How AI Search Is Recommending Small Business Loans: Monthly Trends

Executive Summary

The small business lending category was stable at the top in October 2026. Bluevine held 1st place with 65.1% valid recommendation coverage, a lead of 7.4 points over onDeck at 57.7%. That margin is close to September 2026's 7.6-point gap, so the top of the field did not reposition this month. The competitive consequence sits below the leader: the fintech tier now occupies three of the top four positions, and two of them moved up significantly against the July 2026 baseline.

The largest baseline riser is Fundbox, up 16.8 points from 39.8% in July 2026 to 56.6% in October 2026, a significant move that the benchmark places on a three-month upward streak. onDeck posted the second-largest significant baseline gain, up 11.7 points from 46.0% to 57.7%, also on a two-month upward streak. Both brands pair that coverage growth with stronger top-three placement and more frequent first-position appearances than at baseline.

The sharpest decliner is the tracked entity Bank of America, down 56.5 points from 56.5% in July 2026 to 0.0% by September 2026 and 0.0% again in October 2026. The same pattern appears at Chase, down 49.0 points from 49.0% to 0.0%. Both movements coincide with the appearance of Bank of America Corp. (24.1% in October 2026) and Chase Credit Journey (0.3%) as replacement tracked entities, so the benchmark reads them as a brand-entity tracking change rather than evidence that either bank lost commercial ground.

Against the prior month, nothing crossed the significance threshold. Bank of America Corp. added 5.6 points, Fundbox added 5.1 points, onDeck added 5.4 points, and Bluevine added 5.2 points, each inside normal month-to-month variation. The category-level picture is several smaller movements in the same direction rather than one decisive shift.

Each monthly run begins with 800 prompt-surface observations (538 unique questions in October 2026, against 565 in July 2026) across the benchmark's defined AI/search surface universe. All 800 observations in each month mentioned a tracked brand or competitor; 618 were relevant and 182 were irrelevant in October 2026, against 476 relevant and 324 irrelevant in July 2026. The public metrics use the 378 qualified observations from October 2026 and the 437 qualified observations from July 2026 that survive both qualification stages.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• What kinds of factual inconsistencies did AI platforms produce about small business lenders this month?
• Which lending terms triggered the most conflicting answers across platforms?

The benchmark detected 28 critical or high-severity factual inconsistencies across 6 AI platforms in October 2026. The conflicts cluster around eligibility thresholds, maximum credit limits, pricing, and corporate identity, meaning a small business owner researching the same lender on two different surfaces can receive materially different answers.

Fundbox

Fundbox accounts for 15 of the 28 inconsistencies, spanning all six platforms. AI platforms provided conflicting information about the minimum time in business. When asked what the easiest business line of credit to get is, Copilot stated 6+ months in business while ChatGPT stated 3 months, citing pages including Clear Value Lending, BizLendHub, and NerdWallet. A related conflict surfaced on the same question set, with Gemini stating 3+ months in business and Copilot stating 12+ months when asked who gives the best business line of credit.

Responses differed on the minimum annual revenue requirement across five separate conflicts. When asked whether Fundbox is a legit company, Copilot stated $100,000 annual revenue required while Google AI Mode stated a $30,000 minimum, with flagged sources including a Housecall Pro help article on one side and a LendingTree review on the other. On the easiest business line of credit question, Copilot stated $50K+/year while ChatGPT stated $30,000. On how to get approved for Fundbox, Google AI Mode described a range of $30,000 to $100,000 depending on financial profile while Google AI Overviews stated $100,000 or more, referencing the company's own help center eligibility page. On the same approval question, ChatGPT stated at least $30,000 in annual revenue while Google AI Mode stated $100,000 as the standard benchmark. On how Fundbox works, Google AI Mode stated at least $30,000 while Google AI Overviews described a threshold typically starting around $30,000 to $100,000.

The maximum credit line produced two further conflicts. When asked who has the best small business line of credit, ChatGPT stated up to $250K while Gemini stated up to $150,000, with a flagged Bluevine page citing the $250,000 figure. When asked how Fundbox works, Perplexity described a revolving line commonly cited up to $150,000 to $200,000 while Google AI Mode stated up to $250,000, with flagged sources including BestCompany, Business.org, and LendingTree.

Personal credit score requirements conflicted twice. When asked the easiest business line of credit to get, ChatGPT stated a minimum 600 FICO while Google AI Overviews stated 500+. When asked how to get approved for Fundbox, ChatGPT stated that 600 should not be treated as an absolute cutoff given the company's holistic evaluation while Google AI Overviews stated a minimum of 600, citing the company's help center, Crestmont Capital, and Business.com.

Corporate identity produced three conflicts. When asked whether Fundbox is a lender, Google AI Mode stated the company is a direct lender funding from its own balance sheet rather than a broker or marketplace, while Google AI Overviews stated loans are originated by partner banks such as First Electronic Bank or Lead Bank, citing a Finder review. On the same question, Google AI Mode again described direct balance-sheet funding while Copilot stated loans are funded through First Electronic Bank, again citing Finder. On the same question, Google AI Mode described direct funding while ChatGPT stated the company is not a direct lender and that loans are originated by partner banks including First Electronic Bank and Lead Bank.

Bluevine

Bluevine accounts for 6 inconsistencies across five platforms. Responses conflicted on the APY offered on Bluevine Business Checking. When asked what the best small business bank account is, ChatGPT stated up to 1.3% APY on qualifying balances up to $250,000 while Google AI Mode stated up to 3.00% APY on balances up to $3 million. A second conflict on the same topic appeared when asked whether Bluevine is a good business bank account, with ChatGPT stating the standard plan offers up to 1.3% APY under requirements and Google AI Mode stating up to 3.0% APY depending on account tier and monthly activity requirements. In that second conflict, the company's own business checking page was flagged on both sides of the discrepancy.

Eligibility thresholds conflicted twice. When asked whether it is hard to get approved with Bluevine, Copilot stated 12+ months in business while Perplexity stated 2+ years, with flagged sources including Haystack, Fundnode, and the company's own help center on the 12-month side. On the same question, Copilot stated $10,000 per month in revenue while Perplexity cited sources saying $40,000 in monthly revenue, with Haystack, the company's help center, and Fundnode flagged on the lower figure.

Pricing conflicted once. When asked who has the best business line of credit rates, Copilot stated rates of 14% or higher while Gemini stated starting rates around 7.8%, with Bankrate and CNBC flagged on the 7.80% figure and Axiant Partners flagged on the higher range.

onDeck

onDeck accounts for 3 inconsistencies across four platforms. When asked what the best loan for a small business is, Copilot stated an APR range of 15% to 50% while Gemini stated that online term loans such as onDeck can reach up to 90% or more. When asked whether onDeck is a reputable company, Google AI Mode stated that prepayment penalties or fees may apply depending on product and state while Google AI Overviews stated there is no penalty for early repayment. When asked what the easiest business loan to get approved for is, Google AI Mode stated minimum credit scores often range from 550 to 600 for online lenders including onDeck while Copilot stated the minimum is 625, with LendingTree and ZenBusiness flagged on the 625 figure.

Biz2Credit

Biz2Credit accounts for 2 inconsistencies across two platforms. When asked how Biz2Credit works, Google AI Overviews stated term loans up to $2 million while Perplexity stated up to about $1 million, with a Finder review flagged on the $2 million figure. When asked whether Biz2Credit is easy to get approved for, Google AI Overviews stated a general requirement of at least $100,000 in annual revenue while Perplexity cited common thresholds around $250,000+ annually, with Finder flagged on the higher figure.

National Funding

National Funding accounts for 1 inconsistency across two platforms. When asked how National Funding works, ChatGPT stated many customers have a FICO score of 600+ while Perplexity cited a credit score around 660+, with LendingTree flagged on the 600+ figure.

QuickBridge

QuickBridge accounts for 1 inconsistency across two platforms. When asked who the CEO of QuickBridge is, ChatGPT named Taylor Hilton as CEO per a business-directory profile while Gemini described Ben Gold as founding President leading the company under Dave Gilbert as CEO of National Funding, citing deBanked and Equipment Finance Advisor coverage of the acquisition.

Benchmark Context

Questions This Section Answers

• How many prompt-surface observations qualified for the benchmark this month?
• Why did qualified observations decline even as relevant prompts increased?

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

The raw collection universe held steady at 800 observations in both months while the relevance mix improved, with relevant prompts rising from 476 to 618 and irrelevance falling from 324 to 182. Qualified observations nonetheless declined from 437 to 378, because the reserved pool grew. The August 2026 interim month qualified 464 relevant prompts and the September 2026 month qualified 651, the highest of the series.

Benchmark-Level Metrics

The August 2026 interim month saw valid recommendation shortlist share fall to 65.9%, and September 2026 recovered it to 71.3% before the October 2026 reading of 90.5%. Recommendation density has climbed across the series rather than arriving in a single month.

AI Recommendation Trend

Questions This Section Answers

• Which small business lenders rank highest when AI recommendations are combined?
• How did the fintech tier shift against the July 2026 baseline?

Leadership held as the fintech tier consolidated above half the field

Bluevine retained 1st place, and three of the top four positions are now held by non-bank lenders, with onDeck at 57.7% and Fundbox at 56.6% separated by 1.1 points.

Bank of America and Chase appear in the July 2026 baseline at 56.5% and 49.0% under their prior tracking names and no longer carry coverage under those names in October 2026. No brand exceeded normal month-to-month variation in October 2026. The category-level shift came from the combination of several smaller movements in the same direction rather than from any single brand's repositioning.

What Changed This Month

Questions This Section Answers

• Which small business lenders made the largest baseline gains among AI recommendation coverage?
• What happened to Bank of America and Chase tracking in the benchmark?
• Did any brand's movement cross the significance threshold against the prior month?

Bluevine: leadership held, depth strengthened

Bluevine's valid recommendation coverage moved from 69.8% in July 2026 to 65.1% in October 2026, a 4.7-point decline that sits inside normal variation against baseline. The prior-month comparison is the opposite direction: coverage rose 5.2 points from 59.9% in September 2026, also inside normal variation.

The placement metrics tell a stronger story than the coverage line. The top-three rate climbed from 39.4% at baseline to 56.6%, a 17.2-point gain the benchmark treats as significant, and the rank-one rate rose from 14.4% to 27.8%, a significant 13.4-point gain. Bluevine recorded 105 first-position appearances in October 2026, up from 63 in July 2026, and 214 top-three placements, up from 172.

The distinction is between breadth and depth. Raw mention presence fell from 77.3% to 71.2%, a significant 6.1-point decline, so Bluevine is named in fewer answers than at baseline even as it is placed first far more often. Coverage declined while first-place placement nearly doubled. The consolidation is happening at the top of the recommendation set, not in the middle.

Highest-priority diagnostic: Which answer contexts have stopped naming Bluevine at all, and whether those contexts are concentrated on a specific AI surface or prompt type.

onDeck: significant baseline rise with stronger first-position presence

onDeck's valid recommendation coverage rose from 46.0% in July 2026 to 57.7% in October 2026, an 11.7-point increase the benchmark classifies as significant. The prior-month movement was 5.4 points from 52.3% in September 2026, and the series pattern shows a 41.5% trough in August 2026 before two consecutive monthly gains.

Raw mention presence rose from 54.0% to 65.6%, a significant 11.6-point gain. The top-three rate climbed from 24.9% to 44.4%, a significant 19.5-point gain, and the rank-one rate rose from 5.7% to 15.3%, a significant 9.6-point gain. onDeck recorded 168 top-three placements and 58 rank-one appearances in October 2026, against 109 and 25 in July 2026.

The distinction is that onDeck's gains appear across presence, top-three placement, and rank-one frequency simultaneously. Net sentiment held at 0.9 across the series with a single negative observation in October 2026, so the expansion is not accompanied by any shift in how the brand is framed.

Highest-priority diagnostic: Which competitor held the top-three placements that onDeck now occupies, and whether the gain is concentrated on one AI surface family.

Fundbox: strongest baseline riser on a three-month streak

Fundbox's valid recommendation coverage rose from 39.8% in July 2026 to 56.6% in October 2026, a 16.8-point increase the benchmark classifies as significant, and the largest baseline gain in the category. The prior-month movement was 5.1 points from 51.5% in September 2026. Fundbox has risen every month of the series, from 39.8% to 41.3% to 51.5% to 56.6%.

Raw mention presence rose from 47.6% to 59.5%, a significant 11.9-point gain, and the top-three rate climbed from 21.1% to 42.9%, a significant 21.8-point gain. The rank-one rate rose from 6.6% to 15.3%, a significant 8.7-point gain, taking Fundbox from 29 to 58 first-position appearances. Net sentiment reached 1.0 in October 2026 from 0.9 at baseline, with no negative observations in the current month.

The distinction is that Fundbox now sits 1.1 points behind onDeck for 2nd place. The two brands have converged from a 6.2-point gap at baseline to essentially level, and the split between them is narrower than either brand's gap to 4th place.

Highest-priority diagnostic: Which lending scenarios entered Fundbox's recommendation set during the three-month rise, and whether the 2nd-place position shifted between the two brands within the month.

Bank of America and Chase: entity change continues to define the series

The tracked entity Bank of America reads 56.5% in July 2026, 43.2% in August 2026, and 0.0% in both September 2026 and October 2026. Chase follows the same pattern at 49.0%, 36.2%, and 0.0%. The benchmark classifies both as significant decliners, but the mechanism is the appearance of Bank of America Corp. and Chase Credit Journey as replacement tracked entities.

Bank of America Corp. holds 24.1% coverage in October 2026, up 5.6 points from 18.5% in September 2026 on a two-month upward streak, with 91 valid recommendations out of 378 qualified observations. Its raw mention presence rate is 29.1%, its top-three rate is 17.5%, and its rank-one rate is 4.8%. Chase Credit Journey holds 0.3% coverage, representing a single valid recommendation, and 1 observation of presence out of 378.

The distinction is that the prior-name and new-name series are not directly comparable. The gap between Bank of America and Bank of America Corp. has widened every month of the series, from 56.5 points to negative 24.1 points, and the benchmark presents that as a category observation about tracking identity rather than a competitive repositioning.

Highest-priority diagnostic: How Bank of America Corp.'s 24.1% coverage compares with the prior entity on equivalent prompt sets once the two tracking identities can be matched.

Lendio, National Funding, Biz2Credit and the small-count tier

Lendio holds 36.5% coverage in October 2026, up 5.1 points from 31.4% in July 2026, inside normal variation, though the underlying metrics moved further: top-three rate up 14.0 points to 23.8% and rank-one rate up 4.8 points to 8.5%, both significant. Lendio has risen for two consecutive months, from 26.1% in August 2026 through 33.5% in September 2026.

National Funding holds 9.5% coverage, down 2.2 points from 11.7% at baseline, inside normal variation and on a one-month downward streak. Its top-three rate rose 2.6 points to 5.8% while its rank-one rate was essentially flat at 0.8%. The brand recorded 36 valid recommendations in October 2026, down from 51 in July 2026.

Biz2Credit holds 5.8% coverage, up 2.1 points from 3.7% at baseline, inside normal variation but on a three-month upward streak the benchmark flags as a watch item. Its raw mention presence rate rose 7.0 points to 12.7%, a significant gain, while rank-one frequency fell to 0.0% from 0.5%. Biz2Credit recorded 22 valid recommendations and 0 rank-one appearances in October 2026.

Funding Circle holds 2.1% coverage, flat against baseline, with 8 valid recommendations and 0 rank-one appearances. QuickBridge holds 3.4% coverage, up 0.6 points from 2.8%, with 13 valid recommendations and 5 rank-one appearances. Both operate on counts small enough that single-observation changes move the percentage materially.

Highest-priority diagnostic: Whether Biz2Credit's rising presence is converting into top-three placement in a durable way, and whether the small-count brands appear in a narrow set of lending scenarios.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters did the benchmark actually measure?
• What does the benchmark fail to capture about price sensitivity and head-to-head comparisons?

All 378 qualified observations in October 2026 fell into the Brand Recommendation class. The benchmark found no qualified observations in the Pricing & Value or Multi-Brand Comparison classes. The raw response-type distribution included 38 comparison analysis responses and 1 pricing analysis response, but those did not qualify into the separate buyer-intent clusters.

The practical consequence is that the public benchmark measures who gets named and placed, and does not yet measure at what price or against which specific alternative. The inconsistency data does show pricing and eligibility claims circulating widely, with Bluevine's starting rate disputed between 7.8% and 14% or higher and Fundbox's eligibility thresholds varying by tens of thousands of dollars. Those claims are present in AI answers; they simply did not qualify as pricing or comparison observations. Price sensitivity, value positioning, and head-to-head preference therefore remain questions the aggregate percentages cannot answer.

Brand Opportunity Summary

Questions This Section Answers

• Which small business lenders have the strongest AI recommendation position but unresolved coverage questions?

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

Questions This Section Answers

• What data layer supports the coverage and placement metrics in this report?
• How should the inconsistency findings be interpreted separately from the recommendation metrics?

The aggregate metrics are built from prompt-level observations (query, surface, recommendation outcome, rank, sentiment, and citations where exposed). Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence is not automatically treated as proof of causation. The inconsistency data in this report is a separate signal from the coverage and placement metrics: it identifies where AI platforms gave conflicting answers on a specific factual claim, with the cited source pages on each side recorded where they were exposed.

About This Benchmark

This report is part of the LLM Authority Index AI Visibility Market Discovery research program.

• AI Visibility Industry Market Methodology
• AI Visibility Industry Market Metrics
• AI Visibility Industry Market Standards

Report-Specific Interpretation Notes

Small-count movement is meaningful in this vertical. Funding Circle, Chase Credit Journey, QuickBridge, and Biz2Credit operate on valid recommendation counts in the single digits to low twenties, so a handful of observations can move their percentages materially. Chase Credit Journey's 0.3% coverage represents a single valid recommendation out of 378 qualified observations, and Funding Circle's 2.1% represents 8.

The qualified denominator differs from the raw collection universe. All metrics in this report use the qualified benchmark observations (378 in October 2026), not the 800 raw prompt-surface observations. The two banks tracked under prior names in the July 2026 baseline no longer carry coverage under those names, and the replacement entities are not directly comparable to the prior series.

Directional analysis identifies movement worth investigating; it does not by itself establish cause, and no claim in this report should be read as attributing a brand's movement to any specific action by that brand or its partners. The inconsistency findings describe conflicting AI responses and their cited sources; they do not establish which claim is correct.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

Beneath the aggregate percentage sit the questions that determine competitive position: which high-intent prompts a brand wins, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. The inconsistency findings sharpen those questions, because a lending term that one platform states as $30,000 and another as $100,000 changes which borrower sees the brand as available to them at all.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. Where the public benchmark shows movement or conflicting claims, the audit identifies the mechanism and the path to remediation.

Request an AI visibility audit