Crunchbase Answer-Engine Citation Authority: 2.76% Citation Rate, B-Confidence — Claude Emerges as Dominant Engine

Crunchbase is cited in 2.76% of all observed AI answer engine runs across six engines, ranking 12th among 16,600 measured domains and 2nd among evidence-qualified market databases in the Machine Relations Index v2. Over 76 days of observation (May 10 – July 29, 2026), Crunchbase appeared in 271 answer runs out of 9,810 total. The most recent 30-day rolling window reveals a structural shift: Claude now produces more Crunchbase citations than any other engine, and Crunchbase's average citation position improved to 3.5 — meaning it appears near the top of AI responses when cited.

Last updated: July 29, 2026

Crunchbase MRI v2 Profile: Full Observation Window #

The Machine Relations Index v2 measures source-segment citation rates — how often AI answer engines cite each source domain — across ChatGPT, Claude, Gemini, Perplexity, Google AI Mode, and Google AI Overviews. A domain's citation rate publishes only after it clears the evidence floor: at least 10 observations across at least 7 distinct run dates.

The full-window citation rate decreased from an earlier 3.60% (271/7,532 over 63 days) to 2.76% (271/9,810 over 76 days). This reflects the MRI's expanded observation scope — 30% more runs observed across additional query categories — not a decline in Crunchbase's citation frequency. The absolute citation count (271 runs) remained stable. All top-ranked domains experienced proportional rate compression as the observation base grew.

Engine-Level Citation Distribution: 30-Day Rolling Window #

The most recent 30-day window (through late July 2026) shows 81 Crunchbase citations with a distribution that differs markedly from the engine landscape six months ago.

Claude's position as Crunchbase's dominant citation engine is the most significant structural shift in this update. Claude produces nearly 7x more Crunchbase citations than ChatGPT (20 vs. 3) — a pattern distinct from other top-ranked sources. For comparison, G2's 30-day engine distribution is led by Gemini (35.9%), with Claude contributing only 6.2% of G2 citations. Crunchbase's Claude-dominant profile suggests that Claude's retrieval architecture weights entity-resolved company databases more heavily when answering business and funding queries.

Google's engine family (Gemini, AI Mode, AI Overviews) collectively accounts for 53.1% of Crunchbase citations — a majority, but less concentrated than G2's 66.9% Google-family share. Crunchbase's more even distribution across engines reflects its role as a cross-engine entity reference rather than a format optimized for any single retrieval architecture. The Foglift AI Search Citation Benchmark Q2 2026 — an independent study tracking 1,119 cited domains across 375 buyer-intent responses — found low cross-engine overlap (Jaccard index 0.18), confirming that each engine maintains its own source preferences. Crunchbase's presence across all six MRI-tracked engines at relatively even shares indicates structural authority rather than optimization for any single engine's retrieval pipeline.

The 30-day average citation position of 3.5 represents an improvement over the full-window average of 5.5. When engines cite Crunchbase in recent runs, they place it in the top third of their citation lists — consistent with its function as a foundational entity reference that opens a response before analytical sources provide interpretation.

Citation Rates by Category #

Crunchbase's citation rate varies significantly across subject categories, revealing where AI engines treat it as a primary data source versus an occasional reference. MRI v2 stratum data breaks this into published categories that have cleared the evidence floor.

The 30-day rolling data also shows Crunchbase cited in three categories not yet represented in the v2 stratum table — legal/compliance, PR/earned media, and SaaS — which are still collecting evidence toward the publication floor. These expanding categories align with Crunchbase's growing query coverage: 29 distinct queries triggered Crunchbase citations in the latest 30-day window across 9 observed verticals.

The HR/talent and cybersecurity categories show citation rates more than double Crunchbase's overall average. These are categories where buyer-intent queries frequently involve company comparisons, funding history, and employee-count data — exactly the structured fields Crunchbase maintains. External measurement corroborates this pattern: CapstonAI's Q1 2026 cohort analysis found Crunchbase was the second-most-cited business data source after Wikipedia for B2B brand queries, appearing in 19% of Claude responses, 17% of ChatGPT responses, and 14% of Perplexity responses for B2B SaaS queries. Pendium.ai assigns Crunchbase an AI Visibility Score of 76/100, consistent with its confidence B rating in the MRI.

What Structural Properties Drive Crunchbase's Citation Authority #

Three structural properties distinguish Crunchbase from lower-ranked market databases. The concentration matters: Hexagon's analysis of 10,000 AI-generated citations across ChatGPT, Perplexity, and Claude found that 3% of brands capture 71% of all generative search recommendations. Data-layer sources like Crunchbase are part of the infrastructure that determines which brands appear in that 3%.

Structured, entity-resolved data. Crunchbase profiles map directly to discrete business entities — each company has a canonical profile with standardized fields (founding date, funding rounds, employee count, industry tags, key personnel). AI engines that need to answer "who is Company X" or "compare Company A vs Company B" can pull structured facts without parsing natural language. GetCito's analysis calls Crunchbase profiles a "digital deed" for the Knowledge Graph — the canonical entity reference that AI retrieval systems trust because the data is structured, verifiable, and owner-maintained. Forbes reported that Crunchbase's own AI systems can forecast unicorns with 95% accuracy using this structured dataset — a signal that the data's internal structure is machine-readable enough for AI systems to not only cite but reason over.

Claude's emergence as Crunchbase's dominant citation engine reinforces this structural reading. Claude's retrieval pipeline processes structured entity data differently from Google's engines — where Google's family draws from its existing index and knowledge graph, Claude's retrieval appears to weight canonical entity databases more heavily for company-level queries. The result is that Crunchbase's structured data architecture is most valuable to the engine whose retrieval most directly consumes it.

Funding-round provenance chains. Every funding entry on Crunchbase links to a press source (TechCrunch, Bloomberg, regional outlets). This creates a citation chain: the AI engine cites Crunchbase for the aggregate data, and Crunchbase itself cites the primary reporting source. CapstonAI's data shows that verified Crunchbase profiles receive 1.6x higher weighting in AI engine citation decisions compared to unverified profiles — suggesting that the provenance chain functions as a trust signal. The 5WPR Venture Capital AI Visibility Index, which analyzed 28,400 prompts across five AI engines, found that firm-owned editorial domains rarely appear in the top cited sources for VC queries — only a16z.com made the top 10. Data-layer platforms like Crunchbase fill the gap, serving as the neutral entity reference when AI engines need company facts without vendor bias.

Cross-category coverage. The MRI v2 data shows Crunchbase with published citation rates across six distinct subject categories — and three more collecting evidence. This breadth reflects Crunchbase's role as a horizontal data layer: it contains company records for every vertical, so any buyer-intent query that needs company context can pull from it. Scrunch's citation tracking research confirms that systematic cross-platform citation measurement — not one-off prompt testing — is required to detect this kind of structural authority, because individual engine snapshots miss the cross-category consistency that defines sources like Crunchbase.

How Crunchbase Compares to Other Market Databases #

Among top-ranked market databases tracked by the Machine Relations Index v2, three hold significant positions. The comparison below uses the full 76-day observation window for structural comparisons and the 30-day rolling window for recent engine dynamics.

Full observation window (76 days, 9,810 runs):

G2 leads on citation rate (3.54% vs. 2.76%) and carries confidence A, but Crunchbase earns better average position quality — 5.5 vs. G2's 7.5. When Crunchbase is cited, it appears earlier in the response. G2's higher citation rate comes from its review-aggregation function, which makes it relevant to comparative queries ("X vs Y"). Crunchbase's higher position quality comes from its entity-data function — it provides the foundational company facts that open a response before comparative analysis begins.

30-day rolling window — engine distribution comparison:

The engine distribution reveals a fundamental difference in how Crunchbase and G2 earn citations. Crunchbase's citations are distributed more evenly across engines, with Claude dominant — reflecting its use as a cross-engine entity reference. G2's citations are concentrated in Google's engine family (66.9% combined) — reflecting its optimization for Google's retrieval pipeline through structured review data.

Gartner occupies a distinct position among market databases. With a 3.88% full-window citation rate (381/9,810), confidence A, and rank #6/16,600, Gartner leads both G2 and Crunchbase on rate and confidence. But Gartner's average position (7.6) is the weakest of the three — consistent with Gartner's role as an analytical overlay that AI engines cite for interpretation after the foundational entity facts (Crunchbase) and comparative review data (G2) have been established.

The stratum data reveals where Crunchbase and G2 diverge most sharply:

Crunchbase outperforms G2 in HR/talent (8.88% vs. 5.92%), cybersecurity (7.70% vs. 3.53%), and enterprise software (6.21% vs. 5.03%) — categories where company funding data and market-sizing context are frequently part of the answer AI engines construct. G2 outperforms in fintech (2.44% vs. 1.79%) and martech (6.15% vs. 3.65%) — categories where product-comparison matrices and vendor reviews are the primary value an AI engine retrieves.

What Operators Can Learn from Crunchbase's Citation Profile #

The updated MRI v2 data surfaces four operational lessons:

1. Citation rates are relative to the observation scope. Crunchbase's rate moved from 3.60% to 2.76% not because engines cite it less, but because the MRI expanded its query basket by 30%. Operators tracking their own citation rates should anchor to absolute citation counts and position quality alongside rates — a rate decline can mask stable or improving absolute performance.

2. Engine distribution matters more than aggregate rate. Crunchbase's Claude-dominant citation profile (24.7% of citations) versus G2's Gemini-dominant profile (35.9%) means these sources serve different retrieval architectures. For companies optimizing their entity presence, the engine that matters depends on where their buyers research — and Claude's rapid adoption for B2B research (more than doubling usage per G2's 2026 AI Search Insight Report) makes Crunchbase's Claude-leading profile increasingly relevant.

3. Position quality signals structural function. Crunchbase's 30-day average position of 3.5 — compared to G2's 7.5 and Gartner's 7.6 — reflects its role as a foundational entity reference that AI engines place early in responses. Sources that provide the "who" earn better position than sources that provide the "which one" or the "what does this mean." For operators building citation architecture, this suggests that entity completeness on Crunchbase may influence where their company data appears in AI responses.

4. Cross-category coverage creates compounding authority. Crunchbase's presence across nine verticals in the 30-day data (six with published stratum rates, three collecting) reflects horizontal data architecture. Vertical-specific competitors may win individual categories, but Crunchbase's breadth means any buyer-intent query that needs company context can reach it. Angelfish Marketing's study of over one million AI citations reinforces this finding: structured data sources and review aggregators dominate what AI engines read to recommend B2B software — and Crunchbase sits at the intersection.

The practical implication remains: CapstonAI reports that 64% of B2B brands leave their Crunchbase profile 30–50% incomplete. For any brand competing for AI visibility, incomplete profiles on a source with Crunchbase's citation authority represent a measurable gap — not because optimization games work, but because the structured data that AI engines already retrieve is missing.

How This Connects to Machine Relations #

In the Machine Relations framework, Crunchbase's updated profile demonstrates two properties that define source authority in AI search:

First, citation authority is engine-specific, not universal. The same domain can rank #1 for citations from one engine and #6 from another. Crunchbase's Claude-dominant profile and G2's Gemini-dominant profile are not competing for the same citations — they serve different retrieval architectures. Machine Relations measurement must track engine-level distribution, not aggregate rates alone, to capture where authority actually resides.

Second, entity-resolved data earns better position quality than analytical or review-based data. Crunchbase consistently earns positions 3-5 when cited, while analytical sources (Gartner) and review aggregators (G2) earn positions 7-8. This pattern holds across engines and categories because AI retrieval systems use entity references to ground a response before layering interpretation. For practitioners building source architecture, the implication is that entity completeness on neutral data platforms compounds differently than content quality on owned domains — both matter, but they occupy different positions in how AI engines construct answers.

FAQ #

What is Crunchbase's current MRI citation rate? #

Crunchbase's overall citation rate in the Machine Relations Index v2 is 2.76%, based on 271 cited runs out of 9,810 observed across six AI answer engines over 76 days (May 10 – July 29, 2026). Crunchbase holds B-confidence, ranking #12 in the full universe of 16,600 domains and #2 among evidence-qualified market databases.

Why did Crunchbase's citation rate decrease from 3.60% to 2.76%? #

The rate decrease reflects the MRI's expanded observation scope — 30% more runs observed across additional query categories (9,810 vs. 7,532). Crunchbase's absolute citation count (271 runs) remained stable. All top-ranked domains experienced proportional rate compression as the observation base grew. This is a measurement expansion, not a decline in citation frequency.

Which AI engine cites Crunchbase most? #

Claude produces the largest share of Crunchbase citations — 24.7% in the most recent 30-day window (20 of 81 total), followed by Google AI Overviews (21.0%) and Gemini (19.8%). This Claude-dominant profile is distinct from most other top sources and reflects Claude's retrieval weighting of structured entity databases for business queries.

How does Crunchbase compare to G2 and Gartner? #

Gartner leads on citation rate (3.88%) and confidence (A). G2 ranks second on rate (3.54%) with confidence A. Crunchbase (2.76%, confidence B) leads on position quality — average position 5.5 in the full window and 3.5 in the 30-day window, compared to G2's 7.5 and Gartner's 7.6. When cited, Crunchbase appears earlier in AI responses than either competitor.

Why does Crunchbase rank higher in HR/talent queries than fintech? #

Crunchbase's HR/talent citation rate (8.88%) is nearly five times its fintech rate (1.79%). HR and cybersecurity are categories with high M&A and funding activity where buyer-intent queries frequently need company-level data — founding dates, headcount, funding history. Fintech queries tend to involve product feature comparisons, which AI engines resolve through review platforms and vendor documentation rather than company databases.

Is Crunchbase citation authority something companies can influence? #

Partially. Companies control their own Crunchbase profiles — completeness, accuracy, funding-round sourcing, and industry tags. CapstonAI data shows verified profiles receive 1.6x higher citation weighting. But Crunchbase's domain-level authority (its citation rate) is earned through structural properties — entity resolution, provenance chains, cross-category coverage — that individual companies do not control.

What is the Machine Relations Index methodology? #

The Machine Relations Index v2 measures source-segment citation rates across six production AI engines using standardized buyer-intent questions. A domain's citation rate is the share of observed answer runs in which it was cited. Rates publish only after clearing the evidence floor (10+ observations across 7+ distinct run dates). Confidence tiers (A, B, C, or collecting) reflect how much evidence stands behind each rate. The current observation window covers 16,600 tracked domains and 9,810 observed runs across 76 days.