Fish & Richardson AI Market Strategy Report - Patent Attorneys

Answer Capsule

Fish & Richardson remains the most visible and most recommended patent attorney firm in AI-generated recommendations for patent attorneys, holding 43.42% valid recommendation coverage in September 2026. That leadership is real but narrowing: coverage fell 15.1 points from 58.5% in July 2026, and rank-one placement fell 19.9 points to 18.42% over the same period. The firm still holds a 21.0-point lead over Finnegan, the next closest brand, and leads the category on both top-three rate (36.84%) and rank-one rate (18.42%). The clearest opportunity is converting its unmatched raw presence, 67.11% of qualified observations, back into first-position recommendations.

Who This Report Is For

This report is written for law firm leadership, business development teams, and marketing decision-makers at patent attorney firms who need to understand how AI systems are recommending firms in their category and where competitive position is shifting.

Report Card

Executive Summary

Fish & Richardson holds dominant recommendation power in the patent attorney category, but the September 2026 benchmark shows that power contracting. The firm recorded 33 valid recommendations across 76 qualified observations, a valid recommendation coverage rate of 43.42%. That figure leads all tracked brands by 21.0 percentage points over Finnegan at 22.4%, but it represents a 15.1-point decline from 58.5% in July 2026.

The firm's raw mention presence rate of 67.11% remains the strongest in the category. Fish & Richardson appeared in 51 of 76 qualified observations in September 2026, compared with 80 of 94 in July 2026. The gap between presence and recommendation conversion is the central story: the firm is still surfaced more often than any competitor, but a smaller share of those appearances convert into valid recommendations.

Rank-one placement showed the steepest erosion. Fish & Richardson's rank-one rate fell 19.9 points from 38.3% in July 2026 to 18.42% in September 2026, the largest decline in first-choice positioning among all tracked brands. The firm's rank-one count dropped from 36 to 14 over the same period. Top-three rate fell 15.3 points to 36.84%.

Sentiment remains strongly positive. The firm recorded 45 positive mentions, 6 neutral mentions, and zero negative mentions in September 2026, producing a net sentiment score of 0.8824. No tracked competitor registered a negative mention in the qualified set, so sentiment is not a differentiator in this category.

Google AI Overviews is the strongest platform signal for Fish & Richardson. The firm holds 53.1% valid recommendation coverage on that surface, with 17 valid recommendations from 28 appearances. Google AI Mode follows at 66.7% coverage, though from a smaller observation base of 12. Perplexity shows the highest rank-one rate at 40.0%, but the sample is limited to 5 observations.

The clearest platform gap is Copilot. Fish & Richardson appeared in 4 of 6 Copilot observations but converted only 1 into a valid recommendation, producing 16.7% coverage. Gemini shows a similar pattern: 3 appearances, 2 valid recommendations, 20.0% coverage. These surfaces represent recommendation-stage visibility that is not yet converting at the rate the firm achieves on Google surfaces.

What Fish & Richardson Is Winning

Questions This Section Answers

• Which metrics and platforms is Fish & Richardson leading in patent attorney AI recommendations?
• How does Fish & Richardson's Google AI Overviews coverage compare with its overall category rate?
• What does Fish & Richardson's average recommended rank and sentiment profile show?

Fish & Richardson holds the strongest overall position in the patent attorney category across every primary recommendation metric. The firm leads on valid recommendation coverage (43.42%), top-three rate (36.84%), rank-one rate (18.42%), and raw mention presence (67.11%). No competitor approaches these figures.

The firm's performance on Google AI Overviews is the standout platform result. Fish & Richardson achieved 53.1% valid recommendation coverage on AI Overviews, with 17 valid recommendations from 28 appearances. That coverage rate exceeds its overall category rate by nearly 10 percentage points, indicating that Google's AI-generated overview layer surfaces and recommends the firm at a higher rate than the category average.

Fish & Richardson also holds the strongest average recommended rank in the category at 1.7742. When the firm receives rank-eligible recommendations, it appears closer to first position on average than any competitor. Finnegan follows at 2.1333, and Wolf Greenfield at 2.2308.

The firm recorded zero negative mentions across all qualified observations. Its net sentiment score of 0.8824 reflects 45 positive mentions and 6 neutral mentions. While sentiment is uniformly positive across the category, the absence of any cautionary or negative framing is a baseline strength.

Where Fish & Richardson Has the Clearest AI Visibility Gaps

Questions This Section Answers

• Why is recommendation conversion, not raw presence, the main gap for Fish & Richardson?
• Which platforms show the clearest conversion shortfall for the firm?
• How did rank-one erosion and category contraction affect the firm's position?

The primary gap is recommendation conversion. Fish & Richardson appeared in 51 of 76 qualified observations but converted only 33 into valid recommendations. That 64.7% conversion rate trails Wolf Greenfield, which converted 13 of 19 appearances (68.4%), and Sterne Kessler, which converted all 5 of its appearances into valid recommendations. The firm's presence is not the constraint; the conversion of that presence into shortlist placement is.

Rank-one erosion is the sharpest expression of this gap. The firm's rank-one rate fell from 38.3% in July 2026 to 18.42% in September 2026. Finnegan and Knobbe Martens both held rank-one rates of 3.95% in September, while Wolf Greenfield converted 7.89% of qualified observations into first-position recommendations. Fish & Richardson still leads on rank-one rate, but the margin has compressed.

Copilot represents the clearest platform-specific gap. The firm appeared in 4 of 6 Copilot observations but converted only 1 into a valid recommendation. Finnegan achieved the same 16.7% coverage rate on Copilot from 2 appearances, and Kilpatrick Townsend matched it from 2 appearances. Fish & Richardson's presence on Copilot is not translating into recommendation placement at the rate it achieves elsewhere.

Gemini shows a similar pattern. The firm appeared in 3 of 10 Gemini observations with 2 valid recommendations, producing 20.0% coverage. Finnegan also achieved 20.0% coverage on Gemini, and Kilpatrick Townsend recorded 10.0%. The firm's Gemini presence is limited and its conversion rate does not exceed the category baseline.

The category itself contracted in September 2026. Valid recommendation shortlist share fell to 50.0% from 68.1% in July 2026, meaning fewer qualified observations produced valid recommendation shortlists. Fish & Richardson's decline occurred within a broader category pullback, but the firm's rank-one losses exceeded the category trend.

Biggest Opportunity

Questions This Section Answers

• Why are Google AI Overviews and Google AI Mode the clearest path back to rank-one conversion?
• What actually needs to change for Fish & Richardson to restore first-position recommendations?

The clearest path from reference to recommendation for Fish & Richardson is rebuilding rank-one conversion on Google AI Overviews and Google AI Mode. These two surfaces account for the largest share of qualified observations and represent the firm's strongest platform positions. Google AI Overviews alone produced 17 valid recommendations for the firm, more than any other surface. Google AI Mode produced 8 valid recommendations from 12 observations, a 66.7% coverage rate.

The opportunity is not presence. Fish & Richardson already appears in 87.5% of Google AI Overviews observations and 75.0% of Google AI Mode observations. The opportunity is converting those appearances back into first-position recommendations at the rate the firm achieved in July 2026, when its rank-one rate was 38.3%. That requires understanding which prompt types and evidence sources drove the rank-one decline and correcting the citation and framing layers that AI systems use to determine recommendation order.

Competitive Landscape

Questions This Section Answers

• How wide is Fish & Richardson's lead over its closest competitors on top-three and rank-one rates?
• Which competitors are closest on average recommended rank and sentiment?
• Which tracked firms recorded no presence in the September 2026 benchmark?

Fish & Richardson holds recommendation-stage strength in the patent attorney category, but the gap to the next tier has compressed. The firm leads on every primary metric, yet Finnegan, Knobbe Martens, and Wolf Greenfield all registered meaningful presence in September 2026.

Average recommended rank covers rank-eligible recommendations only.

Fish & Richardson's top-three rate of 36.84% is nearly double Finnegan's 19.74% and more than double Knobbe Martens at 15.79%. Its rank-one rate of 18.42% is more than four times Wolf Greenfield's 7.89%, the next highest. The firm's average recommended rank of 1.7742 places it closer to first position on average than any competitor. Three tracked brands recorded zero presence across all qualified observations.

Prompt Evidence

Google AI Overviews / Brand Recommendation Prompt: "What is the best IP law firm?" Result: Fish & Richardson received a valid recommendation with rank-one placement, consistent with its strongest platform performance.

Google AI Mode / Brand Recommendation Prompt: "Who is the best lawyer in Atlanta, Georgia?" Result: Fish & Richardson appeared in the response but did not convert to a rank-one recommendation, reflecting the firm's rank-one erosion on this surface.

Perplexity / Brand Recommendation Prompt: "best patent law firms" Result: Fish & Richardson received a valid recommendation with rank-one placement, contributing to its 40.0% rank-one rate on Perplexity.

ChatGPT / Brand Recommendation Prompt: "intellectual property law firm" Result: Fish & Richardson appeared in the response with a valid recommendation, though its overall ChatGPT coverage remained at 27.3%.

What CiteWorks Studio Would Do Next

Phase 1: AI Market Discovery Audit Map every prompt, surface, and competitor substitution pattern driving the rank-one decline, with particular focus on Google AI Overviews and Google AI Mode where the firm holds its strongest positions.

Phase 2: Recommendation Readiness Plan Identify which prompt categories and evidence sources correlate with rank-one placement and build a prioritized correction plan for the surfaces where conversion has weakened.

Phase 3: Owned Answer Layer Buildout Strengthen the firm's owned content so AI systems retrieve clearer, more recommendation-ready answers about practice strengths, attorney credentials, and matter experience.

Phase 4: Citation / Authority Layer Development Develop the public evidence layer, including third-party profiles, directory listings, and source pages, that AI systems appear to synthesize when determining recommendation order.

Phase 5: Monthly AI Visibility and Recommendation Tracking Track valid recommendation coverage, top-three rate, rank-one rate, and sentiment by platform and prompt cluster to measure whether corrections restore first-position conversion.

Why This Matters

Questions This Section Answers

• Why isn't Fish & Richardson's raw presence enough if it's not converting to recommendations?
• What does the rank-one decline mean for buyers asking AI for a recommended patent attorney firm?

AI presence alone is not enough. Fish & Richardson appears in 67.11% of qualified observations, more than any competitor, but converts only 64.7% of those appearances into valid recommendations. The firm's rank-one rate fell 19.9 points in two months. Buyers who ask AI systems for a recommended patent attorney firm are seeing Fish & Richardson more often than any other firm, but they are seeing it recommended first less often than they did in July 2026.

The next move is targeted correction of the prompt, page, and citation layers that AI systems use to determine recommendation order. The benchmark shows where the decline occurred. A company-level audit shows why, which prompts and surfaces drove the rank-one losses, and which evidence sources need strengthening to restore first-position conversion.

Core Metrics

Sentiment Score

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

Fish & Richardson's sentiment score for September 2026 is 0.8824, calculated from 45 positive mentions, 6 neutral mentions, and zero negative mentions across 51 total mentions.

This score matters because unclassified mention counts are misleading. A firm that appears in 51 AI responses but is framed neutrally or cautionarily in half of them is not in the same position as a firm that appears in 51 responses with consistently positive framing. 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. Classified sentiment is required before interpreting AI visibility. Fish & Richardson's 0.8824 score indicates that when the firm appears in AI-generated responses, the framing is overwhelmingly positive. The 6 neutral mentions represent factual references without recommendation weight. The absence of negative mentions means no AI system in the qualified set framed the firm unfavorably.

Sentiment by Platform

Methodology

1. This report is a benchmark-based analysis of AI-generated recommendations for patent attorney firms, produced by CiteWorks Studio using data from the LLM Authority Index AI Market Discovery Index for September 2026.
2. The reporting window covers September 2026, with trend comparisons to July 2026 and August 2026 where the benchmark provides historical data.
3. Six AI and search surface families were tracked: ChatGPT, Copilot, Gemini, Perplexity, Google AI Overviews, and Google AI Mode. All six families had at least one qualified observation in September 2026.
4. The September 2026 benchmark analyzed 575 prompt-surface observations, producing 394 unique questions. After qualification, 76 observations formed the public denominator for brand-level metrics.
5. Ten patent attorney firms were tracked: Fish & Richardson, Finnegan, Knobbe Martens, Wolf Greenfield, Kilpatrick Townsend, Sterne Kessler, Harrity & Harrity, Cantor Colburn, Banner Witcoff, and Schwegman Lundberg.
6. All 76 qualified observations fell into the Brand Recommendation cluster. No qualified observations were recorded in the Pricing & Value or Multi-Brand Comparison clusters, meaning the current benchmark cannot answer how AI systems frame pricing or head-to-head comparisons.
7. Stage 0 extraction retained the query, AI/search surface, answer, brand outcome, recommendation placement, sentiment, and, where exposed, citations or attributable evidence sources for each observation.
8. A mention is defined as any appearance of a tracked brand in an AI-generated response, regardless of whether the brand was recommended. Mentions include positive, neutral, and negative references.
9. A valid recommendation is defined as an observation where the brand received a positive recommendation with rank placement between 1 and 10. Neutral references, cautionary mentions, and listed-only appearances do not count as valid recommendations.
10. Brand-level percentages use the 76 qualified observations as the public denominator, not the raw collection of 575 prompt-surface observations. Coverage percentages are calculated within each month's qualified set, so direct comparisons reflect rates, not raw counts.
11. The benchmark does not measure market share, attributable sales, every possible AI response, organic-search ranking, social mention volume, private or sponsored channels, or causality from a metric movement alone. The two-month declines for leading brands should not yet be treated as an established trend.
12. Source presence in the evidence layer is not automatically treated as proof that the source caused a recommendation. Citation frequency is not endorsement.

See How AI Is Recommending Your Brand

The public benchmark shows where Fish & Richardson is winning and where its recommendation position has weakened. A company-level AI visibility audit maps the prompt, surface, competitor, ranking, sentiment, and evidence-source patterns behind those movements into a prioritized strategy. Where the benchmark shows a rank-one decline, the audit identifies whether the loss is concentrated in specific prompts, surfaces, or competitor substitutions.