How AI Search Is Recommending Mortgage Refinance Lenders: Monthly Trends

Executive Summary

Rocket Mortgage remains the category leader in AI search recommendations for mortgage refinance lenders, with valid recommendation coverage of 79.2% in October 2026. That is up from 71.2% in July 2026, a gain of 8.0 points that exceeded normal month-to-month variation. loanDepot holds second position at 49.3%, a gap of nearly 30 points behind Rocket Mortgage on the coverage measure that drives this report.

loanDepot was the strongest sustained mover in the series. Its coverage rose from 33.9% in July 2026 to 49.3% in October 2026, a gain of 15.4 points that exceeded normal variation, and its September-to-October movement of 12.8 points was also significant. The August-to-October trend is a two-month upward streak for the brand, making it the clearest sustained riser in the tracked set.

Bank of America Corp. also extended an upward streak, reaching 35.5% coverage in October 2026 from 27.9% in September. That is up 35.5 points against the July baseline under its current tracked name, and up 7.6 points from the prior month. The sharpest decliner against the July baseline was Better.com, which fell from 33.8% to 27.6%, a decline of 6.2 points that exceeded normal variation. Several brand names that held coverage in July (Chase at 54.5%, Bank of America at 55.4%, Wells Fargo at 20.8%, Guaranteed Rate at 15.2%) recorded zero coverage under those tracked names in October, the same naming reconfiguration the benchmark flagged in September.

The October 2026 run recorded 529 qualified observations, down from 570 in September and from the 607 July baseline. The category was mixed: six brands posted significant gains against the July baseline and five posted significant declines, with the movement concentrated in coverage rather than presence.

Each monthly run begins with 800 prompt-surface observations (584 unique questions in October 2026) across the benchmark's defined AI/search surface universe. Of those, 800 mentioned a tracked brand or competitor; 678 were relevant and 122 were irrelevant. The public metrics use the 529 observations that survive both qualification stages. For comparison, the July 2026 baseline began with 800 prompt-surface observations (541 unique questions), of which 800 mentioned a tracked brand or competitor; 624 were relevant and 176 were irrelevant, yielding 607 qualified observations. The September 2026 intermediate run produced 570 qualified observations from the same 800-prompt collection.

Key Findings

AI Response Inconsistency Alerts

The benchmark detected 23 critical and high-severity factual inconsistencies across six AI platforms in this month's review. The inconsistencies below involve conflicting claims about lender rankings, pricing, and product terms that AI platforms surfaced when queried directly.

United Wholesale Mortgage

Largest FHA lender by loan count (critical). AI platforms provided conflicting information about which lender originates the most FHA loans. When asked "Who is the largest FHA lender?", Google AI Overviews stated that Pennymac leads by number of FHA loans issued, citing Scotsman Guide's 2024 FHA lender rankings, while Copilot stated that United Wholesale Mortgage originated 56,684 FHA purchase loans in 2025, nearly twice as many as the next biggest lender, citing Mortgage Research and Scotsman Guide's 2025 rankings.

Largest FHA lender overall (critical). A second conflict on the same question produced different claims. ChatGPT stated that Pennymac has recently ranked as the largest FHA lender by volume while UWM has been among the largest by loan count, while Gemini stated that UWM is the largest FHA lender overall, consistently leading in total FHA loan volume and purchase loans. Both cited Mortgage Research and Scotsman Guide data.

VA loan volume ranking (high). When asked "Who is the biggest VA lender?", Google AI Overviews stated that UWM issued 36,637 VA loans worth $15.6 billion, ranking second behind Veterans United, citing CNBC and VA Lender Statistics, while ChatGPT stated that UWM ranked at the top in 2025 with about $31 billion in VA loan volume across more than 71,000 loans, citing Mortgage Research and Veterans United's newsroom.

Largest FHA lender by volume, question variant (high). When asked "Who is the best FHA lender?", ChatGPT stated that Pennymac was the largest FHA lender by 2025 volume, citing Bankrate, while Gemini stated that UWM is the largest overall FHA lender, citing Mortgage Research and Credit Karma.

Freedom Mortgage vs. UWM claim (high). A second "Who is the best FHA lender?" query produced contradictory responses: Copilot stated that Freedom Mortgage is the number one FHA lender by volume, citing mortgage-info.com, while Gemini stated that UWM is the largest FHA lender overall, citing Mortgage Research.

Loan volume leadership (high). When asked "What is the best mortgage lending company?", Google AI Mode stated that Rocket Mortgage is the nation's largest mortgage lender by loan volume, citing US News, while Copilot stated that UWM leads in loan volume with $162 billion in 2025, citing The Motley Fool and The Truth About Mortgage.

Pennymac

Largest FHA lender by overall volume (critical). AI platforms differed on FHA leadership. When asked "Who is the largest FHA lender?", Google AI Overviews stated that Pennymac is currently ranked as the largest FHA lender in the United States by overall volume and number of loans issued, citing Scotsman Guide 2024 rankings and Bankrate, while Gemini stated that UWM is the largest FHA lender overall, citing The Truth About Mortgage and Mortgage Research.

Largest FHA lender by volume, question variant (high). When asked "Who is the best FHA lender?", Copilot stated that Freedom Mortgage is the number one FHA lender by volume, citing mortgage-info.com, while ChatGPT stated that Pennymac was the largest FHA lender by 2025 volume in Bankrate's data, citing Bankrate.

Rocket Mortgage

30-year fixed mortgage rate (high). When asked "What lender has the lowest mortgage rates right now?", Google AI Overviews stated a 30-year fixed rate around 7.50%, citing Bankrate and Credit Karma, while Copilot stated a 30-year fixed rate of 6.09%, citing Branchspot and mortgage-info.com. The two rates cannot both be current.

30-year fixed rate, second query (high). When asked "Who is offering the lowest mortgage rates?", Google AI Mode stated 6.875%, citing Rocket Mortgage's own rate page, while Copilot stated 6.09%, citing mortgage-info.com.

Rank by loan count and volume (high). When asked "Who are the top 6 mortgage lenders?", Google AI Mode stated that Rocket Mortgage is ranked second with UWM ranked first overall, citing Bankrate and The Motley Fool, while Google AI Overviews stated that Rocket Mortgage is ranked first overall by volume, citing Bankrate and CNBC.

30-year jumbo fixed purchase rate (high). When asked "What are the current rates at Rocket Mortgage?", Google AI Mode stated a 30-year jumbo fixed rate of 6.125% (6.351% APR), while Google AI Overviews stated 6.5% (6.733% APR). Both cited Rocket Mortgage's own mortgage rates page.

30-year VA purchase rate (high). On the same query, Google AI Mode stated 6.125% (6.523% APR) for the 30-year VA purchase rate, while Google AI Overviews stated 6.625% (7.038% APR). Both cited Rocket Mortgage's rate pages including the Oklahoma state page.

30-year FHA purchase rate (high). On the same query, Google AI Mode stated 6.125% (6.972% APR) for the 30-year FHA purchase rate, while Google AI Overviews stated 6.625% (7.494% APR), again both citing Rocket Mortgage's own rate content.

30-year fixed purchase rate (high). On the same query, Google AI Mode stated 6.75% (7.052% APR) for the 30-year fixed purchase rate, while Google AI Overviews stated 7.125% (7.436% APR), both citing Rocket Mortgage's mortgage rates and rate updates pages.

Minimum home equity loan amount (high). When asked "Is Rocket Mortgage good for HELOCs?", Google AI Mode stated that loan amounts range from $2,000 to $500,000, while Google AI Overviews stated that borrowers must take at least $45,000 (loan amounts of $45,000 to $500,000), citing LendEDU, CNBC, and Finder reviews.

30-year FHA mortgage rate, second query (high). When asked "What are current rates at Rocket Mortgage?", Google AI Mode stated a 30-year FHA rate of 6.25% (7.076% APR), while Google AI Overviews stated 5.875%. Both cited Rocket Mortgage's own content.

Largest overall mortgage lender (high). When asked "What are the major mortgage companies?", Google AI Mode stated that UWM is the nation's largest overall mortgage lender, citing Experian, while Google AI Overviews stated that Rocket Mortgage is the nation's largest overall mortgage lender and top loan originator, citing CNBC and Bankrate.

30-year fixed rate, national average comparison (high). When asked "What bank has the lowest interest rate for mortgages?", Google AI Mode stated there is no single lowest rate, with national averages for a 30-year fixed around 6.65% to 6.78%, while Copilot stated that Rocket Mortgage offers the lowest 30-year fixed rate at 5.99%, citing mortgage-info.com. The 5.99% figure sits well below the range cited in the other response.

30-year VA mortgage rate, second query (high). When asked "What are current rates at Rocket Mortgage?", Google AI Mode stated a 30-year VA rate of 6.25% (6.639% APR), while Google AI Overviews stated 5.875%. Both cited Rocket Mortgage's own content.

30-year fixed rate, third query (high). When asked "What are Rocket Mortgages' rates today?", Google AI Overviews stated 6.75% (7.039% APR), citing Rocket Mortgage's 30-year rate page and refinance rates page, while Perplexity stated a range of around 6.3% to 6.6%, citing SmartAsset and Rocket Mortgage's rate pages.

loanDepot

Top five mortgage lender ranking (high). When asked "What are the top 5 mortgage lenders in the US?", Copilot stated that loanDepot is one of the top 5 mortgage lenders, citing RE Report and mortgage-info.com, while Google AI Mode stated that the top five are Rocket, UWM, CrossCountry, Pennymac, and JPMorgan Chase, with loanDepot not listed, citing Bankrate and The Motley Fool.

Benchmark Context

Questions This Section Answers

• How many prompt observations qualified for the October 2026 mortgage refinance benchmark, and how does that compare with July?
• Why is the qualified observation set the denominator for every lender-level recommendation metric?

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

The qualified benchmark set is the denominator behind every brand-level metric that follows. The August 2026 intermediate run produced 381 qualified observations and September produced 570, so October's 529 sits between those two months and below the July baseline scale.

Benchmark-Level Metrics

AI Recommendation Trend

Questions This Section Answers

• Which mortgage refinance lenders ranked highest in AI recommendation coverage in October 2026?
• Which brands posted significant coverage gains or declines against the July 2026 baseline?

Rocket Mortgage Extended Its Lead While loanDepot Closed on Second

Rocket Mortgage held first position in October 2026 at 79.2% coverage, and loanDepot held second at 49.3%, a gap of nearly 30 points. Bank of America Corp. moved into third at 35.5%. Several brands exceeded normal month-to-month variation against the July baseline: six brands posted significant gains and five posted significant declines. Rocket Mortgage, loanDepot, Bank of America Corp., Rate, United Wholesale Mortgage, and Wells Fargo & Co. all posted significant gains against the baseline, while Chase, Bank of America, Wells Fargo, Guaranteed Rate, and Better.com posted significant declines. The declines for Chase, Bank of America, Wells Fargo, and Guaranteed Rate are tied to the tracked entity-name reconfiguration the benchmark flagged in September, not a clearly established loss of recommendation strength.

What Changed This Month

Questions This Section Answers

• What drove Rocket Mortgage's increase in rank-one placements this month?
• How did loanDepot's October gain differ from earlier movement in the series?
• Why did Bank of America, Wells Fargo, and Chase show zero coverage under their July tracked names?

Rocket Mortgage

Rocket Mortgage's valid recommendation coverage measured 79.2% in October 2026, compared with 71.2% in July 2026. The gain of 8.0 points exceeded normal month-to-month variation, and the September-to-October movement of 15.5 points was also significant. Valid recommendations numbered 419 in October from a qualified base of 529.

Presence remained near saturation at 98.1%, unchanged from 97.7% in July, with the brand appearing in 519 of 529 qualified observations. Placement strengthened alongside coverage: the top-three rate rose from 55.4% to 69.9%, and the rank-one rate rose from 38.1% to 49.0%, with the brand's top-three count rising from 336 to 370.

The distinction to notice: Rocket Mortgage is both present in nearly every response and named first or within the top three more often than in July, a combination that only a handful of brands in any vertical achieve.

Highest-priority diagnostic: Which prompt families drove the increase in rank-one placements, and is that placement concentrated on particular AI surfaces?

loanDepot

loanDepot held second place in October 2026 with valid recommendation coverage of 49.3%, up from 33.9% in July 2026. The gain of 15.4 points against baseline exceeded normal variation, and the September-to-October movement of 12.8 points was also significant, extending a two-month upward streak. Valid recommendations grew from 206 in July to 261 in October.

Presence rose from 43.7% to 56.7%, a gain of 13.0 points, with the brand appearing in 300 of 529 observations. Placement also improved: the top-three rate rose from 13.3% to 23.8%, and the rank-one rate moved from 2.1% to 2.8%.

Unlike the pattern observed earlier in the series, loanDepot's October gain came from both broader presence and stronger placement. The brand is being mentioned in more answers and recommended within the top three more often.

Highest-priority diagnostic: Which competitive comparisons are placing loanDepot in top-three slots, and does that placement hold across all six AI surface families?

Bank of America and Bank of America Corp.

Bank of America, as tracked in July and August 2026, recorded valid recommendation coverage of 55.4% in July and 51.2% in August. In September and October 2026, that entity name recorded 0.0% coverage, while Bank of America Corp. rose from 27.9% coverage in September to 35.5% in October, a gain of 7.6 points that exceeded normal variation. Tracking the two named entities together, the gap between them moved by 90.9 points across the series and widened every month.

Bank of America Corp. achieved its 35.5% coverage from 188 valid recommendations among 529 qualified observations. Presence measured 44.6%, with the entity appearing in 236 observations. Its top-three rate was 18.7%, and its rank-one rate was 3.0%, with 16 first-place finishes.

The commercial reading: the underlying lender shows steady AI presence under the corporate entity name, but the benchmark now captures it under a different tracked name than it did in July. The apparent collapse of Bank of America and the rise of Bank of America Corp. line up with a change in tracked entity names, not necessarily a change in how AI systems treat the lender.

Highest-priority diagnostic: Which prompts drive Bank of America Corp.'s top-three placements, and how should the benchmark track the lender's entity names consistently going forward?

Better.com

Better.com's coverage measured 27.6% in October 2026, down from 33.8% in July 2026, a decline of 6.2 points that exceeded normal variation. The September-to-October movement of 0.4 points was within normal range, indicating the decline occurred between July and August and has since stabilized. Valid recommendations numbered 146 in October versus 205 in July.

Presence fell from 41.3% to 28.7%, a decline of 12.6 points, with the brand appearing in 152 of 529 observations. Placement rose: the top-three rate moved from 16.5% to 19.1%, and the rank-one rate from 2.3% to 2.5%. Net sentiment remained the strongest among tracked brands at 0.97.

Better.com's pattern is distinct: it is mentioned far less often, but when mentioned it is placed in the top three more often than in July. The decline is a presence story, not a placement story.

Highest-priority diagnostic: Which prompt families account for the loss of presence, and does the improved placement signal a narrower but stronger recommendation footprint?

PennyMac

PennyMac's valid recommendation coverage measured 26.5% in October 2026, down from 27.7% in July 2026, a decline within normal variation. The September-to-October movement of 6.5 points was a significant gain, indicating the brand recovered after an August dip. Valid recommendations numbered 140 in October versus 168 in July.

Presence measured 37.4%, down from 40.5% in July. The top-three rate was 7.9% versus 6.9%, and the rank-one rate was 1.7% versus 1.8%. Net sentiment measured 0.71.

PennyMac's classification is stable against the July baseline despite its prior-month dip, and its October placement rates are modestly stronger than July. The brand is being mentioned slightly less often but recommended within the top three slightly more often.

Highest-priority diagnostic: Which surfaces drove the September-to-October recovery, and is the pattern repeatable?

United Wholesale Mortgage

United Wholesale Mortgage's coverage rose from 13.3% in July 2026 to 18.3% in October 2026, a gain of 5.0 points that exceeded normal variation. The September-to-October gain of 9.9 points was also significant. Valid recommendations rose from 81 to 97.

Presence rose from 25.5% to 28.9%, with the brand appearing in 153 of 529 observations. Placement improved: the top-three rate rose from 9.1% to 14.2%, and the rank-one rate from 2.3% to 4.5%, the brand's strongest rank-one rate in the series.

United Wholesale Mortgage's return to growth came through both broader presence and stronger placement, reversing the prior-month decline pattern. The brand's rank-one rate now sits above several higher-coverage brands.

Highest-priority diagnostic: Which prompt families account for the increase in rank-one placements?

New American Funding

New American Funding recorded coverage of 17.2% in October 2026, down from 19.9% in July 2026. The decline of 2.7 points did not exceed normal variation. The September-to-October movement of 1.2 points was also within normal range. Valid recommendations numbered 91 in October versus 121 in July.

Presence measured 18.9%, close to the 21.2% recorded in July. The top-three rate was 4.2% versus 3.8%, and the rank-one rate was 0.8% versus 0.5%. Net sentiment stayed high at 0.91.

New American Funding remains stable against the July baseline despite a smaller qualified observation base. Its placement metrics improved modestly, but its absolute counts remain modest relative to the leaders.

Highest-priority diagnostic: Which surfaces provide the most reliable New American Funding presence, and can that presence grow without a corresponding movement in coverage?

Wells Fargo and Wells Fargo & Co.

Wells Fargo recorded coverage of 20.8% in July 2026 and 24.7% in August 2026. In September and October 2026, Wells Fargo as tracked recorded 0.0% coverage, while Wells Fargo & Co. held 7.2% coverage in September and 6.4% in October, a decline of 0.8 points from the prior month that was within normal variation. The baseline-to-current gain of 6.4 points against a zero baseline is significant, but reflects first appearance rather than an established trend.

Wells Fargo & Co. achieved its 6.4% coverage from 34 valid recommendations among 529 qualified observations. Presence measured 17.6%, with 93 observations. Its top-three rate was 2.8%, and its rank-one rate was 0.8%, with 4 first-place finishes. Net sentiment measured 0.37, the lowest among tracked brands this month.

The distinction to notice: the Wells Fargo & Co. entity is present in responses but almost never recommended in the top three. This reads as a brand that is discussed but not surfaced as a leading option under its full corporate name.

Highest-priority diagnostic: Is the underlying lender losing recommendation strength, or is it being tracked under a name that appears less often in AI responses?

Rate

Rate entered the tracked set in September 2026 and recorded 4.2% coverage in October 2026, down 1.4 points from the prior month within normal variation. Against the July baseline of zero, the gain of 4.2 points is significant by a small threshold. Valid recommendations numbered 22 in October from a qualified base of 529.

Presence measured 5.7%, with 30 observations. The top-three rate was 1.5%, and the rank-one rate was 0.8%, with 4 first-place finishes. Net sentiment measured 0.73.

Rate operates on very small counts: its 4.2% coverage represents 22 valid recommendations, and its 4 first-place finishes are a small absolute base, among the smallest footprints in the tracked set alongside Chase Credit Journey.

Highest-priority diagnostic: Which prompts surfaced Rate, and does the brand's presence persist across months or fluctuate within normal variation?

Chase and Chase Credit Journey

Chase recorded valid recommendation coverage of 54.5% in July 2026 and 50.4% in August 2026. In September and October 2026, Chase as tracked recorded 0.0% coverage, while Chase Credit Journey measured 0.9% coverage in September and 0.6% in October, a decline of 0.3 points from the prior month within normal variation. Chase Credit Journey's coverage is far below the level Chase held through August.

Chase Credit Journey's 0.6% coverage represents only 3 valid recommendations and 4 mentions across 529 observations. Its top-three rate was 0.6%, with 3 top-three placements, and its rank-one rate was 0.0%.

Chase's disappearance from the tracked set under its original name is far larger in magnitude than the emergence of Chase Credit Journey. If AI responses are referring to the lender by other names not yet captured in the tracked set, the benchmark would understate the lender's true presence.

Highest-priority diagnostic: Which entity names are AI systems using when discussing Chase mortgage refinance offerings, and do those names map back to the same underlying brand?

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters did the October 2026 benchmark actually capture?
• Which commercial questions about refinance pricing and head-to-head lender comparison remain outside the current measurement?

All 529 qualified observations in October 2026 fell into the Brand Recommendation cluster, as did all observations in July, August, and September 2026. No observations were captured in the Pricing & Value or Multi-Brand Comparison clusters, which means the public benchmark cannot yet answer price, value, or head-to-head comparison questions. The commercial questions these clusters would answer, such as which lender offers the best refinance rate or how lenders stack up directly, remain outside the current measurement.

Brand Opportunity Summary

Questions This Section Answers

• Which mortgage refinance lenders show the strongest or weakest current AI recommendation signals?
• What is the highest-priority diagnostic for each tracked brand based on its October coverage and placement?

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 prompt-level data is collected to build the mortgage refinance recommendation metrics?
• Does source presence in AI answers establish causation for a lender's recommendation position?

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.

About This Benchmark

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

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

Report-Specific Interpretation Notes

• Small-count movement: Rate (22 valid recommendations in October), Chase Credit Journey (3), and Wells Fargo & Co. (34) operate on small absolute bases. Percentage movements for these entities carry less stability than the leaders.
• Entity-name change: Chase, Bank of America, Wells Fargo, and Guaranteed Rate recorded zero coverage under their July and August tracked names in September and October, while Bank of America Corp., Wells Fargo & Co., and Chase Credit Journey entered the tracked set. The benchmark cannot distinguish whether this reflects a change in AI-generated naming conventions or a change in which entity names the benchmark tracked; the coverage swings described above should be read with that caveat.
• Qualified denominator vs raw collection: all percentages use the qualified set (529 observations in October 2026, 607 in July 2026), not the raw 800 collected prompts.
• Directional analysis: month-over-month movement identifies changes worth investigating; it does not establish cause.

Next Step

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

The aggregate percentage hides the questions that matter: which high-intent prompts are won, which competitor takes the recommendation when a brand loses ground, what attributes AI associates with each option, and which external sources shape those answers. For the brands that fell to zero coverage under their tracked names, the benchmark shows the magnitude of the change but not whether the underlying lender lost recommendation strength or merely shifted to a different entity name.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy.

Request an AI visibility audit