Earnest AI Market Strategy Report - Student Loan Refinance

Answer Capsule

Earnest holds the second-strongest recommendation position in the September 2026 Student Loan Refinance benchmark, with valid recommendation coverage of 56.26% across 455 qualified observations. The brand appears in 92.31% of AI-generated answers, but its top-three placement rate fell 6.7 points month over month to 40.88%, and its rank-one rate dropped to 12.97%. Earnest is visible in nearly every answer but is being recommended lower in the list than its presence suggests. The clearest opportunity is converting that near-universal presence into stronger first-position recommendation share, particularly on the platforms where SoFi is capturing rank-one placements at nearly double Earnest's rate.

Who This Report Is For

This report is for Earnest's marketing, growth, and brand strategy teams, and for any stakeholder responsible for how the lender appears in AI-generated recommendations during the student loan refinance buyer journey.

Report Card

Executive Summary

Earnest is the second most recommended student loan refinance brand in the September 2026 benchmark, with valid recommendation coverage of 56.26% across 455 qualified observations. That figure is down 3.7 points from 60.0% in August 2026, and the gap to category leader SoFi widened from 1.1 points to 2.8 points over the same period.

The brand's raw mention presence rate is 92.31%, meaning Earnest appears in nearly every AI-generated answer about student loan refinance. This presence rate trails only SoFi's 94.7%. The decline in recommendation coverage was not driven by disappearance from answers. It was driven by placement: Earnest's top-three rate fell 6.7 points from 47.6% to 40.88%, and its rank-one rate dropped from 17.2% to 12.97%. The brand is being mentioned but recommended lower in the list.

Earnest's average recommended rank slipped from 2.17 to 2.29, reinforcing that the brand is appearing further down in recommendation sets. Sentiment held steady at 0.6857, so the change is about placement, not perception. The brand recorded 288 positive mentions, 132 neutral mentions, and zero negative mentions across the qualified observation set.

The strongest platform signal for Earnest is Google AI Mode, where the brand holds a 69.6% valid recommendation coverage rate and a 47.3% top-three rate. ChatGPT is also a strong surface, with a 67.4% coverage rate and a 47.8% top-three rate. The weakest platform is Perplexity, where Earnest's coverage drops to 33.3% and its rank-one rate falls to zero.

The clearest gap is in rank-one placement. SoFi's rank-one rate of 25.71% is nearly double Earnest's 12.97%, despite similar coverage levels. This means that when both brands are recommended, SoFi is more often placed first. The benchmark data shows that similar coverage levels can still hide very different first-position rates.

The category itself saw a slight contraction in recommendation-shaped answers. The share of answers that took a recommendation form declined from 46.5% to 44.6%, and the valid recommendation shortlist share fell from 62.4% to 58.9%. September's prompt mix produced more factual or pricing-oriented responses and fewer direct recommendation lists.

What Earnest Is Winning

Questions This Section Answers

• Where does Earnest hold the strongest recommendation position in the category?
• Which platforms come closest to converting Earnest's raw AI presence into actual recommendations?

Earnest holds the second-strongest recommendation position in the category, with a 56.26% valid recommendation coverage rate that places it well ahead of third-place ELFI at 20.00% top-three rate and 32.5% recommendation coverage. The gap between Earnest and the rest of the field beyond SoFi is substantial.

The brand's raw mention presence rate of 92.31% is the second-highest in the benchmark, trailing only SoFi. Earnest appears in 420 of 455 qualified observations, which means AI systems consistently recognize and reference the brand when answering student loan refinance questions.

Earnest's strongest platform is Google AI Mode, where it holds a 69.6% valid recommendation coverage rate and a 47.3% top-three rate across 112 observations. On ChatGPT, the brand holds a 67.4% coverage rate and a 47.8% top-three rate across 46 observations. These are the two platforms where Earnest's recommendation power is closest to its raw presence.

The brand recorded zero negative mentions across the entire qualified observation set. Its net sentiment score of 0.6857 reflects a framing environment that is overwhelmingly positive or neutral, with no cautionary or negative characterizations detected.

Earnest also holds a meaningful rank-one position on Copilot, where its 17.0% rank-one rate slightly exceeds SoFi's 3.4% on the same platform. This is a narrow but notable pocket of first-position strength.

Where Earnest Has the Clearest AI Visibility Gaps

Questions This Section Answers

• What is the gap between how often Earnest is mentioned and how often it is actually recommended?
• Where is Earnest losing first-position recommendations to SoFi?

The primary gap is recommendation placement relative to presence. Earnest appears in 92.31% of qualified observations but receives a valid recommendation in only 56.26% of them. That is a 36-point gap between being mentioned and being recommended. SoFi's gap is narrower at 35.6 points, but SoFi converts more of its presence into top-three and rank-one placements.

The rank-one gap is the most commercially significant. Earnest's rank-one rate of 12.97% means the brand is the first recommendation in 59 of 455 observations. SoFi's rank-one rate of 25.71% means it is first in 117 observations. SoFi is nearly twice as likely to be the first brand named when AI systems recommend a student loan refinance lender.

On Perplexity, Earnest's rank-one rate falls to zero. The brand holds a 33.3% valid recommendation coverage rate on that platform but never appears as the first recommendation. SoFi, by contrast, holds a 27.1% rank-one rate on Perplexity. This is a platform-specific gap where Earnest is present but not competitive for first position.

Earnest's top-three rate declined 6.7 points from August to September 2026, a larger movement than its overall coverage change. This suggests the brand is being displaced from top-three positions by competitors even when it remains in the recommendation set. The benchmark classified Earnest's overall coverage movement as stable, but the top-three decline is a leading indicator worth monitoring.

The category's recommendation-shaped answer share declined from 46.5% to 44.6%, and the valid recommendation shortlist share fell from 62.4% to 58.9%. This means fewer prompts are producing direct recommendation lists. Earnest's presence remains high, but the opportunities to be recommended are becoming slightly less frequent across the category.

Biggest Opportunity

Questions This Section Answers

• Which platform offers the largest first-position gain if Earnest can close the rank-one gap?
• Where does Earnest already outperform SoFi on first-position placement?

Earnest's clearest path from reference to recommendation is closing the rank-one gap on Google AI Mode and ChatGPT, the two platforms where its coverage is strongest and where SoFi's first-position advantage is most pronounced.

On Google AI Mode, Earnest holds a 69.6% coverage rate but a 10.7% rank-one rate. SoFi holds a 73.2% coverage rate and a 32.1% rank-one rate on the same platform. The coverage difference is 3.6 points, but the rank-one difference is 21.4 points. This is the single largest first-position gap in the benchmark for Earnest.

On ChatGPT, Earnest holds a 67.4% coverage rate and a 17.4% rank-one rate. SoFi holds a 65.2% coverage rate and a 13.0% rank-one rate. Earnest actually outperforms SoFi on rank-one placement on ChatGPT, which suggests the brand's first-position strength is platform-dependent rather than uniformly weak.

The opportunity is to understand which prompt types on Google AI Mode are placing SoFi first and Earnest second or third, and what evidence sources are supporting those first-position recommendations. The benchmark data shows the gap but cannot identify the specific prompts or sources causing it. A company-level analysis can isolate those patterns.

Competitive Landscape

Questions This Section Answers

• How does Earnest's top-three and rank-one performance compare with the rest of the competitive set?
• What separates Earnest from SoFi in the rankings besides sentiment?

SoFi and Earnest hold the strongest recommendation-stage positions in the category, with SoFi leading on coverage, top-three rate, and rank-one rate. Earnest is the clear second, well ahead of ELFI and the rest of the field. The table below shows the full competitive set sorted by top-three rate.

Average recommended rank covers rank-eligible recommendations only.

Earnest's position in the table shows a brand with strong top-three presence but a rank-one rate that is less than half of SoFi's. The average recommended rank of 2.29 places Earnest firmly in second position on average, but the gap to SoFi's 1.88 is meaningful. The sentiment score of 0.6857 is essentially tied with SoFi's 0.7007, indicating that framing quality is not the differentiator between the top two brands.

Prompt Evidence

Google AI Mode / Best Student Loan Refinance Companies & Options Prompt: "What is the best lender for student loans?" Result: Earnest appeared in the recommendation set with a top-three placement, but SoFi captured the rank-one position.

Perplexity / Best Student Loan Refinance Companies & Options Prompt: "student loan refinance lenders" Result: Earnest received a valid recommendation but was not placed in the top three, and the brand recorded zero rank-one appearances on this platform.

ChatGPT / Best Student Loan Refinance Companies & Options Prompt: "What are the best options for student loans?" Result: Earnest appeared as a top-three recommendation with strong positive framing, one of the brand's strongest prompt-level outcomes.

Google AI Overviews / Best Student Loan Refinance Companies & Options Prompt: "best student loan refinance" Result: Earnest appeared in the recommendation set with a top-three placement, contributing to its 53.6% top-three rate on this platform.

What CiteWorks Studio Would Do Next

Phase 1: AI Market Discovery Audit Map the specific prompts on Google AI Mode and ChatGPT where Earnest is recommended but not placed first, and identify which competitors are capturing those rank-one positions.

Phase 2: Recommendation Readiness Plan Prioritize the prompt clusters and platforms where Earnest's coverage is strong but its rank-one rate is weak, focusing on the 21-point first-position gap on Google AI Mode.

Phase 3: Owned Answer Layer Buildout Strengthen the owned content and structured data that AI systems retrieve when forming first-position recommendations, particularly for the comparison and evaluation prompts where Earnest is present but not leading.

Phase 4: Citation / Authority Layer Development Develop the public evidence layer that supports first-position recommendations, including the source types and citation patterns that AI systems appear to favor when placing a lender at rank one.

Phase 5: Monthly AI Visibility and Recommendation Tracking Track Earnest's top-three and rank-one rates month over month across all six platforms, with particular attention to whether the 6.7-point top-three decline in September 2026 continues or reverses.

Why This Matters

Earnest appears in nearly every AI-generated answer about student loan refinance. That presence is not the problem. The problem is that presence is not converting into first-position recommendations at the same rate as SoFi. When a borrower asks an AI system for the best student loan refinance lender, Earnest is almost always mentioned, but SoFi is more often named first.

The difference between being mentioned and being recommended first is the difference between being on the shortlist and being the default choice. AI systems are increasingly where recommendations are formed, and the brand that occupies the first position captures the buyer's attention before the comparison even begins. Earnest's 12.97% rank-one rate means it is the first recommendation in roughly one in eight qualified observations. SoFi's 25.71% rate means it is first in roughly one in four.

The next move is targeted correction of the prompt, page, and citation layers that determine first-position placement. The benchmark shows where Earnest is losing ground. A company-level analysis shows why, and what specific levers can close the gap.

Core Metrics

Sentiment Score

Questions This Section Answers

• Why is mention count alone a misleading measure of Earnest's AI recommendation strength?
• What does Earnest's zero-negative-mention profile say about its framing versus its placement?

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

Earnest's sentiment score for September 2026 is 0.6857. This is calculated from 288 positive mentions, 132 neutral mentions, and zero negative mentions across 420 total mentions.

This matters because unclassified mention counts are misleading. A brand that appears in 420 answers but is framed negatively in half of them is not in the same position as a brand that appears in 420 answers with uniformly positive or neutral framing. Earnest's zero negative mentions mean its framing environment is clean, but the 132 neutral mentions represent references where the brand is named without a clear recommendation signal.

Share of voice is a diagnostic metric, not a business KPI. A positive recommendation, a neutral reference, a cautionary mention, and a competitor-displaced mention are not equal. Counting all mentions as wins is bad measurement. Earnest's 92.31% presence rate looks strong, but the 56.26% valid recommendation coverage tells a more precise story about how often the brand is actually being recommended rather than just referenced.

Classified sentiment is required before interpreting AI visibility. Earnest's sentiment score of 0.6857 is essentially tied with SoFi's 0.7007, which means framing quality is not the differentiator between the top two brands. The differentiator is placement: SoFi converts its presence into first-position recommendations at nearly double Earnest's rate.

Sentiment by Platform

Questions This Section Answers

• Which platforms show the strongest positive framing for Earnest?
• Where do neutral mentions dominate despite high sentiment scores?

Note: Platform-level mention counts reflect mentions in answers where a brand outcome was classified. They do not sum to the 420 total brand mentions because a single mention can span multiple platform observations, and several observations produced no classified mention.

Methodology

1. This report is a benchmark-based analysis of Earnest's AI recommendation visibility in the Student Loan Refinance category for September 2026. It is not a client implementation case study.
2. The reporting window is September 2026, with August 2026 baseline data included for month-over-month comparison.
3. Six AI/search surface families were tracked: ChatGPT, Copilot, Gemini, Perplexity, Google AI Overviews, and Google AI Mode. All six were represented in the qualified observation set.
4. The September 2026 benchmark produced 455 qualified observations from an initial collection of 800 prompt-surface observations. The August 2026 baseline produced 447 qualified observations from the same raw collection volume.
5. Ten brands were tracked in the competitive set: SoFi, Earnest, ELFI, Citizens, LendKey, RISLA, Splash Financial, Laurel Road, MEFA, and PNC Bank.
6. One public high-intent cluster was qualified for analysis: Best Student Loan Refinance Companies & Options. The Pricing & Value and Multi-Brand Comparison clusters contained zero qualified observations in the public benchmark.
7. Stage 0 extraction retained the query, AI/search surface, answer, brand outcome, recommendation placement, sentiment, and, where exposed, citations or attributable evidence sources.
8. A mention is counted when a brand appears in a qualified observation in any capacity, including neutral references and comparison anchors.
9. A valid recommendation is counted when a brand receives a genuine, non-cautionary recommendation. Negative, neutral, cautionary, comparison-anchor, and listed-only mentions are not counted as valid recommendations.
10. The qualified denominator of 455 observations is smaller than the raw collection universe of 800 prompts because only prompts that were relevant and delivered a usable AI answer enter the public metrics.
11. Several brands in this vertical operate with small observation counts, so percentage movements can appear large even when the absolute number of recommendations changed by only a handful.
12. Month-over-month movement identifies changes worth investigating. It does not by itself establish the cause of those changes. Directional analysis requires prompt-level review.

See How AI Is Recommending Your Brand

The public benchmark shows where Earnest is winning and losing in AI-generated recommendations. A company-level AI visibility audit maps the specific prompts, competitors, platforms, and evidence sources driving those outcomes, and identifies the levers that can improve first-position recommendation share.