown as such, until something better arrives. Run alongside the calculation platform you already use, DitchCarbon is the counterparty data feeding it, so the model, the factor set and the published baseline stay where they are and the DQ 4 and DQ 5 tail moves up inside them. The Portal calculator is verified to ISO 14064-3, limited assurance, by UL Solutions, renewed annually.

What you still need afterwards: nothing for the baseline, or for assured emissions calculations. Requests go only to the holdings where a reported figure would lift the score.

2. Persefoni

Persefoni is a calculation platform: it applies the PCAF methodology to the book and produces the reporting, and SOCOTEC International Certification has verified that the platform calculates emissions against the PCAF Standard, which is rare in this market. For a private holding the input is primary collection from the portfolio company rather than a database lookup, so what goes into the verified calculation for an unlisted counterparty is whatever that company sends. Run alongside Persefoni, DitchCarbon supplies the counterparty figures the calculation needs. The verification claim is undated on Persefoni's certifications page; ask for the certificate.

What you still need afterwards: the data for the private book. The calculation is verified; the input for an unlisted holding is whatever the company sends you.

3. S&P Global Sustainable1 (Trucost)

A dataset built on the Trucost research universe, index-constituent based, with a large modelled private company universe alongside it. Where a company has not reported, S&P maps it to business activities, apportions revenue and applies a modelled factor, and its documentation says gaps are filled so that no gaps remain. Under the PCAF ladder that filled gap is DQ 4: a sector average with a company name on it. The record does not separate reported from modelled holdings unless you ask. We checked S&P for our verification benchmark and found no public software-level verification; its PCAF accreditation is not a methodology review. Our S&P Trucost and MSCI comparison works through the reported and modelled shares.

What you still need afterwards: to know which holdings are reported and which are modelled. A filled gap is a DQ 4 however many holdings it covers.

4. MSCI

Portfolio footprinting and climate analytics over a listed universe plus a private capital dataset. Figures are reported where available, otherwise estimated from production or revenue, otherwise a sector average, in MSCI's own order; its factsheet marks private equity and debt funds as economically estimated and unlisted corporates as on-demand coverage. "Where available" is the part that decides the score, and the estimated remainder is produced by a model you cannot inspect. MSCI is not in our verification benchmark.

What you still need afterwards: the share of your book that fell outside "where available", and a source you can show the assurance provider for each holding inside it.

5. ISS STOXX (formerly ISS ESG)

Carbon footprinting that outputs PCAF quality scores directly, over a universe of companies and corporate debt issuers. Private coverage is not published. Alignment with PCAF is stated by ISS; no independent verification of the calculation is published.

What you still need afterwards: the private and unlisted holdings. The universe is issuers.

6. Morningstar Sustainalytics

Carbon emissions data combining reported and modelled figures over a listed universe; private coverage is not published. Alignment with PCAF is stated by Sustainalytics; no independent verification of the calculation is published.

What you still need afterwards: the unlisted book, and the split between reported and modelled for the listed one.

7. Clarity AI

Portfolio emissions and climate risk analytics, with coverage benchmarked to a listed index. Private coverage is not published. Alignment with PCAF is stated by Clarity AI; no independent verification of the calculation is published.

What you still need afterwards: the unlisted book. Climate risk analytics on the listed holdings do not produce a counterparty figure for the rest.

8. Novata

Private markets carbon data collection and roll-up, with a financed emissions output table by asset class. There is no pre-built database: coverage is whatever the general partner collects from its own portfolio companies, and collection is the only path to DQ 3, 2 or 1 for a holding that has published nothing. It works at the scale of relationships a GP can actively manage. Alignment with PCAF is stated by Novata; no independent verification of the calculation is published. DitchCarbon starts from what portfolio companies have already published, so collection goes only where it would change the score.

What you still need afterwards: everything not yet collected, which on day one is the whole book. Coverage is the collection programme.

9. Watershed

An enterprise carbon platform with a portfolio offering, PCAF accredited, with private coverage not published. Run alongside Watershed, DitchCarbon becomes the counterparty data layer feeding the portfolio module. We checked Watershed for our verification benchmark: its corporate calculation methodology was validated against ISO 14064-3 by Apex Companies, and the PCAF accreditation is not a methodology review.

What you still need afterwards: the portfolio data itself. The platform is built for the corporate inventory, and the portfolio coverage is not published.

Verification status from our benchmark, Who's Really Verified? A Reality Check on Carbon Software Assurance, evidence links last checked 19 August 2026. PCAF accreditation is recorded separately because PCAF itself describes it as database access and support, not a review of software or calculations. Other columns reflect each vendor's public documentation as at August 2026.

What actually moves a holding from DQ 5 to DQ 2?

A change in the class of input, not a better model. The PCAF Standard's data quality tables set it out: DQ 5 is a sector factor on the outstanding amount; DQ 4 adds the company's own revenue and the attribution denominator with the factor still a sector average; DQ 3 needs primary physical activity data on production; DQ 2 needs energy consumption by source with specific factors, or the company's own unverified calculated emissions; DQ 1 is the company's verified emissions. A vendor database can move a private holding from DQ 5 to DQ 4. Moving higher requires something only the company has produced, which is why the useful question about any vendor is how many of your holdings it can lift above DQ 4, and by what route. Our DQ 5 to DQ 2 page works through each step and what it costs.

Being private does not cost a holding a point. The PCAF scale for business loans and unlisted equity is structurally identical to the listed one; only the attribution denominator differs. What drives private holdings to DQ 4 and DQ 5 is that private companies disclose less often, so the inputs the higher scores need are missing. That is a data availability problem, and it is solvable in two directions: find the disclosures that already exist, and engage the holdings that have not made any, in that order.

Do you have to replace the platform you run?

No. The model, the factor set and the published baseline stay where they are. DitchCarbon feeds the counterparty figures into the calculation platform you already run, so the reported holdings carry their own figures rather than a sector average and the DQ 4 and DQ 5 tail moves up the hierarchy inside your existing approach, with the same fallback ladder underneath it: a disclosed figure first, a figure modelled from disclosures where there is none, a regionalised industry factor last. Nothing is restated. Run on its own, DitchCarbon is the platform: measure the counterparties, engage them for better data, forecast against each organisation's own target trajectory and track at counterparty, account or portfolio level. Every figure carries its source and change history, so an assurance provider can follow a sampled holding from the number back to what the company published; the calculator behind those figures is verified to ISO 14064-3, limited assurance, by UL Solutions, and the reports are in the trust centre.

If you are a portfolio company being asked

Many companies reading this are on both sides of the request: measuring their own holdings or suppliers, and being asked by an investor or a customer for the same thing. Claim your organisation's profile, check what is already held against your name, and answer once. A reported figure is what lifts your investor's data quality score, and it costs you nothing to provide.

The short version

Financed emissions software is a calculation platform, a dataset, or a collection tool, and the PCAF score of a book is decided by the class of data going into whichever you run. Datasets model the private book to DQ 4 and call it coverage; collection tools reach higher and scale with the relationships you can manage; calculation platforms are only as good as what is fed to them. DitchCarbon finds the disclosures that already exist for over 2 million organisations, shows which holdings are reported and which are modelled, asks only where a reported figure would change the score, and feeds the result into the platform you already run.