AI Visibility Index

What this index is

The AI Visibility Index ranks companies inside high-demand AI, software, SaaS, fintech, and tech categories. Each board lists about 15 companies with a transparent 0 to 100 score. The index has 23 boards.

This is not a paid directory. Scores come from documented public sources scraped during the data week of 3 November 2026 (scraped 5 November 2026, Europe/Warsaw). Austin Heaton may advise brands in these markets. That relationship never buys a rank. Listed companies are not clients by default. Unlisted companies are not bad. They may simply fall outside this sample.

The index is independent research for austinheaton.com. It is not a white-label feed from any other AI visibility product.

How scores work (v1)

Version v1-listicle-review. Two pillars only.

Pillar	Weight	What we count
Listicle consensus	60%	Share of scraped independent best-of pages that name the vendor
Review standing	40%	Public G2 (and similar) category footprint mapped to documented bands

The formula is:

• listicle_score = 100 times (listicle mentions divided by the row denominator)
• review_score = the documented band for that vendor
• composite = round(0.6 times listicle_score + 0.4 times review_score)

AI-answer presence is omitted in v1. ChatGPT, Claude, Gemini, Perplexity, and Google AI Overview citations will be added only after raw answers are captured and saved as an audit trail. Until then this index does not invent weekly AI percentages. A citation-surface pillar is also deferred. When answer data is added, weights will be republished and historical composites will be versioned so readers can see the break.

Limits: listicles disagree, review sites lag, and brands with quiet public relations can be under-counted. Read the methodology before you treat a one-point gap as decisive.

Methodology

Last updated 5 November 2026 (Europe/Warsaw). Data week of 3 November 2026. Version v1-listicle-review.

Purpose

Rank companies inside high-demand AI, software, SaaS, fintech, and tech categories using public evidence that other buyers and writers already publish. The research steps are:

1. Search for independent "best [category] 2026" pages and category review hubs.
2. Scrape main content. This release used Firecrawl search and scrape.
3. Count whether each candidate vendor is named (listicle consensus). A hit is the company name or a documented alias in the page main content.
4. Map public review standing into bands.
5. Compute the composite with the 60/40 weights above.
6. Publish dated boards with sources. Refresh monthly. Each board shows the data week and the scrape time.

Review bands

Bands are assigned from public G2 category pages and well-known editorial security and software reviews. They are qualitative mappings, not scraped star averages. Consumer and business finance boards (online banks, crypto cards, business bank accounts, payment processors, and corporate cards) do not have a single G2 grid for every vendor. Review bands on those boards are assigned from app store rating standing and well-known editorial bank and card reviews (Bankrate, NerdWallet, Forbes Advisor, CNBC Select), in addition to G2 where a category exists. The bands stay qualitative mappings, not scraped star averages.

Band	Score	Meaning
leader_high	95	Clear category leader with a large public review footprint
leader	88	Consistent leader or top-grid presence
high	80	Strong public review presence
solid	70	Regularly reviewed, mid-grid
moderate	58	Present but not dominant
emerging	45	Newer or thinner public review evidence
thin	32	Sparse public review evidence
na	40	No clear category page signal; neutral placeholder

Name matching

A listicle hit is the company name or a documented alias in the page main content. Ambiguous brand words (Current, Close, Spark, Found, Discover, Fold, Mesh, monday) are matched only on qualified aliases, for example Close CRM, Spark Mail, or Discover Bank, so an ordinary use of the word does not count as a hit.

Pillars not used in v1

Planned pillar	Status
AI answer presence (target 40% when active)	Omitted in v1. No raw answers were captured and saved this session.
Citation surface (target 15%)	Deferred to a later refresh.

What we do not do

• Invent AI visibility percentages or weekly tracker numbers without raw answers on file.
• Sell placements or let payment change ranks.
• Put Austin service prices on these pages.
• Claim a one-point score gap is statistically decisive.

Disclosure

Austin Heaton advises B2B and SaaS brands on AI search and answer visibility. That work can involve companies that appear on these boards. Advisory relationships do not purchase ranks. Sources are listed on each board. Fintech boards are not financial advice. Rates, fees, rewards, and regional availability change often and vary by country.

Changelog

• 5 November 2026: v1 launch dataset. 15 boards, about 15 companies each. Weights: listicle 60%, review 40%. AI-answer pillar omitted because no raw answers were captured. Hub path /ai-visibility-index.
• 5 November 2026: Wave 2 adds 8 boards (online banks, crypto cards, payment processors, AI email assistants, business bank accounts, corporate cards and expense management, payroll software, and CRM software). The index total is 23 boards. Finance review bands and the qualified-alias rule are documented above. Formula unchanged (v1-listicle-review). AI-answer pillar still omitted.

Frequently asked questions

If you need mention rates inside live AI answers, that is a different measurement. Start with Measuring AI visibility. This index does not provide those rates yet.