How AI Search Is Recommending Home Equity Loans: Monthly Trends

Executive Summary

The top of the home equity loans category held firm in October 2026 while the field beneath it reshuffled. Navy Federal Credit Union leads with 72.8% valid recommendation coverage, ahead of Figure at 58.2% and Aven at 40.7%. Navy Federal Credit Union's lead over Figure widened to 14.6 points from 10.3 points in September 2026. Two significant decliners and four significant risers moved the category's middle and lower tiers this month.

Bank of America Corp. is the month's largest single gain, entering the tracked set at 55.5% valid recommendation coverage in October 2026. No earlier baseline value exists for this entity (its series reads 0.0, 0.0, 55.5 across Aug 2026, Sep 2026, and Oct 2026), a shift reflecting a change in how the entity is labelled rather than a measured baseline-to-current move. Aven rose 9.6 points from 31.1% in August 2026 to 40.7% in October 2026, its second consecutive month of significant gain, and Spring EQ rose 5.2 points from 3.9% to 9.1%. U.S. Bancorp entered the tracked set at 23.6% coverage.

The sharpest decliner is Bank of America, which no longer appears in the October 2026 tracked set: its series fell from 66.3% in August 2026 to 69.6% in September 2026 to 0.0% in October 2026. U.S. Bank followed the same pattern, falling from 31.1% in August 2026 to 35.7% in September 2026 to 0.0% in October 2026. Both declines reflect a change in entity labelling across the tracked set, not a month-over-month coverage loss for the brands themselves.

Against the full baseline-to-current series, October 2026 was a month of entity reclassification and mid-field movement rather than a change at the top. Navy Federal Credit Union extended a two-month upward streak, Figure's coverage was effectively flat, and the most notable commercial story is the broadening of the mid-tier, where Aven, Spring EQ, and U.S. Bancorp all registered significant gains.

Each monthly run begins with 800 prompt-surface observations (619 unique questions in October 2026) across the benchmark's defined AI/search surface universe. In October 2026, 800 mentioned a tracked brand or competitor; 418 were relevant and 382 were irrelevant. The public metrics use the 364 observations that survive both qualification stages. For comparison, August 2026 began with 800 prompt-surface observations (626 unique questions); 374 were relevant and 426 were irrelevant, with 338 qualified observations forming the public denominator.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• Which lenders had conflicting rate and fee information across AI platforms this month?
• Where did AI platforms disagree about Figure's APR, loan limits, and credit score requirements?
• What inconsistencies surfaced around Aven's credit line limits and fee structure?

The benchmark detected 7 critical or high-severity factual inconsistencies across 7 AI platforms in October 2026, spanning Figure, Aven, and PNC Bank.

Figure

AI platforms provided conflicting information about Figure's HELOC APR. When asked "What is a good HELOC rate today?", AI Overviews stated fixed APRs starting around 6.05% to 6.40%, citing Bankrate, The Mortgage Reports, and a Reddit thread, while Copilot stated 7.49% APR, citing HELOCalculator, RateAPI, and LendingTree. One flagged source on the HELOCalculator page read "The best available rate is 7.49% from Figure Home." These are incompatible non-overlapping values for the same lender's rate.

Responses also differed on Figure's maximum loan amount. When asked "Which Bank is best for a HELOC?", Gemini stated credit lines "up to $750,000", citing The Mortgage Reports, while Copilot stated a range of "$50k–$500k". A flagged Bankrate source was cited on both sides with an excerpt referencing competing lender ranges.

AI platforms further differed on Figure's minimum credit score requirement. When asked "best heloc lenders", Google AI Mode stated an estimated minimum credit score of 600, while Google AI Overviews stated a minimum credit score around 640, citing Bankrate, The Mortgage Reports, and CNBC. A flagged source was referenced on both sides with an excerpt showing both a 620 and a 600 threshold.

Aven

AI platforms provided conflicting information about Aven's maximum credit line. When asked "Who gives the best HELOC loans?", Gemini stated "credit lines up to $750,000", citing two Mortgage Reports pages, while ChatGPT stated "lines up to $400,000 in eligible states", citing Bankrate, NerdWallet, and Forbes. A flagged Mortgage Reports source on the Gemini side referenced Figure's credit line specifically at up to $750,000.

Responses also differed on Aven's origination and annual fees. When asked "best heloc lenders", Google AI Mode stated that fintech and online lenders like Aven frequently charge an upfront origination fee up to 4.99%, while Google AI Overviews stated that Aven has no origination or annual fees, citing Bankrate, Credible, and Investopedia. These two positions cannot both be accurate for the same lender.

PNC Bank

AI platforms provided conflicting information about PNC Bank's personal loan starting APR. When asked "Which bank has the lowest interest rates for personal loans?", Google AI Overviews stated rates start around 6.99% APR including an autopay discount, citing Wells Fargo, Credible, and CNBC, while Google AI Mode stated rates start around 7.89% to 8.14% depending on zip code, citing TD Bank, Credible, and Upstart. A flagged NerdWallet source cited an estimated APR range starting at 6.99%.

Responses also differed on PNC Bank's HELOC rate range. When asked "home equity line of credit rates today", Google AI Mode stated variable rates ranging from 7.49% to 14.50%, citing Bank of America, Navy Federal Credit Union, and Bankrate, while Google AI Overviews stated a 5.74% introductory APR on lines of $50,000 or more, citing Navy Federal Credit Union, Bank of America, and U.S. Bank. These are incompatible starting rates for the same product.

Benchmark Context

Questions This Section Answers

• How many prompt observations survived qualification for the October 2026 benchmark?
• What does the qualified benchmark denominator represent compared to the raw collected prompts?

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

The following benchmark-level metrics summarize category-wide behavior across the tracked brands.

Benchmark-Level Metrics

The valid recommendation shortlist share rose to its highest level across the three-month series. September 2026 sat between the two endpoints, with a valid recommendation shortlist share of 78.5% and a recommendation-shaped answer share of 46.9%.

AI Recommendation Trend

Questions This Section Answers

• Who leads the home equity loan category, and how did the top tier hold up this month?
• Which brands had the largest coverage movement between August and October 2026?
• How does the entity labelling change affect how the Bank of America and U.S. Bank rows should be read?

A Stable Top Tier Over a Reshuffled Mid-Field

Bank of America Corp. (55.5%), U.S. Bancorp (23.6%), and Bank of America (0.0%) and U.S. Bank (0.0%) are carried in the table because they appear in the tracked set across the series. The Bank of America and U.S. Bank rows reflect a change in how these entities are labelled in the October 2026 run; their coverage did not fall to zero in a measured month-over-month sense. Aven, Bank of America Corp., Spring EQ, and U.S. Bancorp are each classified as significant risers for the baseline-to-current period. Rocket Mortgage's 6.3-point baseline-to-current gain was not classified as significant, but its prior-to-current move of 7.3 points was.

What Changed This Month

Questions This Section Answers

• Which prompts converted added presence into recommendations for Navy Federal Credit Union?
• What drove Aven's rank-one placement gains, and did they concentrate in one buyer question type?
• Why do the top-of-category coverage gaps look wider this month?

Navy Federal Credit Union

Navy Federal Credit Union's valid recommendation coverage rose 6.8 points, from 66.0% in August 2026 to 72.8% in October 2026, extending a two-month upward streak. The movement is classified as stable for the period. Its prior-to-current move of 5.0 points from September 2026 was also not classified as significant.

The brand's raw mention presence moved the other way, down 2.6 points from 81.7% in August 2026 to 79.1% in October 2026, while its top-three rate rose 0.8 points to 28.0% and its rank-one rate rose 1.5 points to 3.6%. It registered 265 valid recommendations in October 2026 against 223 in August 2026 on a base of 364 qualified observations.

Navy Federal Credit Union leads the category on coverage while its raw presence slipped. The brand is seen slightly less often but is converted into a recommendation more often, and its placement rates improved modestly alongside the coverage gain. Strength here is about conversion rather than visibility.

Highest-priority diagnostic: Which prompts converted added presence into recommendations, and whether the top-three rate can be pushed toward the coverage rate.

Aven

Aven's valid recommendation coverage rose 9.6 points, from 31.1% in August 2026 to 40.7% in October 2026, its second consecutive month of significant gain. The prior-to-current move from September 2026 was 8.8 points, also classified as significant.

The gains were broad. Raw mention presence rose 9.2 points to 42.3%, the top-three rate rose 14.4 points from 9.2% to 23.6%, and the rank-one rate rose 8.3 points from 4.1% to 12.4%. The brand registered 148 valid recommendations and 45 rank-one placements in October 2026, against 105 valid recommendations and 14 rank-one placements in August 2026.

Aven is the clearest riser in the category, and the gains span presence, shortlist placement, and top-rank placement at once. The gains span presence, shortlist placement, and top-rank placement at once, distinguishing Aven from brands whose coverage rose without a matching placement gain.

Highest-priority diagnostic: Which prompts and surfaces account for the rank-one gains specifically, and whether they concentrate in one buyer question type.

Bank of America and Bank of America Corp.

Bank of America Corp. appears in the October 2026 tracked set at 55.5% valid recommendation coverage, placing third in the category. Its raw mention presence is 64.3%, its top-three rate is 44.2%, and its rank-one rate is 21.4%. It registered 202 valid recommendations and 78 rank-one placements in October 2026.

Bank of America does not appear in the October 2026 tracked set, following coverage of 66.3% in August 2026 and 69.6% in September 2026. Its August 2026 supporting figures were 89.0% raw presence, 47.0% top-three, and 23.1% rank-one on 224 valid recommendations.

Bank of America Corp. is not a new entrant to the category in the commercial sense; it is a relabelled entity carrying the same market position Bank of America held. The October 2026 figures sit somewhat below Bank of America's September 2026 coverage, and the benchmark records them under the new label without asserting that the underlying market position changed.

Highest-priority diagnostic: How much of the difference between the two labels reflects genuine coverage change versus a difference in entity naming alone.

U.S. Bancorp and U.S. Bank

U.S. Bancorp appears in the October 2026 tracked set at 23.6% valid recommendation coverage, ranking seventh. Its raw mention presence is 29.7%, its top-three rate is 11.8%, and its rank-one rate is 3.9%. It registered 86 valid recommendations and 14 rank-one placements in October 2026.

U.S. Bank does not appear in the October 2026 tracked set, following coverage of 31.1% in August 2026 and 35.7% in September 2026. Its August 2026 supporting figures were 53.2% raw presence, 13.6% top-three, and 3.0% rank-one on 105 valid recommendations.

U.S. Bancorp covers a materially smaller share of qualified observations than U.S. Bank did in September 2026, and the gap between the two labels is larger than the gap between Bank of America and Bank of America Corp. As with Bank of America, the comparison is complicated by the labelling change, and the benchmark does not assert that the underlying market position moved by the full difference.

Highest-priority diagnostic: Whether any of the difference between the two labels reflects lost prompt coverage, and whether that pattern concentrates on particular surfaces.

Bank of America's Widened Gap with the New Labels

Notable gaps widened across the top of the category this month. Bank of America's gap over U.S. Bancorp moved from 66.3 points to -23.6 points, a widening of 89.9 points, driven by the labelling change and U.S. Bancorp's lower coverage. Bank of America's gap over Bank of America Corp. widened by 121.8 points on the same basis. Bank of America Corp.'s gap over U.S. Bank moved from -31.1 points to 55.5 points, a widening of 86.6 points.

The category's gaps at the top look larger this month primarily because the tracked set has been re-labelled, not because the underlying competitive positions diverged by that magnitude. The benchmark records the gap widening as an observation about the present tracked set. The category-level repositioning is worth reading with the labelling change in mind rather than as a signal of one brand's problem or another's breakthrough.

Highest-priority diagnostic: Which entity label the surfaces themselves use when they return either Bank of America or U.S. Bank variants, and whether surface-level naming splits the coverage.

Rocket Mortgage

Rocket Mortgage's valid recommendation coverage rose 6.3 points, from 23.4% in August 2026 to 29.7% in October 2026; the baseline-to-current move was not classified as significant. The prior-to-current move from September 2026 was 7.3 points, which was classified as significant.

Its top-three rate rose 6.2 points to 22.5%, while its rank-one rate fell 2.4 points from 10.1% to 7.7%. Raw mention presence rose 4.0 points to 37.4%. The brand registered 108 valid recommendations in October 2026 against 79 in August 2026.

Rocket Mortgage is being added to more shortlists without a matching gain in top-rank placement, a pattern that separates shortlist visibility from lead-position strength. The prior-to-current coverage move was the sharper of the two windows.

Highest-priority diagnostic: Which prompts added Rocket Mortgage to the top three without elevating it to rank one, and which brand typically takes the lead position.

Spring EQ

Spring EQ's valid recommendation coverage rose 5.2 points, from 3.9% in August 2026 to 9.1% in October 2026, classified as a significant riser. The prior-to-current move from September 2026 was 5.0 points, also classified as significant.

Raw mention presence rose 6.3 points to 10.7%, and the top-three rate rose 2.7 points to 3.3%. The rank-one rate rose 0.3 points to 0.3%. The brand registered 33 valid recommendations and one rank-one placement in October 2026.

Spring EQ's counts remain small, and each observation carries meaningful weight in its percentage. The brand moved from 13 to 33 valid recommendations across the two months, and the small absolute base means rates should be read alongside those counts.

Highest-priority diagnostic: Whether the added recommendations cluster on a narrow set of prompts or spread across surfaces.

PNC Bank and TD Bank

PNC Bank's valid recommendation coverage fell 3.2 points, from 45.0% in August 2026 to 41.8% in October 2026, a stable movement. Its prior-to-current move from September 2026 was 2.7 points, also stable, marking a second consecutive month of decline. Its top-three rate rose 2.0 points to 20.6% while its rank-one rate and raw presence both slipped slightly. It registered 152 valid recommendations in October 2026.

TD Bank's valid recommendation coverage was effectively flat, up 0.1 points from 13.6% in August 2026 to 13.7% in October 2026. Its prior-to-current move from September 2026 was 4.9 points, classified as significant; the September 2026 figure of 8.8% was the low point in the series. TD Bank registered 50 valid recommendations in October 2026, against 46 in August 2026.

PNC Bank's coverage has drifted lower over two months while TD Bank has recovered the loss it registered in September 2026. Both movements are modest, and neither brand's position moved relative to the coverage leader.

Highest-priority diagnostic: For PNC Bank, which prompts dropped out of its coverage; for TD Bank, whether the September 2026 dip was temporary and what prompts returned.

Buyer-Intent Interpretation

Questions This Section Answers

• What buyer-intent clusters did the October 2026 qualified observations capture?
• Can the benchmark answer pricing, value, or head-to-head comparison questions for home equity loans?

In October 2026, all 364 qualified observations fell into the Brand Recommendation cluster, with no qualified observations captured in the Pricing & Value or Multi-Brand Comparison clusters. The August 2026 run showed the same distribution, with all 338 qualified observations in the Brand Recommendation cluster. The benchmark can therefore describe which brands AI systems recommend for home equity loans, but it cannot yet answer price, value, or head-to-head comparison questions in this vertical, even though seven pricing and factual inconsistencies surfaced across the AI surfaces this month.

Brand Opportunity Summary

Questions This Section Answers

• Which brands are the significant risers and decliners in the October 2026 tracked set?
• What diagnostic should each brand investigate to understand its current position?

The benchmark identifies where attention is warranted across the field; a company-level analysis is needed to explain why each brand sits where it does.

Evidence Behind the Benchmark

Questions This Section Answers

• What prompt-level data is used to calculate the aggregate metrics?
• How should small-count movements and the entity labelling change be interpreted?

The aggregate metrics are built from prompt-level observations capturing the 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.

Report-Specific Interpretation Notes

• Small-count movement: Brands such as Discover Home Loans (9 valid recommendations) and Spring EQ (33 valid recommendations) have rates that can move sharply with a small number of changes. Their percentages should be read with the absolute counts in mind.
• Qualified denominator: All coverage rates are calculated against the qualified observations, not the raw 800 prompts or the unique question counts. August 2026 used 338 qualified observations and October 2026 used 364.
• Entity labelling: The October 2026 tracked set uses Bank of America Corp. and U.S. Bancorp where earlier months used Bank of America and U.S. Bank. Coverage movements between the two labels carry the labelling change as well as any genuine movement.
• Directional analysis: Month-over-month movement identifies changes worth investigating. It does not by itself establish the cause of those changes. Aven's gains and PNC Bank's two-month decline are signals for further inspection, not conclusions.

About This Benchmark

This report is part of the CiteWorks Studio AI Visibility Industry Market research program.

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

Next Step

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

The aggregate percentage masks the questions that matter: which high-intent prompts are won, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. For a brand moving 9.6 points across the series, or one whose coverage depends on an entity label, the difference between a strong and a weak month often lives in the prompts, surfaces, and evidence sources beneath the number.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. It answers why a brand rose or fell, and what would need to change to move the number in the intended direction.

Request an AI visibility audit