How AI Search Is Recommending LLC Formation Services: Monthly Trends

Executive Summary

Northwest Registered Agent remains the clear leader in AI recommendation coverage for LLC formation services, closing October 2026 at 67.6% valid recommendation coverage. That is 9.3 points ahead of second-place ZenBusiness at 58.3%. The leader's baseline-to-current move of up 7.9 points from July 2026 is the only movement in the series that goes beyond the category's normal month-to-month variation, and it came with a rank-one rate that rose from 22.9% to 30.6% over the same span. The gap between Northwest Registered Agent and Bizee (Incfile) has now widened every month since July, from 8.0 points to 18.8 points.

The strongest upward mover among brands outside the top tier is Harbor Compliance, which reached 11.4% coverage in October 2026, up 2.8 points from 8.6% in July 2026. Its top-three placement rate rose from 3.2% to 8.9% across the series, a 5.7-point gain, the largest shift in top-three placement recorded outside the top tier. Firstbase posted its third consecutive monthly increase, reaching 0.9% coverage from 0.0% in July 2026, though the cumulative move remains within normal variation.

The sharpest decliner across the full series is Bizee (Incfile), down 2.9 points from 51.7% in July 2026 to 48.8% in October 2026. That aggregate decline masks a two-month recovery: the brand fell from 51.7% to 44.0% between July and August, then climbed back to 48.8% by October. Its raw mention presence rate, however, tells a different story, falling 7.3 points from 71.8% to 64.5% across the series, a decline that stands out against the rest of the category's movement this period. The brand is appearing in fewer AI answers even as its recommendation coverage recovers.

Against the prior month, September 2026, the category was stable. No brand moved beyond its normal month-to-month range in October. Northwest Registered Agent added 4.0 points from September's 63.6%, continuing a two-month uptrend. Bizee (Incfile) added 2.9 points from 45.9%, LegalZoom added 4.1 points from 46.8%, and Rocket Lawyer added 3.3 points from 11.2%. These are recoveries from a soft August, not structural breaks.

Each monthly run begins with 800 prompt-surface observations (520 unique questions in October 2026) across the benchmark's defined AI/search surface universe. Of those, 800 mentioned a tracked brand or competitor; 567 were relevant and 233 were irrelevant. The public metrics use the 324 qualified observations that survive both qualification stages. For context, July 2026 began with 800 prompts (455 unique questions) with 315 qualified observations, August 2026 began with 800 prompts (516 unique questions) with 384 qualified observations, and September 2026 began with 800 prompts (519 unique questions) with 357 qualified observations.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• What kinds of factual conflicts are appearing across AI platforms for LLC formation services?
• Which brands and pricing claims are most affected by these inconsistencies?

The benchmark's conflict detection identified 9 critical or high-severity factual inconsistencies across 6 companies and 5 AI platforms: ChatGPT, Copilot, Gemini, AI Mode, and AI Overviews. The vast majority of conflicts involve pricing figures, where AI systems returned incompatible cost claims for the same service. These inconsistencies matter commercially because a buyer asking a pricing question may receive materially different answers depending on which AI surface they use.

Rocket Lawyer

Rocket Lawyer accounted for 4 of the 9 detected inconsistencies, spanning all five platforms. The first conflict involves the availability of a standalone will purchase. When asked "How much does Rocket Lawyer charge for a will?", ChatGPT stated that Rocket Lawyer has historically charged a one-time fee for a standalone Last Will and Testament without a membership, citing the company's own homepage. AI Mode stated the opposite, that Rocket Lawyer doesn't typically sell a Last Will and Testament as a standalone purchase and that a membership subscription is required. The conflict was flagged against sources including ailawyer.pro and checkthat.ai, which provided contradictory excerpts on whether individual documents can be purchased without a subscription.

The second conflict involves membership pricing. When asked "Which is better, Rocket Lawyer vs LegalZoom?", ChatGPT stated memberships range from $149 per year to $349 per year, citing the company's pricing plans page. AI Mode stated the membership model is all-inclusive at typically around $39.99 per month, citing NerdWallet, Forbes, and a YouTube source. A flagged source from the company's own pricing page showed a Standard plan at $149 per year, supporting the ChatGPT claim.

The third conflict involves registered agent pricing in North Carolina. When asked "How much does a registered agent cost in NC?", Gemini stated up to $299 annually, while Copilot stated $374 with membership. The Copilot response cited a source showing a $125 agent fee plus a required $249 Plus membership, which totals $374. The Gemini response cited llcuniversity.com and ncregisteredagent.org.

The fourth conflict again involves membership pricing, this time between ChatGPT and AI Overviews. When asked "Which is better, Rocket Lawyer vs LegalZoom?", ChatGPT stated memberships range from $149 to $349 per year, while AI Overviews stated a monthly membership model at $39.99 to $64.99 per month, equivalent to roughly $480 to $780 per year. Both responses cited the company's pricing plans page, but reached incompatible conclusions about the pricing structure.

Harbor Compliance

When asked "What is the cheapest registered agent service?", ChatGPT stated Harbor Compliance charges $99 for the first year and then $149 per year, citing registeredagentguides.com and registeredagentcost.com. Gemini stated the annual fee is $89 to $99 per year, citing a LegalZoom article. The two claims cannot both be true for the same service.

LegalZoom

When asked "Is Northwest Registered Agent better than LegalZoom?", ChatGPT stated LegalZoom's registered agent service costs $249 per year. AI Overviews stated it costs around $199 for the first year. ChatGPT cited the company's business formation page and Northwest Registered Agent's incorporation service page. AI Overviews cited llcuniversity.com and topconsumerreviews.com, the latter of which explicitly stated the service is "pricier at $249 per year," contradicting the AI Overviews claim.

Northwest Registered Agent

When asked "What is the cheapest registered agent service?", ChatGPT stated Northwest Registered Agent costs $125 in the first year, citing registeredagentguides.com and llc.org. Copilot stated the service is free with a $39 formation, citing registeredagenthub.com and venturesmarter.com. The Copilot response's own cited source showed both figures: "$125/yr" and "Free with $39 formation," indicating the platform resolved the ambiguity differently than ChatGPT did.

Swyft Filings

When asked "What is the difference between Bizee and Swyft Filings?", ChatGPT stated Swyft Filings' registered agent service costs generally $149 per year. Gemini stated it costs roughly $596 per year if maintained through the company. The Gemini response cited a source explaining that Swyft Filings prices the service at $149 or more per quarter, not per year, which works out to roughly $596 annually. The ChatGPT response cited the company's own LLC page, which did not clarify the quarterly versus annual distinction.

ZenBusiness

When asked "How much is a registered agent in New Mexico?", AI Overviews stated ZenBusiness charges $99. AI Mode stated ZenBusiness can range from $199 to $300 or more per year. The AI Mode response cited speedyregisteredagents.com and tailorbrands.com. A flagged source from speedyregisteredagents.com showed both figures simultaneously: "$99 standalone (free with Premium)" and "$199/yr," suggesting the two platforms resolved the same source data differently.

Benchmark Context

Questions This Section Answers

• How does the October 2026 qualified observation set compare to earlier months?
• What changed in the recommendation-shaped answer share over the series?

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

August 2026 had 384 qualified observations, and September 2026 had 357. The October figure of 324 is the lowest qualified denominator in the series, which means every brand's coverage rate is calculated against a smaller base than in the prior two months. This denominator shift does not by itself explain directional movement, but it is a factor in interpreting month-over-month percentage changes.

Benchmark-Level Metrics

The recommendation-shaped answer share rose from 34.6% in July 2026 to 48.8% in October 2026, an increase of 14.2 points. The valid recommendation shortlist share rose from 63.8% to 73.1%, up 9.3 points. These category-level shifts indicate that a larger proportion of qualified observations are producing recommendation-style outputs over time, which raises the stakes for brands competing for placement inside those answers.

AI Recommendation Trend

Questions This Section Answers

• Which brands lead AI recommendation coverage for LLC formation services?
• Which brands moved beyond the category's normal month-to-month variation?

The leader pulled away while the top tier held its shape

Northwest Registered Agent entered October 2026 with a widening lead, reaching 67.6% coverage against ZenBusiness at 58.3%. The top tier of four brands above 48% coverage remains intact, though the gap between first and second place has grown from 1.3 points in July 2026 to 9.3 points in October 2026. The long tail below 15% coverage has not closed.

Only Northwest Registered Agent moved beyond the benchmark's normal month-to-month variation across the full July-to-October series. The category-level changes came from the combination of several smaller movements, most of which reversed the August dip rather than establishing new directional trends. Two brands, Firstbase and Harbor Compliance, logged three consecutive monthly increases, but both remain within their respective normal ranges of month-to-month movement.

What Changed This Month

Questions This Section Answers

• What changed for Northwest Registered Agent, Bizee, Harbor Compliance, and Rocket Lawyer this month?
• Where did presence and recommendation placement move in different directions?

Northwest Registered Agent: presence and rank-one placement both climbed

Coverage reached 67.6% in October 2026, up 4.0 points from September's 63.6% and up 7.9 points from July's 59.7%. That baseline-to-current gain of 7.9 points is the only movement in the category that stands out from normal month-to-month variation across the full series. The prior-to-current move of 4.0 points is within normal variation for a single month.

Raw mention presence rose from 82.5% in July 2026 to 91.0% in October 2026, an 8.5-point gain, among the larger shifts recorded in the series. The brand is now mentioned in more than nine out of ten qualified observations. Its top-three rate rose from 54.9% to 61.7% across the series, a 6.8-point gain. Rank-one rate rose from 22.9% to 30.6%, a 7.7-point gain, the largest rank-one movement of any brand in the category.

The distinction to notice is between presence and placement. Northwest Registered Agent's presence is already near saturation at 91.0%, which limits how much more it can gain from simply being mentioned. Its rank-one rate, however, rose to 30.6%, meaning it is being named first in nearly one in three qualified observations. That is where further gains would come from, not from additional mention presence.

The absolute counts: the brand held 99 rank-one placements out of 324 qualified observations in October 2026, up from 72 out of 315 in July 2026. Its valid recommendation count rose from 188 to 219 across the series.

Highest-priority diagnostic: which prompt categories are driving the rank-one gains, and whether the October pattern reflects changes in how AI systems surface the brand or in the source material those systems draw from.

Bizee (Incfile): coverage recovered while presence continued to erode

Coverage reached 48.8% in October 2026, up 2.9 points from September's 45.9% and within normal variation of July's 51.7%. The prior-to-current gain makes two consecutive months of recovery. The baseline-to-current decline of 2.9 points remains within the category's normal month-to-month variation.

Raw mention presence tells a different story. It fell from 71.8% in July 2026 to 64.5% in October 2026, a 7.3-point decline, among the larger moves recorded for any brand in the series. The brand is appearing in fewer AI answers even as its recommendation coverage stabilizes. Top-three rate held essentially flat, from 34.0% to 34.9%. Rank-one rate rose from 2.2% to 4.9% across the series.

The distinction between coverage and presence is central here. Bizee (Incfile) is being recommended in a similar share of qualified observations as in July, but it is being mentioned at all in a significantly smaller share. The recommendation rate among observations that do mention it has risen, which suggests the brand is appearing less often but converting those appearances into recommendations more efficiently.

The absolute counts: valid recommendation count held at 158 in October 2026, essentially flat with July's 163. Rank-one placements rose from 7 to 16 across the series.

Highest-priority diagnostic: which prompt types no longer mention Bizee (Incfile) at all, and which competitors are absorbing those lost appearances.

Harbor Compliance: three consecutive monthly gains

Coverage reached 11.4% in October 2026, up 2.8 points from July's 8.6% and up 1.3 points from September's 10.1%. That cumulative move remains within the category's normal month-to-month variation; Harbor Compliance is a brand to watch rather than one that has broken out of its typical range.

The sharper movement is in top-three placement. Harbor Compliance's top-three rate rose from 3.2% in July 2026 to 8.9% in October 2026, a 5.7-point gain, the largest shift in top-three placement recorded for the brand across the series. Raw mention presence rose from 23.8% to 25.3%. Rank-one rate rose from 0.3% to 1.8%.

The distinction to notice is that Harbor Compliance is being positioned inside top-three shortlists far more often than it was in July, even though its overall mention presence has moved only modestly. It is converting a similar presence base into better placements.

The absolute counts are small. Harbor Compliance held 29 top-three placements in October 2026 (out of 324 qualified observations), up from 10 in July 2026 (out of 315). Its valid recommendation count rose from 27 to 37. These are small numbers, so the percentage movements should be treated as directional rather than precise.

Highest-priority diagnostic: which prompt types are producing the top-three gains, and whether the increase is concentrated on specific AI surfaces.

Rocket Lawyer: presence fell while sentiment improved

Coverage reached 14.5% in October 2026, down 0.4 points from July's 14.9% and up 3.3 points from September's 11.2%. Neither move is outside the brand's typical month-to-month range. The brand ranks fifth in the category.

Raw mention presence fell from 25.1% in July 2026 to 19.8% in October 2026, a 5.3-point decline. Rank-one rate fell from 1.6% to 0.0% across the series. The brand held zero rank-one placements in October 2026, down from 5 in July 2026.

Net sentiment score rose from 0.7 to 0.8 across the series, meaning the tone of answers that do mention Rocket Lawyer improved slightly even as presence declined. The distinction is between visibility and description: the brand is mentioned less often, but when it is mentioned, the language is more positive.

The conflict data adds context here. Rocket Lawyer was the subject of 4 of the 9 detected factual inconsistencies, all related to pricing. Different AI platforms returned incompatible claims about whether the brand offers standalone purchases, what its membership costs, and what it charges for registered agent services. The presence decline in the benchmark data may be related to these inconsistencies, though the benchmark does not establish causation.

The absolute counts: valid recommendation count held at 47 in October 2026, flat with July's 47. Top-three placements fell from 17 to 13.

Highest-priority diagnostic: whether the pricing inconsistencies identified across platforms are contributing to reduced mention presence, and which prompt categories are driving the decline.

Inc Authority: coverage stable with improving presence

Coverage reached 11.4% in October 2026, up 0.6 points from July's 10.8% and up 2.2 points from September's 9.2%. The prior-to-current move is the sharper story, though neither move moves beyond the brand's normal range.

Raw mention presence rose from 14.3% to 18.2% across the series, a 3.9-point gain. Top-three rate held nearly flat at 3.7% from 3.5%. Rank-one rate fell from 1.6% to 0.6%, a 1.0-point decline. Net sentiment eased from 0.8 to 0.6.

The distinction to notice is that Inc Authority improved its presence in AI answers without converting that presence into top-three or rank-one placements. The brand is being mentioned more often, but not being recommended more prominently within those mentions.

The absolute counts: valid recommendation count rose from 34 to 37. Rank-one placements fell from 5 to 2. Top-three placements rose from 11 to 12.

Highest-priority diagnostic: whether the presence gains are concentrated on surfaces where the brand is less likely to be recommended, and which prompts are producing mentions without placements.

LegalZoom: stable coverage with a rank-one uptick

Coverage reached 50.9% in October 2026, down 1.2 points from July's 52.1% and up 4.1 points from September's 46.8%. Neither move is outside the brand's normal range. The brand ranks third in the category.

Raw mention presence held at 90.4% in October 2026, up 0.2 points from July's 90.2%. The brand is present in more than nine out of ten qualified observations, essentially tied with Northwest Registered Agent for the highest presence rate in the category. Top-three rate held at 26.2% from 26.4%. Rank-one rate rose from 5.7% to 6.2%.

The distinction is between presence and recommendation. LegalZoom is nearly always mentioned in AI answers about LLC formation services, but it converts that presence into top-three placements at roughly 26%, less than half the rate of Northwest Registered Agent at 61.7%. The brand is visible without being recommended as prominently.

The absolute counts: valid recommendation count held at 165 in October 2026, flat with July's 164. Top-three placements rose from 83 to 85. Rank-one placements rose from 18 to 20.

Highest-priority diagnostic: which surfaces or prompt types are failing to translate LegalZoom's near-universal presence into top-three placements, and which brands occupy those slots instead.

ZenBusiness: coverage stable, rank-one rate down from July

Coverage reached 58.3% in October 2026, down 0.1 points from July's 58.4% and up 0.6 points from September's 57.7%. Neither move is outside the brand's normal range. The brand ranks second in the category.

Raw mention presence rose from 82.2% to 85.8% across the series. Top-three rate held flat at 53.1% from 53.3%. Rank-one rate fell from 30.5% to 25.6%, a 4.9-point decline.

The distinction is between overall coverage and first-position wins. ZenBusiness holds the second-highest coverage rate in the category and a top-three rate of 53.1%, but it is being named first less often than it was in July. The brand's coverage is stable; its ability to win the top slot is not.

The absolute counts: valid recommendation count rose from 184 to 189. Top-three placements rose from 168 to 172. Rank-one placements fell from 96 to 83 across the series.

Highest-priority diagnostic: which prompts previously returned ZenBusiness first and now return another brand, and whether Northwest Registered Agent is the primary beneficiary.

Tailor Brands: stable with small-count sensitivity

Coverage reached 7.1% in October 2026, down 0.2 points from July's 7.3% and down 0.7 points from September's 7.8%. Neither move is outside the brand's normal range. The brand ranks eighth in the category.

Raw mention presence fell from 13.0% to 12.0% across the series. Top-three rate fell from 1.3% to 0.3%, a 1.0-point decline. Rank-one rate fell from 0.3% to 0.0%. Net sentiment rose from 0.6 to 0.7.

The distinction is between presence and recommendation conversion. Tailor Brands appears in roughly 12% of qualified observations but converts that presence into top-three placements less than 1% of the time. The brand is visible without being recommended.

The absolute counts: valid recommendation count held at 23 in October 2026, flat with July's 23. Top-three placements fell from 4 to 1. These are small counts, so percentage movements should be treated as directional.

Highest-priority diagnostic: whether the near-zero top-three conversion rate is consistent across all surfaces or concentrated on specific prompt types.

Swyft Filings: minimal presence with sentiment below neutral

Coverage reached 1.2% in October 2026, down 0.4 points from July's 1.6% and up 0.1 points from September's 1.1%. Neither move is outside the brand's normal range. The brand ranks ninth in the category.

Raw mention presence rose from 5.4% to 7.4% across the series. Top-three rate held at 0.3%. Rank-one rate held at 0.0%. Net sentiment fell from 0.4 to 0.2, meaning the tone of answers that mention Swyft Filings is only marginally positive.

The distinction is between presence and sentiment. Swyft Filings improved its mention presence across the series but saw its sentiment score decline. The brand is appearing in more AI answers, but the language surrounding those appearances is less favorable than in July. The conflict data shows that when asked about the difference between Bizee and Swyft Filings, ChatGPT and Gemini returned incompatible registered agent pricing figures for Swyft Filings, which may contribute to buyer confusion.

The absolute counts: valid recommendation count held at 4 in October 2026, flat with July's 5. Top-three placements held at 1.

Highest-priority diagnostic: which prompt types are producing the presence gains, and whether the sentiment decline is concentrated on specific surfaces or pricing-related queries.

Firstbase: three consecutive monthly increases from zero

Coverage reached 0.9% in October 2026, up 0.9 points from July's 0.0% and up 0.3 points from September's 0.6%. The cumulative move remains within the category's normal month-to-month variation. The brand ranks tenth in the category.

Raw mention presence rose from 0.0% to 1.5% across the series, a 1.5-point gain, the largest relative shift recorded for the brand across the series. Top-three rate rose from 0.0% to 0.3%. Rank-one rate held at 0.0%. Net sentiment rose from 0.0 to 0.6.

The distinction to notice is that Firstbase went from no presence at all in July 2026 to 5 present mentions in October 2026, of which 3 converted to valid recommendation placements. The absolute counts are extremely small. A single prompt changing outcome would move the percentage materially.

Highest-priority diagnostic: which specific prompts are surfacing Firstbase and whether those prompts represent a repeatable category or isolated queries.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters does the current benchmark actually capture?
• What pricing and comparison questions are buyers asking that the benchmark cannot yet answer?

In October 2026, all 324 qualified observations fell into the Brand Recommendation class. The benchmark's current prompt set can answer "which service should I choose?" but cannot yet answer pricing, value, or head-to-head comparison questions at the cluster level. This is a meaningful limitation given the conflict data, which shows that 8 of the 9 detected factual inconsistencies involve pricing figures. Buyers asking pricing questions through AI platforms are receiving contradictory answers, and the public benchmark does not yet capture those interactions in its qualified analysis set. The recommendation landscape is measured; the pricing and comparison landscape is not.

Brand Opportunity Summary

Questions This Section Answers

• Which brands have the clearest opportunity signals in the October 2026 benchmark?
• What diagnostic question should each brand investigate next?

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 data sits behind the recommendation coverage and conflict findings?
• Does source presence in the dataset establish causation?

The aggregate metrics are built from prompt-level observations capturing the query, the AI surface, the recommendation outcome, the rank, sentiment, and citations where exposed. Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence in the current dataset is not treated as proof of causation. The conflict data referenced in this report is drawn from the same observation set and reflects incompatible factual claims returned by different AI platforms for the same question.

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

Report-Specific Interpretation Notes

• Brands including Firstbase, Swyft Filings, and Rocket Lawyer have small qualified observation counts. Their percentages move on a handful of prompts and should be treated as directional rather than precise.
• The public metrics use the qualified observation count (324 in October 2026), not the raw prompt collection (800). All percentages reflect the qualified set.
• Month-over-month movement identifies where attention is warranted; it does not by itself establish the cause of that movement.
• The conflict data reflects factual inconsistencies detected across AI platforms. It is reported as an observation about platform behavior, not as a claim about any brand's actual pricing or service terms.

Next Step

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

Beneath the aggregate percentages sit the questions that matter for strategy: which high-intent prompts are won and lost, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. The benchmark shows the outcome; it does not reveal the mechanism.

A company-specific AI visibility audit maps prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. It answers the "why" behind the movement and identifies the specific queries and sources where intervention would change the recommendation outcome.

Request an AI visibility audit