Carbon & ESG Software SEO: Why Content Beats Authority in AI Search

The product race in carbon accounting and ESG software is maturing fast. Audit-grade data, Scope 3 depth, and agentic AI are becoming table stakes. The race that is still wide open is distribution: who gets found when a buyer searches Google, and who gets named when that same buyer asks ChatGPT, Perplexity, or Google's AI.

This teardown uses live search and AI-visibility data across ten leading platforms to show where the opportunity sits, and what to do about it. Brands appear as examples to learn from, not to rank or rate.

01What is the short version?

The seven takeaways

• Authority is not the bottleneck. Eight of ten sites already carry a domain rating of 63+. More links will not move the needle; more published pages will.
• Content footprint decides traffic. Ranked-keyword counts range from ~250 to nearly 12,000 across near-identical authority. The spread is a publishing choice.
• AI answers cite content, not authority. The two highest-authority domains earn the fewest AI citations; the biggest content library earns the most, by an order of magnitude.
• The buyer terms are cheap and unclaimed. Terms like carbon accounting software and csrd software are low difficulty and high commercial value, and most vendors rank for none of them.
• Paid search is filling the gap. The vendors thinnest on organic content spend most on ads for exactly the terms they could own.
• GEO and SEO are the same work. Structured, answer-first content wins Google features and AI citations at once. One program, not two.
• This is a founder problem too. Proprietary data, a defensible point of view, and engineering time for schema and site health are set at the top, not delegated to a content calendar.

02Why is distribution the new battleground?

When a sustainability lead or a CFO's team goes looking for a platform, they do two things: they Google a category term, and increasingly they ask an AI assistant to shortlist vendors. Both surfaces are won with published, structured content. Neither is won with a better demo or a bigger funding round. So the question is simple: for a category with surging, search-led demand, who has actually built for search, and who is leaving the channel open?

03How did we measure this?

04Is authority the moat? (It isn't.)

Look at the outliers. One well-known enterprise brand carries a domain rating of 73, higher than several competitors, yet sits near the bottom of the traffic range with a tiny keyword footprint, the same authority-without-traffic gap as the field service software teardown. The traffic leaders are the two firms with the largest content libraries, not the strongest link profiles. The lesson: stop buying links and start shipping pages. You have already cleared the authority bar. This is the same pattern behind ranking isn't visibility.

05What actually decides traffic?

06Do AI answers cite authority, or content?

AI engines are not consulting a link-authority score when they build an answer; they are retrieving the clearest, most relevant passage they can find and citing its source. If you have not written the passage, you cannot be the source, the mechanism detailed in how your page gets retrieved.

Where the citations actually come from

Google's AI Overviews and AI Mode are where most citation volume lives, so the same content that wins a Google featured snippet tends to win the AI citation, your existing SEO work compounds here. And breadth matters: the leaders are cited across every engine, while thinner sites get a stray citation on one or two. Being citable everywhere is a function of having enough well-structured pages that every engine finds something to quote, which is why engines recommend different vendors.

How an AI answer gets built

07Are the buyer terms cheap and unclaimed?

08Is paid search a moat?

Paid search has its place, it is fast and measurable, but it is a bridge you cross while the durable asset gets built underneath it. The strategic mistake is treating paid as the strategy rather than the bridge. This is the same trap covered in getting found on Google and AI.

09How do you make your pages citable?

Step 1: let the AI crawlers in

AI engines can only cite what they can fetch. Confirm your robots.txt permits the named AI bots, and do not let an over-aggressive bot wall block them, one hyperscaler sustainability page in this study returned a bot challenge to non-browser requests, which can suppress citations. See how AI crawlers index your site.

# robots.txt - allow the major AI answer engines
User-agent: GPTBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Google-Extended   # Gemini / AI Overviews
Allow: /

User-agent: ClaudeBot
Allow: /

Sitemap: https://yourdomain.com/sitemap.xml

Step 2: publish an llms.txt

An llms.txt file is a plain-text index that tells AI crawlers which pages matter. It is a five-minute file that most of the field has not shipped, more on whether llms.txt actually does anything yet.

# llms.txt - https://yourdomain.com/llms.txt
# Guide for AI crawlers: the pages worth citing.

## Product
- [Carbon accounting software](/product): measure Scope 1-3, audit-ready.
- [CSRD reporting](/csrd): map data to ESRS and file with confidence.

## Methodology
- [How GHG accounting works](/learn/ghg-accounting): plain-English guide.
- [Scope 3 explained](/learn/scope-3): the 15 categories, with examples.

## Proof
- [Benchmark report 2026](/research/benchmark): original emissions data.

Step 3: add structured data (JSON-LD)

Schema is how a page self-describes to machines. The clearest pattern in this study: the sustainability section that ships Product and FAQPage schema already ranks number one for a non-branded buyer term, while a far larger competitor that ships no schema does not. Put Organization, Product and FAQPage on your key pages, the full pattern is in schema markup for AI citations.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "What is carbon accounting software?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "Carbon accounting software measures an organization's
       greenhouse-gas emissions across Scopes 1, 2, and 3, then turns
       that data into audit-ready reports for CSRD, ISSB, and the SEC."
    }
  }]
}
</script>

Step 4: write answer-first

Structure invites the citation; the words earn it. Lead each key section with a direct, self-contained answer of roughly 40 to 60 words that an engine can lift verbatim, then expand below it, the exact shape of a high-citation page.

## What is Scope 3?

Scope 3 is all the indirect emissions in a company's value
chain, both upstream and downstream: purchased goods, business
travel, and the use of sold products. It is usually the largest
and hardest-to-measure share of a company's footprint,
spanning 15 defined categories under the GHG Protocol.

<!-- then expand: the 15 categories, examples, how to measure -->

10What is the technical foundation checklist?

11Why does one good page pay out three times?

12How should the motion match the buyer?

13What does the first 90 days look like?

14Who owns what: founders or marketers?

15Which metrics actually matter?

16What is the takeaway?

For founders, the move is to treat distribution as a product surface: fund the site health, protect the engineering time, and decide which category terms you intend to own. For marketers, the move is to stop renting visibility and start building the compounding asset, one structured, answer-first, buyer-intent page at a time.