Bank of America Corp. AI Market Strategy Report - Mortgage Refinance Lenders

Answer Capsule

Bank of America Corp. entered the September 2026 Mortgage Refinance Lenders benchmark with 27.9% valid recommendation coverage, a substantial increase from the 0.0% recorded under its prior tracked name in July 2026. The brand holds the third-strongest recommendation position in the category, behind Rocket Mortgage and loanDepot, with a presence rate of 44.9%. Its clearest weakness is a rank-one rate of just 0.9%, meaning the lender is frequently shortlisted but almost never named as the first or primary recommendation. The clearest opportunity is converting its strong mid-list recommendation presence into higher placement through targeted prompt and citation work.

Who This Report Is For

This report is for mortgage and home lending executives, digital strategy leaders, and brand teams at Bank of America Corp. tracking how AI systems recommend lenders during mortgage refinance discovery.

Report Card

Executive Summary

Questions This Section Answers

• What valid recommendation coverage did Bank of America Corp. record in September 2026, and where does it rank among tracked lenders?
• How does the September 2026 entity-name reconfiguration affect comparison with Bank of America Corp.'s prior monthly results?
• What does the conversion of mentions into valid recommendations reveal about Bank of America Corp.'s AI presence?

Bank of America Corp. recorded valid recommendation coverage of 27.9% in September 2026, placing it third among the ten tracked mortgage refinance lenders. This marks a significant entry into the benchmark under its full corporate name, following the July 2026 baseline in which the similarly named Bank of America entity recorded no qualified coverage. The benchmark shows the lender with 159 valid recommendations from 570 qualified observations, alongside a presence rate of 44.9% and 256 total mentions.

The strongest signal for Bank of America Corp. is its recommendation coverage relative to presence. The brand converts a meaningful share of its mentions into valid recommendations, with 159 of 256 appearances producing shortlist inclusion. Its top-three rate of 11.2% places it fourth in the category on that measure, while its rank-one rate of 0.9% is among the weakest of the leading brands, with only five first-place finishes recorded.

The clearest platform strength appears in Google AI Mode, where Bank of America Corp. achieved 34.0% valid recommendation coverage, its strongest platform-level performance. The clearest gap is rank-one placement across all platforms, where the brand rarely appears as the first or primary recommendation despite consistent shortlist presence.

The benchmark's September 2026 entity-name reconfiguration complicates direct comparison with prior months. Bank of America, as tracked in July and August 2026, recorded coverage of 55.4% and 51.2% respectively, then fell to 0.0% in September under that name. Bank of America Corp. entered at 27.9% in the same month. The combined reading suggests the underlying lender retains meaningful AI presence, but the benchmark captured it under a different tracked name in September.

What Bank of America Corp. Is Winning

Questions This Section Answers

• Which platforms and metrics show the strongest evidence of Bank of America Corp.'s AI recommendation visibility?
• How does Bank of America Corp.'s presence-to-recommendation conversion compare with competitors?
• What does Bank of America Corp.'s net sentiment score indicate about how AI systems frame the lender?

Bank of America Corp. holds the third-strongest valid recommendation coverage in the category at 27.9%, trailing only Rocket Mortgage at 63.7% and loanDepot at 36.5%. This positions the lender ahead of Better.com, PennyMac, and the remaining tracked brands.

The brand shows a strong presence-to-recommendation conversion pattern. With 256 mentions and 159 valid recommendations, Bank of America Corp. converts roughly six of every ten mentions into shortlist inclusion, a stronger conversion profile than several competitors with comparable presence.

Google AI Mode is the clearest platform win. Bank of America Corp. achieved 34.0% valid recommendation coverage on that surface, its highest platform-level result, with a top-three rate of 17.7% and 48 valid recommendations from 141 observations.

Net sentiment of 0.66 is positive and consistent with the category's leading brands, indicating the lender is framed favorably when it appears in AI responses.

Where Bank of America Corp. Has the Clearest AI Visibility Gaps

Questions This Section Answers

• Which AI visibility measure shows Bank of America Corp.'s most significant weakness relative to leading competitors?
• How does the entity-name measurement gap affect interpretation of Bank of America Corp.'s apparent coverage change?
• What does the comparison with Wells Fargo & Co. and leading lenders reveal about Bank of America Corp.'s placement profile?

The most significant gap is rank-one placement. Bank of America Corp. recorded a rank-one rate of just 0.9%, with only five first-place finishes across 570 observations. Rocket Mortgage, by comparison, holds a 30.5% rank-one rate. The lender is being shortlisted consistently but is almost never named as the first or primary recommendation, suggesting AI systems treat it as a credible option rather than a default choice.

The entity-name question creates a measurement gap that matters for strategy. Bank of America, as tracked in July 2026, held 55.4% valid recommendation coverage. Bank of America Corp. entered at 27.9% in September 2026. The benchmark cannot determine whether this reflects a change in how AI systems name the lender or a change in which entity names the benchmark tracked. The apparent decline of 55.4 points for Bank of America and the rise of 27.9 points for Bank of America Corp. line up with a naming reconfiguration rather than a clearly established loss of recommendation strength.

Wells Fargo & Co. provides a useful contrast. Both bank brands entered the September tracked set under full corporate names, but Bank of America Corp. achieved 27.9% coverage while Wells Fargo & Co. recorded just 7.2%. Within the leading tier, Bank of America Corp.'s top-three rate of 11.2% trails Rocket Mortgage at 44.2% and loanDepot at 13.3%, indicating the brand is recommended but less often placed in the most visible positions.

Biggest Opportunity

The clearest opportunity for Bank of America Corp. is converting its strong mid-list recommendation presence into top-three placement. The brand already achieves meaningful shortlist inclusion at 27.9% coverage, but its top-three rate of 11.2% and rank-one rate of 0.9% show that most recommendations place the lender below the most visible positions. The path forward is identifying which high-intent prompts produce mid-list placement and strengthening the owned answer layer and citation architecture that AI systems draw on when ranking lenders as first or primary options.

Competitive Landscape

Questions This Section Answers

• Where does Bank of America Corp. rank against tracked competitors on top-three rate and rank-one rate?
• Which competitor closely matches Bank of America Corp.'s top-three rate with a stronger average recommended rank?

Rocket Mortgage holds dominant recommendation-stage strength in the category, with loanDepot and Bank of America Corp. forming the next tier. Bank of America Corp. sits third by valid recommendation coverage but trails the top two brands substantially on top-three and rank-one placement.

Average recommended rank covers rank-eligible recommendations only.

Bank of America Corp. holds the third position in the tracked set by valid recommendation coverage, but its top-three and rank-one rates are closely matched by Better.com, which achieves a similar top-three rate with a stronger average recommended rank of 2.93. The numbers show Bank of America Corp. is recommended at scale but not yet placed with the prominence its coverage level might suggest.

Prompt Evidence

Google AI Mode / Brand Recommendation Prompt: "Which bank is best for refinancing?" Result: Bank of America Corp. appeared in a valid recommendation shortlist with a top-three placement, contributing to its strongest platform-level coverage.

ChatGPT / Brand Recommendation Prompt: "Is it better to go through a bank or mortgage lender?" Result: Bank of America Corp. was mentioned and recommended but did not achieve a rank-one placement, reflecting the brand's broader pattern of mid-list positioning.

Perplexity / Brand Recommendation Prompt: "Who are the big 5 mortgage lenders?" Result: Bank of America Corp. appeared as a valid recommendation with a rank-one finish, one of only five first-place results recorded across all platforms.

What CiteWorks Studio Would Do Next

Questions This Section Answers

• What first phase should Bank of America Corp. undertake to map its AI recommendation gaps?
• Which later phases address the owned content and citation layers that affect rank-one placement?

Phase 1: AI Market Discovery Audit Map which high-intent mortgage refinance prompts produce Bank of America Corp. mentions versus valid recommendations, and identify the prompt families where the brand is shortlisted but not placed in the top three.

Phase 2: Recommendation Readiness Plan Build a prioritized plan to strengthen the brand's rank-one and top-three positioning, focusing on the comparison and selection prompts where the lender currently appears as a mid-list option.

Phase 3: Owned Answer Layer Buildout Develop owned content that directly answers refinance comparison, bank versus lender, and lender selection questions, giving AI systems clearer material to cite when ranking Bank of America Corp. as a primary option.

Phase 4: Citation / Authority Layer Development Strengthen the public evidence layer that AI systems retrieve when forming refinance recommendations, ensuring third-party sources describe Bank of America Corp. in terms that support first-position placement.

Phase 5: Monthly AI Visibility and Recommendation Tracking Track monthly changes in valid recommendation coverage, top-three rate, and rank-one rate, with particular attention to whether AI systems increasingly reference the lender under its full corporate name.

Why This Matters

AI-generated recommendations are becoming a primary input into mortgage refinance decisions. Bank of America Corp. has established a meaningful presence in this channel, appearing in nearly half of all qualified observations and earning shortlist inclusion in more than a quarter of them. But presence alone is not enough. The lender is recommended far less often than it is mentioned, and when it is recommended, it rarely appears in the most influential positions.

The next move for Bank of America Corp. is targeted correction of the prompt, page, and citation layers that determine whether AI systems name the lender first, second, or third in a refinance shortlist. The benchmark evidence shows the brand has the visibility foundation. Converting that foundation into top-three and rank-one placement is the strategic opportunity.

Core Metrics

Sentiment Score

Sentiment Score = (positive mentions × 1 + neutral mentions × 0 + negative mentions × -1) / total mentions

For Bank of America Corp., the calculation is (168 × 1 + 88 × 0 + 0 × -1) / 256, producing a net sentiment score of 0.6562.

This score matters because unclassified mention counts are misleading. Share of voice is a diagnostic metric, not a business KPI. A positive recommendation, neutral reference, cautionary mention, and competitor-displaced mention are not equal, and counting all mentions as wins is bad measurement. Classified sentiment is required before interpreting AI visibility, because it distinguishes whether a brand is being recommended favorably, mentioned in passing, or surfaced with negative framing.

Sentiment by Platform

Methodology

1. Report orientation: This is a benchmark-based analysis of AI-generated recommendations for mortgage refinance lenders, produced from the LLM Authority Index AI Market Discovery Index and supporting CiteWorks Studio interpretation. It is not a client implementation case study.
2. Reporting window: Data reflects the September 2026 measurement cycle, with comparison to the July 2026 baseline and August 2026 intermediate run where relevant.
3. Platforms tracked: ChatGPT, Copilot, Gemini, Perplexity, Google AI Overviews, and Google AI Mode, representing six canonical AI surface families.
4. Observation count: 570 qualified benchmark observations in September 2026, derived from 800 source prompt-surface observations.
5. Competitor universe: Ten tracked brands, including Bank of America Corp., Better.com, Chase Credit Journey, loanDepot, New American Funding, PennyMac, Rate, Rocket Mortgage, United Wholesale Mortgage, and Wells Fargo & Co.
6. Public clusters used: All qualified observations fell into the Brand Recommendation buyer-intent class. No qualified observations were captured for pricing, value, or multi-brand comparison prompts.
7. Stage 0 role: Raw prompt-surface observations were collected and qualified before brand-level metrics were calculated. The qualified set of 570 observations serves as the public denominator.
8. Definition of a mention: A brand mention is recorded when the brand appears in an AI response to a qualified observation, regardless of whether it is recommended.
9. Definition of a valid recommendation: A valid recommendation is recorded when the brand appears in a recommendation shortlist within an AI response. Mentions that are neutral, negative, cautionary, or listed only without recommendation credit are not counted as valid recommendations.
10. Entity-name note: The September 2026 benchmark reconfigured several tracked brand names. Bank of America, as tracked in July and August 2026, recorded no qualified coverage in September, while Bank of America Corp. entered the tracked set at 27.9% coverage. The benchmark cannot distinguish whether this reflects a change in AI naming conventions or a change in tracked entity names.
11. Limitations: The benchmark measures the sampled prompt set, not every possible AI response. It does not measure market share, sales attribution, organic-search ranking performance, social media volume, or private AI channels. Metric movements identify changes worth investigating but do not establish cause.
12. Citation and source layer: Prompt-level observations retain query, surface, answer, brand outcome, recommendation placement, sentiment, and citations where exposed. Source presence is evidence about the information environment and is not automatically proof that a source caused a recommendation.

Get Your AI Visibility Audit

The public benchmark shows where Bank of America Corp. stands in AI-generated mortgage refinance recommendations. A company-level audit goes deeper, mapping the specific prompts, competitor displacement patterns, platform differences, and evidence sources that determine whether the lender is named first, third, or not at all in AI responses.