AI Search Audit · AEO · GEO · Sample report
AI Visibility Audit
Industry context, AI search performance, competitive position and growth opportunities for clientdomain.com, a B2B company building infrastructure for AI agent commerce.
Scale 0 to 100 across ChatGPT, Claude and Gemini. Band: Emerging.
About the audit
Is your brand part of the answer when buyers ask AI?
B2B buyers now research vendors inside ChatGPT, Claude, Gemini and Google AI Overviews, and form their shortlist before they ever visit a website. The BuzzIQ Labs AI Visibility Audit shows you exactly where your brand is named, recommended or missing in those answers, who is winning instead, and what to fix first.
Hand-built query panel
50 buyer-style queries across awareness, consideration and decision, run live on 4 AI platforms and coded answer by answer.
AI Visibility Index
A 0 to 100 score with mention, recommendation, position and citation data by platform and funnel stage.
Competitor share of voice
See which vendors set the default answer and how AI frames every comparison with your brand.
A 90-day roadmap
Prioritized actions, timing and KPI targets, re-tested at weeks 6 and 13 against your baseline.
Report contents
This sample shows every section of a real audit. The client is anonymized; competitor names and figures are unchanged.
01 · Executive summary
Where the client stands in AI answers
The client wins when buyers already know its name. It is largely absent when buyers are still defining the category, which is where shortlists form.
AI Visibility Index across ChatGPT, Claude and Gemini (Google AIO: mention rate only)
Non-branded answers on 4 platforms that name the client (24 of 108)
Branded answers on 4 platforms that name the client: strong once buyers know the name
Google AI Overviews name the client on 3 of 27 non-branded queries (11.1%)
Labels use the report bands: Good 50+ (branded mentions 90%+), Average 25–49 (70–89%), Poor under 25 (under 70%). Benchmark: Coinbase / x402 and Stripe 54.3% share of voice, the client 25.9%.
What we found
Known-name strength
The client is named in all 56 branded-comparison answers and 34 of 35 brand-specific answers across the four platforms.
Category discovery gap
On awareness (TOFU) queries the client appears in 2 of 40 answers (chat AVI 3.3). Visibility peaks at consideration (MOFU, chat AVI 33.3).
Content exists, citations lag
Published articles cover protocols, identity and bot access with FAQs, yet 0 of 4 platforms named the client for protocol-comparison or bot-control queries.
Competitors frame the comparison
ChatGPT and Claude both recommended Nevermined over the client for pay-per-use API billing. Coinbase / x402 and Stripe each appear in 54.3% of non-branded chat answers.
Non-branded mention rate by platform
Answers naming the client ÷ 27 non-branded queries per platform (5, 9, 7 and 3 queries). Source: BuzzIQ Labs query panel.
02 · Industry overview
Agentic commerce: a new buying interface
Analysts measure different things, so treat the figures as directional. All agree that AI agents are becoming a buying interface, and that the infrastructure that verifies and pays for them is a new layer.
US agentic commerce by 2030: analyst ranges (USD billions)
Sources: McKinsey, The Agentic Commerce Opportunity (Oct 2025); Morgan Stanley, Bain and Gartner as compiled by Stellagent, Hexagon and MetaRouter; ICSC / McKinsey via Retail Dive (May 2026).
McKinsey: global agentic commerce opportunity by 2030, including up to $1 trillion of US B2C retail revenue
Gartner: B2B spend that could flow through AI agent exchanges by 2028
Morgan Stanley: share of US e-commerce that agentic channels could capture by 2030
B2B buyers using AI in vendor research (%)
Sources: G2, The Answer Economy, Mar 2026, n=1,076 (via Demand Gen Report); Forrester 2026 Buyers’ Journey Survey; Gartner 2026 (via SuperMarketers).
03 · Customer understanding
Who buys, who decides and what hurts
The ideal customer profile and personas below are a BuzzIQ Labs synthesis of the client’s public materials, to be validated with the client’s CRM. The company serves both sides of the agent economy: teams that build agents and the services that sell to them.
Positioning
Identity, access and payment infrastructure for AI agents.
Name collision
An unrelated company with a similar name exists. On the identity-check query, Claude separated the two; ChatGPT and Gemini answered about the wrong company; Google merged them.
Primary market
United States (English), where Semrush shows the strongest AI visibility signal. Secondary: Europe and APAC developer and fintech ecosystems.
Ideal customer profile
| Segment | Who they are | Buying trigger |
|---|---|---|
| Agent builders | AI-native startups and developer teams shipping autonomous or shopping agents | Agents blocked by captchas, login walls and human-tied billing |
| API, data and MCP providers | Services that want to sell access to agents | Monetise agent traffic without bank-account onboarding |
| SaaS platforms and marketplaces | Platforms opening sign-up and checkout to agents | Verified agent sign-up and agent-to-agent payments |
| Merchants and commerce platforms | Retail and travel brands accepting agent-initiated orders | Verify delegated authority before charging a card |
| Enterprise risk and security teams | Owners of identity, fraud and payments governance | Audit trail, spend control and revocable authority |
Priority verticals: AI-native SaaS, data and API providers, commerce platforms, payments and fraud.
Buyer personas and decision-makers
AI agent developer
RoleHands-on evaluator and internal champion
PainAgents hit captchas, 403 errors, login walls and API keys tied to human accounts
Convinced byDocs, quickstart and a working token flow
CTO or VP Engineering
RoleTechnical decision-maker
PainIntegration effort, security posture and OAuth2 / OIDC compatibility
Convinced byStandards-based design and partner integrations
Head of Product or BD, API provider
RoleRevenue owner and economic buyer
PainMonetising agent traffic without bank-account onboarding
Convinced byPay-per-use models and transparent pricing
Head of Payments, Fraud or Security
RoleRisk gatekeeper with veto power
PainDelegated authority, fraud on agent orders, audit evidence and KYC versus agent verification
Convinced byMandates, spend controls and network partnerships
04 · Methodology
Custom research, not a tool export
Queries were written by hand, run live, read answer by answer and coded manually, then cross-checked against the client website and Semrush data.
1 · Scope
Define the category, buyers and competitors.
2 · Query design
Write 50 buyer-style queries in a 27 / 14 / 9 mix.
3 · Funnel mapping
Classify each query as TOFU, MOFU or BOFU.
4 · Live testing
Run every query on ChatGPT, Claude, Gemini and Google AI Overviews.
5 · Coding
Read each answer and log mentions, rank, recommendation and citations.
6 · Scoring
Compute rates, AVI and share of voice, then validate.
Query allocation
| Query type | Queries | Share | TOFU | MOFU | BOFU |
|---|---|---|---|---|---|
| Non-branded | 27 | 54.0% | 10 | 10 | 7 |
| Branded comparison | 14 | 28.0% | 0 | 10 | 4 |
| Brand-specific | 9 | 18.0% | 4 | 2 | 3 |
| Total | 50 | 100% | 14 | 22 | 14 |
TOFU: learning (what is, how do). MOFU: comparing (best, vs, alternatives). BOFU: choosing or acting (pricing, integration, which should I choose). Six near-duplicate non-branded queries were dropped for overlap, not by result.
Platform test conditions
| Platform | Mode and notes |
|---|---|
| ChatGPT | Temporary chat. Some branded answers referred to account research and are flagged. |
| Claude | Incognito, web search on Auto. |
| Gemini | Saved chats (temporary mode unavailable). Source links not readable. |
| Google AIO | Manual run in Google Search: named or not named; branded answers coded from screenshots. |
How the metrics work
Mention: the client is named in the answer body. Recommendation: named as top pick, #1 in a list or recommended for the use case (non-branded only). Citation: the answer links to a client-owned domain. AVI: (mention rate + recommendation rate) ÷ 2, with no answer counted as not visible. Share of voice: share of 81 non-branded chat answers naming a vendor.
Limitations
Single runs on two days (16 and 19 Oct 2026): AI answers vary. ChatGPT account context may have inflated one non-branded answer; without it ChatGPT AVI is 15.4 and overall AVI 20.0. Google results are single manual runs, so Google is excluded from AVI and share of voice. Sentiment is classified for 50 of 91 branded answers.
05 · AI visibility audit
The AI Visibility Index, by platform and funnel stage
AVI = (non-branded mention rate + non-branded recommendation rate) ÷ 2, measured on 27 non-branded queries across 3 chat platforms (81 answers). Google AI Overviews reported mentions only, so it is excluded.
AVI by platform (0 to 100)
Chart scaled to the 0–100 index.
| Platform | Mention | Recommend | AVI | Avg pos. | Owned cites |
|---|---|---|---|---|---|
| Claude | 9/27 · 33.3% | 4/27 · 14.8% | 24.1 | 4.7 | 1/27 |
| Gemini | 7/27 · 25.9% | 4/27 · 14.8% | 20.4 | 1.9 | Not captured |
| ChatGPT | 5/27 · 18.5% | 4/27 · 14.8% | 16.7 | 2.4 | 4/27 |
| Google AIO | 3/27 · 11.1% | Not recorded | n/a | n/a | Not recorded |
| Overall | 24/108 · 22.2% | 12/81 · 14.8% | 20.4 | 3.2 | 5/54 |
Reading the index (BuzzIQ Labs interpretation, not an industry standard)
Non-branded mention rate by funnel stage (%)
TOFU · 10 queries
MOFU · 10 queries
BOFU · 7 queries
Query type results (4 platforms)
| Query type | Queries | Answers | Named | Rate |
|---|---|---|---|---|
| Non-branded | 27 | 108 | 24 | 22.2% |
| Branded comparison | 14 | 56 | 56 | 100.0% |
| Brand-specific | 9 | 35 | 34 | 97.1% |
Owned-domain citations: 5 of 54 non-branded answers (ChatGPT, Claude); 19 of 42 branded comparison and 15 of 26 brand-specific answers.
Rank and citations when named
In the 21 non-branded answers that name the client, it is listed first in 11. Average position is 3.2 (median 1). Gemini ranks it highest (average #1.9, first in 5 of 7), ChatGPT averages #2.4, and Claude names it most often but deeper in long lists (average #4.7).
When owned pages are not cited, third parties fill the gap: wire services, financial news sites and protocol sites appeared as sources on branded queries.
Sample of non-branded query results
| Query | ChatGPT | Claude | Gemini | Google AIO |
|---|---|---|---|---|
| best platforms for AI agent payments | #1 · Rec · Cited | #10 · Rec | #1 · Rec | Not named |
| top agentic commerce infrastructure startups | #1 · Rec · Cited | #4 | #1 · Rec | Named |
| best alternatives to Nevermined for agent payments | #6 · Rec · Cited | #1 · Rec | #1 | Not named |
| which agent identity solution works with Visa and Mastercard agent payments | #1 · Rec · Cited | #1 · Rec | #1 · Rec | Not named |
| pricing for AI agent payment platforms | Not named | #6 | #4 | Not named |
| best bot access control solutions for AI agents | Not named | Not named | Not named | Not named |
| agentic payment protocols compared: AP2, ACP, x402 and others | Not named | Not named | Not named | Not named |
| how can my website accept payments from AI agents | Not named | Not named | Not named | Not named |
#n = the client’s position in the answer · Rec = recommended · Cited = owned domain cited. Position, recommendation and citations were not recorded for Google AI Overviews. Source: BuzzIQ Labs query panel, 16 and 19 Oct 2026.
06 · Competitive analysis
Who sets the default answer
Rails and incumbents set the defaults: Coinbase / x402 and Stripe each appear in 44 of 81 non-branded chat answers, with Visa in 34 and Mastercard in 27.
Share of 81 non-branded answers naming each vendor (%)
The client ranks #6 of 33 vendors and #1 among agent-native startups: Crossmint 13 answers, Nevermined 12, Payman 5, Nekuda 4 (the client 21). A vendor counts once per answer; share = answers naming the vendor ÷ 81. Google AIO excluded because vendor lists were not recorded.
Nearest competitors: positioning and opportunities
| Competitor | Category | Answer share | How AI and the market frame it | Opportunity for the client |
|---|---|---|---|---|
| Nevermined | Agent-native billing | 14.8% | Recommended over the client for pay-per-use API billing (ChatGPT and Claude). Raised $4M in Jan 2025 ($7M total). | Publish an owned comparison page covering billing and identity. |
| Crossmint | Agent wallets | 16.0% | Named beside the client in startup and platform lists. Raised $23.6M led by Ribbit Capital. | Keep leading startup roundups; differentiate on identity. |
| Coinbase / x402 | Protocol and rail | 54.3% | Owns how agents pay for APIs. Coinbase and Cloudflare founded the x402 Foundation. | Explain the identity layer with x402, not against it. |
| Stripe | Incumbent payments | 54.3% | Default all-round answer: ACP with OpenAI, Shared Payment Tokens. | Own identity and authority before checkout. |
| Visa and Mastercard | Card networks | 42.0% / 33.3% | Named in rails and protocol answers. The client already works with both networks. | Publish partner pages so answers can cite them. |
| Okta, Auth0, Persona, Cloudflare | Identity and bot control | 6.2% / 3.7% / 6.2% / 33.3% | Fill identity and bot-access answers. The client’s own framing is not yet the one engines use. | Create category content for best identity platforms for agents. |
Buyer pain points versus what AI said
| Test query | Stage | Client named |
|---|---|---|
| why do websites block AI agents with captchas and 403 errors | TOFU | 0 of 4 |
| best tools to let AI agents make purchases with spending limits | MOFU | 0 of 4 |
| how do I let my AI agent check out on merchant websites without being blocked | BOFU | 0 of 4 |
| how can my website accept payments from AI agents | BOFU | 0 of 4 |
The pains the client solves best are the ones buyers search for, and none of the four platforms connected them to the client. Complex purchases involve 6 to 10 decision-makers (Gartner), so each persona needs its own answer.
Branded query sentiment: how AI describes the client
Of 50 classified branded answers: 44 neutral, 3 positive, 3 negative. The negatives are use-case verdicts, not criticism.
Counts shown as positive / neutral / negative. Sentiment is BuzzIQ Labs’ reading of each answer, not a platform-reported score.
Notable branded answers
| Branded query | Platform | Sentiment | What the answer said |
|---|---|---|---|
| Brand or Nevermined: which is better for pay-per-use API billing for agents | Google AIO | Negative | Says Nevermined is generally better for pay-per-use API billing; frames the client as strongest at outbound agent spending. |
| Brand vs Stripe agentic commerce suite: which is better for a merchant | ChatGPT | Negative | Defaults to Stripe for a merchant; the client is worth evaluating for agent-credential use only. |
| does Brand support x402, ACP and AP2 | Claude | Negative | Says none of the three is documented as natively supported. Gemini on the same query said yes, with no sources. |
| Brand explained: wallet and checkout product | Google AIO | Positive | Describes critical financial and identity infrastructure that lets AI agents act as economic actors. |
| is Brand the same company as another brand | ChatGPT, Gemini, Google AIO | Neutral | Wrong entity on 3 of 4 platforms: answers described an unrelated company with a similar name. |
07 · Website analysis
A healthy foundation that engines are not yet using
Technical health is strong and open to AI crawlers. The gaps are content structure, entity clarity and authoritative links.
Site health (top 10% of sites: 92%)
AI search health (293 issues)
Authority Score (Good)
Referring domains / backlinks
Largest issue groups
1,051 total: 13 errors, 298 warnings, 740 notices across 100 crawled pages.
100 crawled pages by status
What is working
ChatGPT-User, OAI-SearchBot, Googlebot and Google-Extended are all allowed. HTTPS 98%, crawlability 96%, site performance 96%, markup 99% (JSON-LD on 99% of pages), 93% of URLs return 2xx. Internal linking is the lowest thematic score at 88%. Also flagged: 2 invalid structured-data items, 20% of pages without a canonical tag, and only 71% of crawled pages in the sitemap.
Content, FAQ and structure review
Homepage
No FAQ blockClear positioning and partner logo wall. No canonical or Open Graph tags returned in our fetch.
Product and use-case pages
No FAQ blockFeature blocks and CTAs. Use-case page last modified Apr 2025; product copy still reads as early-access.
Industry articles (3 sampled)
FAQ on 3 of 3Each has a TL;DR, comparison table and 3–5 Q&As, but no author or publish date in head metadata.
Docs and llms.txt
llms.txt presentThe docs site publishes an llms.txt index of core pages.
Content-to-citation gap: articles that exist versus who was named
| Query theme | Client article already published | Platforms naming the client |
|---|---|---|
| Agentic payment protocols compared | Protocol comparison explainer | 0 of 4 |
| Best bot access control solutions | Bot access control buyer’s guide | 0 of 4 |
| How a website accepts agent payments | Merchant checkout guide | 0 of 4 |
| Best agent identity verification tools | Agent identity tools overview | 1 of 4 (Claude) |
Two articles also state that the client has not verified public support for x402, ACP or AP2. Engines may read that as a reason to leave the client out of protocol roundups. Articles were refreshed on 2 Oct 2026, so indexing lag may play a part (not verified).
Organic search and AI search signals
Organic traffic, worldwide (−33%)
Organic keywords (−10%)
Semrush AI Visibility, 9 mentions
Keywords by intent
Six brand-name keywords drive about 88% of US organic visits; only 2 of 42 tracked keywords are commercial. Semrush AI mentions: ChatGPT 3, AI Mode 4, AI Overview 1, Gemini 1. Source: Semrush Domain Overview (15 Oct 2026), supplied by the client.
Backlink profile review
Needs audit A single exact-match anchor text appears across 1,157 domains (1,163 links), a footprint that rarely occurs naturally. Manual review is needed before any disavow decision.
Worth building on Follow links from standards-body member directories, industry-forum pages and developer-ecosystem listings match the sources AI engines trust.
08 · Recommendations
Priority actions and impact matrix
Most of the high-impact work is low effort and sits in content structure. Third-party citations take longer but compound the most. Placements are BuzzIQ Labs judgement, not measured values.
Vertical axis: impact, low (bottom) to high (top)
Make industry articles citable
Add author, publish date, Article schema and internal links from core pages. Articles exist but 0 of 4 platforms named the client on three target queries.
Publish owned comparison and alternatives pages
Nevermined, Stripe, x402, Crossmint. Nevermined was recommended and third parties write the comparison.
Add FAQs and FAQPage schema to core pages
FAQs exist only on articles; none on home, product, use-case or identity pages.
Link the company and its open protocol as one entity
Organization schema, cross-links and one-line definitions. The protocol was explained without the company on a Gemini query.
Disambiguate from the similarly named company
ChatGPT and Gemini answered about the wrong company.
Fix Semrush errors
10 broken links, 66 meta descriptions, 84 alt texts and 2 schema items.
Earn third-party citations
Partner pages, roundups and digital PR. Press and protocol sites are cited more than owned pages.
Audit and disavow the anchor cluster if toxic
1,157-domain exact-match anchor footprint.
Re-run the 50-query panel monthly
Three runs each, to control for single-run variance.
09 · Roadmap and partnership
A 90-day plan, three phases, two re-tests
Timing is a planning assumption: AI answers can lag publishing, so week 6 and week 13 re-tests check results. Impact and effort are BuzzIQ Labs judgement.
Foundation
- Entity facts page, Organization schema and a not-the-other-company line (wk 1–2)
- Pricing and protocol-support pages, signed off by the client (wk 1–2)
- Article and FAQ schema, canonicals, sitemap fixes (wk 2–3)
- Baseline protocol: 3-run panel, Google fields (wk 1)
Build authority
- 4 comparison pages: Nevermined, Stripe, x402, Crossmint (wk 3–6)
- 6 category pages for top and middle funnel gaps (wk 4–8)
- Backlink audit of the 1,157-domain cluster (wk 3–6)
- Interim re-test, then fix weak pages (wk 6)
Earn and measure
- Pitch roundups, partner and protocol pages (wk 6–13)
- 6 third-party pages naming the client (wk 13 target)
- Refresh pages from re-test findings (wk 9–12)
- Final re-test and KPI report against baseline (wk 13)
Roadmap timeline
Roadmap detail: priorities, dependencies and deliverables
Ranked by impact on AI visibility ÷ effort, then ordered by dependency. Impact and effort are BuzzIQ Labs judgement; the evidence column is from this audit.
| # | Action and deliverable | Audit gap it addresses | Impact / effort | Timing | Target |
|---|---|---|---|---|---|
| 1 | Publish one canonical entity page (what the company is, founders, funding, products, protocol) with Organization schema, sameAs links and a not-the-other-company line. | ChatGPT, Gemini and Google described the wrong company or merged it with a similarly named one; the client’s protocol was explained without the client on two platforms. | High / Low | Weeks 1–2 · none | Live by wk 2. Wrong-entity platforms: 3 of 4 to 1 or fewer. |
| 2 | Publish dated pricing and protocol-support pages that state what is and is not supported. | Answers disagree on protocol support: Claude says undocumented, Gemini says supported with no sources, Google says x402 only. Google quotes a fee range on the pricing query. | High / Low | Weeks 1–2 · client confirms facts | Live by wk 2. Support and pricing queries match the pages on 4 of 4 platforms. |
| 3 | Publish 4 comparison pages with a feature table and FAQ: versus Nevermined, Stripe, Coinbase x402 and Crossmint. | Nevermined is recommended over the client on the billing comparison (ChatGPT, Claude, Google); Nevermined pages are cited sources in Google answers to three queries. | High / Medium | Weeks 3–6 · needs 1 and 2 | Live by wk 6. Billing comparison no longer recommends Nevermined over the client. |
| 4 | Publish 6 category pages that open with a direct answer: identity basics, agent access gateways, bot access control, protocol comparison, accepting agent payments, agent checkout. | Top funnel: named in 2 of 40 answers. 0 of 4 platforms named the client on six queries although articles exist. | High / Medium | Weeks 4–8 · needs 5 | Live by wk 8. Top-funnel mentions 2 to 5 of 40; Google 3 to 6 of 27. |
| 5 | Make industry articles citable: author, date, Article and FAQPage schema, internal links; fix 2 invalid markup items, canonicals and sitemap gaps. | Owned domain cited in 5 of 54 non-branded answers. 20% of pages lack a canonical; 71% of crawled pages are in the sitemap; internal linking 88%. | Medium / Low | Weeks 2–3 · none | All articles marked up; 0 invalid items; owned citations 5 to 8 of 54 (with 4). |
| 6 | Earn third-party mentions: roundups and listicles, card-network and protocol partner pages, trade media. | Without owned citations, wire services, financial news and protocol sites fill the gap. Competitors appear in 54.3% (Coinbase, Stripe) of answers. | High / High | Weeks 6–13 · needs 1 and 3 | 6 third-party pages naming the client by wk 13. |
| 7 | Re-test the 50 queries on 4 platforms, 3 runs each, recording Google position, recommendation and citations. | Baseline is a single run; Google position, recommendation and citations were not recorded. | High / Low | Wk 1, 6 and 13 · none | KPI report at wk 6 and wk 13 against the October baseline. |
| 8 | Audit the backlink cluster and document a keep or disavow decision. | One exact-match anchor appears on 1,157 domains (1,163 links). | Low / Low | Weeks 3–6 · none | Review delivered by wk 6. |
Roadmap KPIs: baseline and targets
| KPI | How it is measured | Baseline | Week 13 target |
|---|---|---|---|
| AI Visibility Index (3 chat) | (Mention rate + recommendation rate) ÷ 2, 27 queries × 3 platforms | 20.4 | 27.8 |
| Non-branded mentions | Answers naming the client ÷ 108 | 24 of 108 (22.2%) | 33 of 108 (30.6%) |
| Recommendations (3 chat) | Non-branded answers recommending the client ÷ 81 | 12 of 81 (14.8%) | 18 of 81 (22.2%) |
| Google AI Overview appearances | Non-branded queries where the overview names the client ÷ 27 | 3 of 27 (11.1%) | 6 of 27 (22.2%) |
| Top-funnel visibility | Top-funnel answers naming the client, 4 platforms ÷ 40 | 2 of 40 (5.0%) | 5 of 40 (12.5%) |
| Visibility across tested queries | Non-branded queries named by at least one platform ÷ 27 | 12 of 27 (44.4%) | 16 of 27 (59.3%) |
| Citation frequency, non-branded | ChatGPT and Claude answers citing the owned domain ÷ 54 | 5 of 54 (9.3%) | 8 of 54 (14.8%) |
| Citation frequency, branded | Branded answers citing owned domains ÷ 68 | 34 of 68 (50.0%) | 40 of 68 (58.8%) |
| Branded sentiment | Share of classified branded answers (50 now; all 91 at re-test) | Positive 6% · Negative 6% | Positive 12%+ · Negative 2% or lower |
| Wrong-entity answers | Platforms describing the wrong company, of 4 | 3 of 4 | 1 of 4 or fewer |
Proposed targets by week 13 (planning assumptions, not forecasts)
Non-branded mentions, 4 platforms (24 to 33 of 108)
Google AI Overview appearances, non-branded
Owned-domain citation, non-branded (5 to 8 of 54)
AI Visibility Index, 3 chat platforms
Baseline versus week 13 target (%)
Week 13 adds 6 chat mentions, 6 chat recommendations and 3 Google appearances to the baseline; AVI target = (27 ÷ 81 + 18 ÷ 81) ÷ 2 = 27.8. AI answers vary between runs, so each re-test uses 3 runs. Other KPIs tracked: branded citation frequency (50.0% to 58.8%), branded sentiment, and wrong-entity answers (3 of 4 to 1 of 4 or fewer).
10 · Sources
Sources and references
Figures from secondary compilations are attributed to the original analyst and flagged as such. The client’s own articles are used for how the client describes itself, not as independent proof.
Market and buyer research
- McKinsey & Company, The Agentic Commerce Opportunity, Oct 2025
- Morgan Stanley (Dec 2025), Bain & Company (Dec 2025) and Gartner (Nov 2025) forecasts, as compiled by Stellagent, Hexagon and MetaRouter
- Retail Dive, US Agentic Commerce Revenue Forecast To Reach $1 Trillion By 2030, May 2026 (ICSC and McKinsey)
- G2, The Answer Economy, Mar 2026, n=1,076, via Demand Gen Report
- Forrester 2026 Buyers’ Journey Survey and Gartner 2026 GenAI finding, via machinerelations.ai and supermarketers.ai
- BuzzIQ Labs, A Comprehensive Guide To Account-Based Marketing For SaaS, Sep 2026 (Gartner buying-group figure)
Competitors
- FinTech Global (Jan 2025): Nevermined $4M raise; Finovate: $7M total
- FinTech Global: Crossmint $23.6M, led by Ribbit Capital
- Competitor product pages and public announcements for Coinbase / x402, Stripe, Visa, Mastercard, Cloudflare, Okta, Auth0 and Persona
- Client website, product, use-case and industry-article pages and documentation, reviewed live
Primary data in this report
- BuzzIQ Labs AI visibility panel: 50 queries × 4 platforms, run 16 and 19 Oct 2026; Google AI Overview results collected by BuzzIQ Labs
- Semrush Domain Overview (15 Oct 2026), Organic Rankings (14 Oct 2026) and Site Audit (16 Oct 2026) screenshots supplied by the client
- BuzzIQ Labs direct review of the client website, 16 Oct 2026
11 · Appendix A
Query panel and observed results
Every query in the panel and what each platform returned. #n = the client’s position in the answer · Rec = recommended · Cited = owned domain cited · ×n = mentions (branded) with sentiment where classified · † ChatGPT used account context · ‡ answer described a different company. Gemini citations were not captured. Google AIO: named or not, branded cited and sentiment only. Branded queries are shown with the client name blurred; protocol and product terms are redacted.
| Query | ChatGPT | Claude | Gemini | Google AIO |
|---|---|---|---|---|
| Non-branded · TOFU (awareness) · 10 queries | ||||
| 1 what is term for AI agents | Not named | #1 · Cited | Not named | Not named |
| 2 how do AI agents pay for things online | Not named | Not named | Not named | Not named |
| 3 what is agentic commerce | Not named | Not named | Not named | Not named |
| 4 why do websites block AI agents with captchas and 403 errors | Not named | Not named | Not named | Not named |
| 5 how do AI agents authenticate to websites and APIs | Not named | Not named | Not named | Not named |
| 6 what are the risks of letting AI agents make purchases | Not named | Not named | Not named | Not named |
| 7 what is the x402 protocol | Not named | Not named | Not named | Not named |
| 8 term vs KYC for AI agents | #3 † | Not named | Not named | Not named |
| 9 how can I give an AI agent its own wallet | Not named | Not named | Not named | Not named |
| 41 what is delegated authority for AI agents making payments | Not named | Not named | Not named | Not named |
| Non-branded · MOFU (consideration) · 10 queries | ||||
| 10 best platforms for AI agent payments | #1 · Rec · Cited | #10 · Rec | #1 · Rec | Not named |
| 11 best AI agent identity verification tools | Not named | #7 · Rec | Not named | Named |
| 12 top agentic commerce infrastructure startups | #1 · Rec · Cited | #4 | #1 · Rec | Named |
| 13 best tools to let AI agents make purchases with spending limits | Not named | Not named | Not named | Not named |
| 14 how to choose an agent payments platform for developers | Not named | #7 | #4 | Not named |
| 15 best bot access control solutions for AI agents | Not named | Not named | Not named | Not named |
| 16 agentic payment protocols compared: AP2, ACP, x402 and others | Not named | Not named | Not named | Not named |
| 17 best agent access gateway platforms | Not named | Not named | Not named | Named |
| 18 stablecoin vs card rails for AI agent payments | Not named | Not named | Not named | Not named |
| 43 best alternatives to Nevermined for agent payments | #6 · Rec · Cited | #1 · Rec | #1 | Not named |
| Non-branded · BOFU (decision) · 7 queries | ||||
| 19 which agent payment platform should a startup building AI shopping agents choose | Not named | Not named | #1 · Rec | Not named |
| 20 how do I let my AI agent check out on merchant websites without being blocked | Not named | Not named | Not named | Not named |
| 21 how can my website accept payments from AI agents | Not named | Not named | Not named | Not named |
| 22 pricing for AI agent payment platforms | Not named | #6 | #4 | Not named |
| 23 how do I integrate agent identity and payments into my AI agent | Not named | #5 | Not named | Not named |
| 47 how do I charge AI agents for access to my API | Not named | Not named | Not named | Not named |
| 48 which agent identity solution works with Visa and Mastercard agent payments | #1 · Rec · Cited | #1 · Rec | #1 · Rec | Not named |
| Branded comparison · MOFU (consideration) · 10 queries | ||||
| 24 Brand vs Nevermined | ×17 · Cited | ×12 · Cited · Neutral | ×6 · Neutral | Named · Neutral |
| 25 Brand vs Coinbase x402 | ×23 · Cited † | ×11 · Neutral | ×10 · Neutral | Named · Neutral |
| 26 Brand vs Stripe for AI agent payments | ×19 · Cited | ×9 | ×9 | Named · Neutral |
| 27 Brand vs Visa Intelligent Commerce | ×25 · Cited | ×11 | ×15 | Named · Cited · Neutral |
| 28 Brand vs Crossmint | ×15 · Cited † | ×10 | ×8 | Named · Neutral |
| 29 Brand alternatives for AI agent payments | ×7 † | ×6 | ×2 | Named · Neutral |
| 30 Brand identity product vs Okta and Auth0 for AI agent identity | ×15 | ×11 · Cited | ×8 | Named · Cited · Neutral |
| 49 Brand vs Payman | ×8 · Cited · Neutral | ×10 · Neutral | ×6 · Neutral | Named · Cited · Neutral |
| 50 Brand vs Mastercard Agent Pay | ×17 · Cited · Neutral | ×9 · Neutral | ×6 · Neutral | Named · Neutral |
| 51 Brand vs Cloudflare for AI agent access and payments | ×18 · Cited · Neutral | ×9 · Cited · Neutral | ×13 · Neutral | Named · Cited · Neutral |
| Branded comparison · BOFU (decision) · 4 queries | ||||
| 31 Brand or Nevermined: which is better for pay-per-use API billing for agents | ×13 · Cited | ×9 · Cited | ×7 | Named · Negative |
| 32 Brand vs Stripe vs Coinbase: which should I use so my agents can buy things | ×6 · Cited | ×3 | ×7 | Named · Neutral |
| 52 Brand or Crossmint: which is better for agent wallets | ×10 · Cited · Neutral | ×5 · Neutral | ×11 · Neutral | Named · Neutral |
| 53 Brand vs Stripe agentic commerce suite: which is better for a merchant | ×11 · Negative | ×7 · Neutral | ×6 · Neutral | Named · Neutral |
| Brand-specific · TOFU (awareness) · 4 queries | ||||
| 33 what is Brand and what does it do | ×22 · Cited | ×17 · Cited | ×15 | Named · Cited · Neutral |
| 34 who is behind Brand and how much funding has it raised | ×11 · Cited † | ×6 | ×3 | Named · Neutral |
| 35 what is the client protocol | ×2 | ×9 | Not named | Named · Positive |
| 54 is Brand the same company as another brand | ×11 · Neutral ‡ | ×10 · Neutral | ×8 · Neutral ‡ | Named · Neutral ‡ |
| Brand-specific · MOFU (consideration) · 2 queries | ||||
| 36 Brand reviews, pros and cons | ×21 † | ×14 · Cited | ×5 | Named · Neutral |
| 55 Brand product explainer | ×13 · Cited · Neutral | ×10 · Cited · Neutral | ×12 · Neutral | Named · Cited · Positive |
| Brand-specific · BOFU (decision) · 3 queries | ||||
| 37 Brand pricing and fees | ×15 · Cited | ×14 · Cited | ×19 | Named · Cited · Neutral |
| 38 how do I get started integrating Brand | ×36 · Cited | ×11 · Cited | ×27 | Named · Cited · Neutral |
| 56 does Brand support x402, ACP and AP2 | Not run | ×13 · Negative | ×13 · Positive | Named · Cited · Neutral |
Totals by platform
| Platform | Non-branded named (of 27) | Branded comparison (of 14) | Brand-specific (of 9) | All 50 | AVI |
|---|---|---|---|---|---|
| ChatGPT | 5 · 18.5% | 14 · 100.0% | 8 of 8 · 100.0% | 27 of 49 · 55.1% | 16.7 |
| Claude | 9 · 33.3% | 14 · 100.0% | 9 of 9 · 100.0% | 32 of 50 · 64.0% | 24.1 |
| Gemini | 7 · 25.9% | 14 · 100.0% | 8 of 9 · 88.9% | 29 of 50 · 58.0% | 20.4 |
| Google AIO | 3 · 11.1% | 14 · 100.0% | 9 of 9 · 100.0% | 26 of 50 · 52.0% | n/a |
| All 4 platforms | 24 · 22.2% | 56 · 100.0% | 34 of 35 · 97.1% | 114 of 199 · 57.3% | 20.4 (3 chat) |
Totals = answers naming the client ÷ answers observed (199 = 149 chat + 50 Google; one ChatGPT answer not run). AVI uses the 27 non-branded queries on the 3 chat platforms only. Excluded: 22 earlier prompts and 6 near-duplicate queries. Source: BuzzIQ Labs query panel, 16 and 19 Oct 2026.
12 · Appendix B
Branded sentiment evidence
Positive = favourable or preferred framing · Neutral = descriptive, balanced or complementary · Negative = inferior, not recommended or doubtful. Chat answers re-run 19 Oct 2026; chat answers not listed were not re-run for sentiment. Google AIO: all 23 branded answers coded from BuzzIQ Labs screenshots. 50 of 91 branded answers classified: 44 neutral, 3 positive, 3 negative. Evidence is paraphrased and anonymized.
| Query | Platform | Sentiment | Evidence (from the answer) |
|---|---|---|---|
| 24 Brand vs Nevermined | Claude | Neutral | Less head-to-head than it looks: client identity-first on the buy side, Nevermined monetization-first on the sell side; neither preferred. |
| 24 Brand vs Nevermined | Gemini | Neutral | Complementary angles: spend and trust stack versus monetization and metering; comparison matrix, no preference. |
| 24 Brand vs Nevermined | Google AIO | Neutral | Both leading platforms with different focus, but a readiness table scores the client lower on agent readiness and merchant monetization. |
| 25 Brand vs Coinbase x402 | Claude | Neutral | Different layers of the stack: open protocol versus a full-stack network that can speak it; no preference. |
| 25 Brand vs Coinbase x402 | Gemini | Neutral | Different layers: open protocol standard versus managed identity and payment platform; comparison matrix, no preference. |
| 25 Brand vs Coinbase x402 | Google AIO | Neutral | Sit on different layers: x402 is a payment protocol, the client is the identity and credential layer. |
| 26 Brand vs Stripe for AI agent payments | Google AIO | Neutral | Split between agent-native identity and wallets (client) and traditional commerce (Stripe); client for autonomous spending. |
| 27 Brand vs Visa Intelligent Commerce | Google AIO | Neutral | Not direct competitors but complementary parts of agentic commerce, used together in pilots. |
| 28 Brand vs Crossmint | Google AIO | Neutral | Both lead their niches: client for strict identity and spend controls, Crossmint for cards and multi-rail wallets. |
| 29 Brand alternatives for AI agent payments | Google AIO | Neutral | Lists Nevermined, Payman, Agentspay, Catena Labs, Crossmint and Coinbase x402 as alternatives; the client is the reference product. |
| 30 Brand identity product vs Okta and Auth0 | Google AIO | Neutral | Okta, Auth0 and the client serve different roles and complement each other; the client covers agent identity plus payment control. |
| 31 Brand or Nevermined: pay-per-use API billing | Google AIO | Negative | Says Nevermined is generally better for pay-per-use API billing; frames the client as strongest at outbound agent spending. |
| 32 Brand vs Stripe vs Coinbase | Google AIO | Neutral | Different sides: Stripe for card merchants, Coinbase or the client for web3 and agent-to-agent use. |
| 33 what is Brand and what does it do | Google AIO | Neutral | Defines the company as financial and identity infrastructure for agentic commerce and lists its features; notes unrelated products with the same name. |
| 34 who is behind Brand and its funding | Google AIO | Neutral | Separates the payments company (seed funding, two named founders) from an unrelated company with a similar name; correct disambiguation. |
| 35 what is the client protocol | Google AIO | Positive | Credits the client as primary creator of its open protocol and lists supporting partners. |
| 36 Brand reviews, pros and cons | Google AIO | Neutral | Balanced list: pros (autonomous payments, identity, spending safeguards) and cons (nascent market, AI risk, crypto dependency). |
| 37 Brand pricing and fees | Google AIO | Neutral | States a transaction fee range with card, ACH and USDC funding; adds a note about an older unrelated app with the same name. |
| 38 getting started integrating Brand | Google AIO | Neutral | Step-by-step onboarding: dashboard sign-up, wallet, API keys and framework integrations. |
| 49 Brand vs Payman | ChatGPT | Neutral | Balanced: choose the client if building agents that buy online services; Payman for authorized money management. |
| 49 Brand vs Payman | Claude | Neutral | Less competitors than complements: different sides of the agent wallet; card-network partnership noted. |
| 49 Brand vs Payman | Gemini | Neutral | Different lifecycle points: open-web commerce and identity versus banking and treasury; choose by use case. |
| 49 Brand vs Payman | Google AIO | Neutral | Opposite sides: client for agents paying digital services, Payman for paying human workers; Payman can use the client. |
| 50 Brand vs Mastercard Agent Pay | ChatGPT | Neutral | Compares focus areas; has a section on where the client has the advantage but also notes Mastercard strengths (network scale, disputes). |
| 50 Brand vs Mastercard Agent Pay | Claude | Neutral | Different layers of the same stack; the client is named among the card network’s agent-payment program participants. |
| 50 Brand vs Mastercard Agent Pay | Gemini | Neutral | Both pioneering; differ in rails (startup and stablecoin versus card network); no preference stated. |
| 50 Brand vs Mastercard Agent Pay | Google AIO | Neutral | Not direct competitors; the client has partnered with Mastercard to supply identity to Agent Pay. |
| 51 Brand vs Cloudflare | ChatGPT | Neutral | Agent-payment platform versus internet infrastructure; notes some features in early access; no winner declared. |
| 51 Brand vs Cloudflare | Claude | Neutral | Compared a lot; client is agent-side and identity-first, Cloudflare is site-side; how they fit together as a stack. |
| 51 Brand vs Cloudflare | Gemini | Neutral | Different layers rather than strictly competing; choose-by-use-case guidance. |
| 51 Brand vs Cloudflare | Google AIO | Neutral | Complementary layers (identity and payments versus network-edge enforcement), not competitors. |
| 52 Brand or Crossmint: agent wallets | ChatGPT | Neutral | Split verdict: client best for agent-to-service payments, Crossmint best for programmable wallets. |
| 52 Brand or Crossmint: agent wallets | Claude | Neutral | No clean winner; client identity-first with API micropayments, Crossmint wallets and cards across rails. |
| 52 Brand or Crossmint: agent wallets | Gemini | Neutral | Choose by use case: Crossmint for broader web and merchants, client for identity plus API micropayments. |
| 52 Brand or Crossmint: agent wallets | Google AIO | Neutral | Depends on priority: client for agent identity and crypto-native micro-transactions, Crossmint for cards and multi-chain wallets. |
| 53 Brand vs Stripe agentic commerce suite | ChatGPT | Negative | Default recommendation is Stripe for a merchant; the client is worth evaluating for agent-credential use; product discovery is not its primary documented focus. |
| 53 Brand vs Stripe agentic commerce suite | Claude | Neutral | Less competitors than different layers; Stripe for new demand, the client for existing agent traffic. |
| 53 Brand vs Stripe agentic commerce suite | Gemini | Neutral | Depends on merchant type: Stripe for retail and fiat, the client for API and machine-to-machine sellers. |
| 53 Brand vs Stripe agentic commerce suite | Google AIO | Neutral | Not either-or: the client covers agent identity, Stripe covers checkout; the client is best for merchants worried about bot fraud. |
| 54 is Brand the same company as another brand | ChatGPT | Neutral | Wrong entity: treats the client as an unrelated company with a similar name; the payments company is not identified. |
| 54 is Brand the same company as another brand | Claude | Neutral | Correct disambiguation: two unrelated companies. |
| 54 is Brand the same company as another brand | Gemini | Neutral | Wrong entity: explains the unrelated company; the payments company appears only in a closing note. |
| 54 is Brand the same company as another brand | Google AIO | Neutral | Wrong entity: says the two are directly connected through an acquisition and mixes the companies. |
| 55 Brand product explainer | ChatGPT | Neutral | Explainer of the wallet and checkout product with spending rules; neutral and descriptive, cites the client docs. |
| 55 Brand product explainer | Claude | Neutral | Factual explainer of the products; no comparative judgement. |
| 55 Brand product explainer | Gemini | Neutral | Descriptive explainer of the product stack; no comparative judgement. |
| 55 Brand product explainer | Google AIO | Positive | Says the client provides critical financial and identity infrastructure that lets AI agents act as economic actors. |
| 56 does Brand support x402, ACP and AP2 | Claude | Negative | None of the three is documented as natively supported; only partial overlap with x402; the client produces its own protocol. |
| 56 does Brand support x402, ACP and AP2 | Gemini | Positive | Opens with yes: support and interoperability for all three; no sources captured (claim unverified). |
| 56 does Brand support x402, ACP and AP2 | Google AIO | Neutral | Integrates with x402 through its identity layer but runs its own rails instead of natively supporting ACP or AP2. |
FAQs
AI Visibility Audit: common questions
Quick answers about what the audit covers, how it is run and what you receive.
What is an AI Visibility Audit?
It measures whether AI assistants such as ChatGPT, Claude, Gemini and Google AI Overviews name, recommend or cite your brand when buyers research your category, and which competitors they name instead.
How is the audit run?
We build a hand-written panel of buyer-style queries across awareness, consideration and decision stages, run each query live on every AI platform, and code each answer for mentions, recommendations, position and citations. The sample shows 50 queries, 4 platforms and 199 observed answers.
What is the AI Visibility Index?
A 0 to 100 score that combines mention, recommendation, position and citation data by platform and funnel stage. It gives you a baseline to track against after the fixes go live.
Which competitors are included?
We benchmark the competitors AI actually names for your category, including share of voice and how each answer frames the comparison with your brand.
What do I receive?
A full report with your AI Visibility Index, funnel and platform gaps, competitor analysis, a website and content review, and a prioritized 90-day roadmap with KPI targets.
Is the client in this sample real?
The sample comes from a real engagement. The client’s identity is anonymized; competitor names and audit figures are unchanged.
How do I get an audit for my brand?
Use the Get Your AI Visibility Report button at the top of this page or contact us at buzziqlabs.com/contact-us/.
How BuzzIQ Labs helps
See where your brand stands in AI answers
BuzzIQ Labs is a B2B SaaS and tech growth partner focused on SEO, AEO and GEO, content marketing and demand generation. We would own the entity, facts, comparison and category pages, schema fixes and third-party outreach, and re-run the panel at weeks 6 and 13 against your baseline.
US: +1-307 291 0384India: +91-750 791 2211Sample report prepared by BuzzIQ Labs from a real engagement. Client identity is anonymized; competitor names and audit figures are unchanged. Market figures trace to the named analyst reports; secondary compilations are attributed to the original analyst. Impact and effort placements, bands and targets are BuzzIQ Labs judgement and planning assumptions, not measured values or forecasts.
