How AI Search Is Recommending ABA Therapy Providers: Monthly Trends

Executive Summary

Action Behavior Centers remains the category's coverage leader in October 2026, but its lead is now materially narrower than at the start of the series. Its valid recommendation coverage stands at 33.3% in October 2026, down 20.7 points from 54.0% in July 2026, and the movement is large enough to register as a significant decline rather than ordinary month-to-month variation. The gap to the next brand, Hopebridge at 17.6% coverage in October 2026, narrowed to 15.7 points from 22.9 points at the July 2026 baseline, as both leaders contracted. No brand displaced Action Behavior Centers from first place.

LEARN Behavioral posted the strongest upward movement in the series. Its coverage rose 0.7 points from 8.1% in July 2026 to 8.8% in October 2026, and the prior-to-current step was sharper: up 3.9 points from 4.9% in September 2026. Its rank-one rate moved from zero at baseline to 3.9% in October 2026, the only significant improvement in a placement metric anywhere in the category this month. LEARN Behavioral recorded 9 valid recommendations in October 2026.

The sharpest decliners are concentrated in the category's upper tier. Behavioral Innovations fell 18.0 points from 29.8% in July 2026 to 11.8% in October 2026, a significant decline that deepened from the prior month (down 12.8 points from 24.6% in September 2026). Hopebridge fell a significant 13.5 points from 31.1% to 17.6% over the same span. Trumpet Behavioral Health fell a significant 5.2 points from 6.2% to 1.0%, leaving it with a single valid recommendation in October 2026.

The category sits in a broad contraction rather than a reshuffling. Four brands are classified as significant decliners against the July 2026 baseline, none as significant risers, and six as stable. October 2026 is the fourth month of the current series, and the qualified denominator in October 2026 is smaller than at the July 2026 baseline.

Each monthly benchmark run begins with the full prompt-surface collection before qualification. July 2026 collected 614 prompt-surface observations (475 unique questions) across the benchmark's defined AI/search surface universe; October 2026 collected 677 (496 unique questions). August 2026 collected 674 (543 unique questions) and September 2026 collected 720 (547 unique questions). All prompts in July 2026 mentioned a tracked brand or competitor; October 2026 recorded 672 brand or competitor mentions. Of the July 2026 collection, 213 prompts were relevant and 401 were irrelevant; of the October 2026 collection, 234 were relevant and 438 were irrelevant. The public metrics use the 161 qualified observations from July 2026 and the 102 from October 2026 that survived both qualification stages.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• Where did AI platforms give conflicting answers about ownership and compensation in the ABA therapy category?
• Which claims about Action Behavior Centers and Hopebridge contradicted each other across platforms?

The benchmark detected four critical or high-severity factual inconsistencies across three AI platforms. The conflicts involve two tracked companies and centre on corporate ownership and compensation questions, the two topics where prospective clients, employees, and referring partners are most likely to encounter contradictory answers.

Action Behavior Centers

Three high-confidence factual conflicts involve Action Behavior Centers, spanning ChatGPT, Copilot, and Perplexity, all triggered by ownership questions.

On the topic of acquisition deal value, AI platforms provided conflicting information about the size of the transaction. When asked "Who owns action behavior centers?", ChatGPT stated the company was "acquired in 2022 in a deal reportedly valuing the company at about $840 million," citing the company's own about page, Crunchbase, and a Becker's Behavioral Health report. Copilot stated the company was "sold in a deal valued at over $2B," citing the company's leadership page, its team page, and a LinkedIn profile. A flagged source on the ChatGPT side, an ABA industry news page, corroborated the approximately $840 million figure with a 0.98 confidence rating. The same sale cannot have been valued at both approximately $840 million and over $2 billion.

On the topic of current ownership, responses differed on whether the ownership structure is knowable at all. ChatGPT stated that "ownership structure is not publicly disclosed; no public filing identifies all equity owners or investors," citing a Better Business Bureau profile, the company's about page, and CorporationWiki. In a separate response to the same question, ChatGPT stated the company is "owned by Charlesbank Capital Partners, which acquired it in 2022," citing the company's about page, a NexPhase Capital announcement, and the same Becker's report. One answer claims current ownership is undisclosed while the other identifies a specific current owner, which cannot both be true.

On the topic of the current owner, a third conflict pairs Copilot and Perplexity. When asked "Who owns action behavior centers?", Copilot stated the company was "sold to private equity investors but details of the buyer are not widely publicized," citing the leadership page, the team page, and a LinkedIn profile. Perplexity stated the company is "owned by Charlesbank Capital Partners (private equity)," citing an ABA industry news page, the Charlesbank investment page, and the company's about page. The buyer cannot be both unpublicized and publicly identified as Charlesbank Capital Partners. Three flagged sources supported the Perplexity side, including the Charlesbank investment page and an ABA trade publication describing a landmark acquisition at an $840 million valuation in 2022.

Hopebridge

One high-confidence factual conflict involves Hopebridge, spanning Copilot and Perplexity, triggered by a compensation question.

On the topic of average annual salary, the claimed ranges do not overlap. When asked "How much does Hopebridge pay?", Copilot stated that "average annual salary across all roles is around $68,000–$113,000," citing ZipRecruiter and two PayScale pages. Perplexity stated that "annual salaries often cited around $37,939 on average, with range roughly $29,015–$60,990," citing the same two PayScale pages and an Indeed salary listing. Because the claimed ranges are $68,000–$113,000 versus $29,015–$60,990, they cannot both be true for the same company. A flagged source on the Copilot side, a PayScale salary page, showed an average base salary of $68k per year.

Benchmark Context

Questions This Section Answers

• How does the raw prompt collection differ from the qualified analysis set this month?
• Why did the qualified denominator shrink from July 2026 to October 2026?

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 surface breadth widened from four canonical AI surface families in July 2026 to all six in October 2026. ChatGPT and Perplexity entered the qualified set during the series, joining Copilot, Gemini, AI Mode, and AI Overviews, which carried qualified observations across the full period.

Benchmark-Level Metrics

August 2026 recorded a recommendation-shaped answer share of 43.4% and a valid recommendation shortlist share of 55.9%, and September 2026 recorded 53.3% and 51.6%. The recommendation-shaped share peaked at the July 2026 baseline and has not returned to that level in any subsequent month, while the valid recommendation shortlist share held near or slightly above half of qualified observations in every month. The qualified denominator contracted from 161 observations to 102 across the period even though raw collection and relevant-prompt counts both grew, reflecting the additional qualification criteria applied after the relevance filter.

AI Recommendation Trend

Questions This Section Answers

• Who leads the category in AI recommendation coverage, and how did the leadership tier change since July 2026?
• Which brands moved up, down, or held flat in October 2026?

A falling leader kept first place while the tier beneath it compressed toward single digits

Action Behavior Centers retained the top coverage position in October 2026, but the category now has one brand above 20% coverage where July 2026 had three.

Four brands crossed the benchmark's significance threshold against the July 2026 baseline, all of them downward, and no brand crossed it upward. The category-level contraction therefore reflects a combination of several brand-level movements rather than one dominant shift, although the declines at Action Behavior Centers and Behavioral Innovations account for the largest share of the total.

What Changed This Month

Questions This Section Answers

• Which brands posted significant coverage declines this month, and how did their presence and placement layers move?
• What signals distinguish Action Behavior Centers' decline from Behavioral Innovations' broader deterioration?
• Where is LEARN Behavioral's upward movement coming from in the ABA therapy recommendation set?

Action Behavior Centers

Action Behavior Centers' valid recommendation coverage fell 20.7 points from 54.0% in July 2026 to 33.3% in October 2026, a significant decline. The prior-to-current step was down 10.1 points from 43.4% in September 2026, meaning the contraction continued for a second consecutive month rather than stabilising. Raw mention presence fell a significant 27.3 points from 78.3% to 51.0% over the baseline-to-current span.

The brand still recorded 34 valid recommendations in October 2026 across 52 qualified observation appearances, including 33 top-three placements and 23 rank-one results. Its rank-one rate of 22.6% remains the highest in the category by a wide margin, and its average recommended rank of 1.5 is its best of the series.

The distinction to notice: presence is falling faster than placement. Action Behavior Centers is named less often in AI answers, but when it is named it still occupies the top slot more reliably than any other provider, and its top-three count held at 33. The gap that matters commercially is narrowing: its lead over LEARN Behavioral narrowed from 45.9 points at baseline to 24.5 points in October 2026, and its lead over BlueSprig Pediatrics narrowed by 20.7 points, shrinking in every month of the series.

Highest-priority diagnostic: Which prompts and surfaces reduced their mentions of Action Behavior Centers, and whether the same prompts now return a different provider or no provider at all.

Behavioral Innovations

Behavioral Innovations' coverage fell 18.0 points from 29.8% in July 2026 to 11.8% in October 2026, a significant decline that also deepened against the prior month, down 12.8 points from 24.6% in September 2026. Both the baseline and prior comparisons cross the significance threshold, making this the sharpest single-month move in the category this month.

The retreat is visible across every placement layer. Raw mention presence fell a significant 21.6 points from 42.2% to 20.6%; top-three rate fell a significant 16.3 points from 26.1% to 9.8%; and rank-one rate fell a significant 7.1 points from 8.1% to 1.0%. The brand recorded 12 valid recommendations in October 2026, down from 30 in September 2026 and 48 in July 2026.

The distinction to notice: Behavioral Innovations lost ground on presence, placement, and top-slot position at once, which is a broader deterioration than the leader's pattern. Its net sentiment score slipped from 0.8 to 0.6 across the period.

Highest-priority diagnostic: Whether the drop concentrated in a small number of surfaces, and which providers now occupy the top-three slots Behavioral Innovations previously held.

Hopebridge

Hopebridge's coverage fell a significant 13.5 points from 31.1% in July 2026 to 17.6% in October 2026. The prior-to-current movement was down 2.9 points from 20.5% in September 2026 and did not cross the significance threshold, so the largest part of the decline occurred earlier in the series.

The presence layer explains most of it. Raw mention presence fell a significant 13.4 points from 49.7% to 36.3% over the baseline-to-current span, while top-three rate fell 3.9 points to 14.7% and rank-one rate actually improved 1.6 points to 7.8%. Hopebridge recorded 18 valid recommendations in October 2026, 15 top-three placements, and 8 rank-one results.

The distinction to notice: Hopebridge is being mentioned less often but is holding its position at the top of the list when mentioned. Its rank-one count of 8 in October 2026 is two fewer than its July 2026 count of 10, a smaller change than the drop in overall coverage. Net sentiment softened from 0.8 at baseline to 0.5, the largest sentiment decline among the significant decliners.

Highest-priority diagnostic: Which surfaces stopped mentioning Hopebridge, and whether the retained rank-one placements sit in the prompts most likely to precede a provider inquiry.

Trumpet Behavioral Health

Trumpet Behavioral Health's coverage fell a significant 5.2 points from 6.2% in July 2026 to 1.0% in October 2026, with a prior-to-current movement up 0.2 points from 0.8% in September 2026 that did not cross the threshold. Raw mention presence fell a significant 6.4 points from 7.4% to 1.0%.

The absolute counts are now minimal. Trumpet Behavioral Health recorded 1 valid recommendation and 1 top-three placement in October 2026 across a single qualified observation appearance, a similarly minimal level to September 2026, down from 10 valid recommendations and 2 top-three placements in July 2026. It holds no rank-one results in any month of the series.

The distinction to notice: with one valid recommendation in the qualified set, a single additional observation would move this brand's coverage percentage by roughly a full point, so its month-to-month readings carry less information than the counts themselves. Its net sentiment score of 1.0 reflects one positive mention, not a broad sentiment position.

Highest-priority diagnostic: Whether the brand is absent from entire surfaces rather than losing position within them, and which providers occupy any slots it previously held.

LEARN Behavioral

LEARN Behavioral is the only brand in the category showing a sustained upward signal. Coverage rose 0.7 points from 8.1% in July 2026 to 8.8% in October 2026, and the prior-to-current movement was up 3.9 points from 4.9% in September 2026. Neither comparison crosses the significance threshold, but the direction is consistent across multiple layers.

Rank-one rate moved from 0.0% at baseline to 3.9% in October 2026, the only significant placement improvement in the category this month. Raw mention presence rose to 22.6%, and top-three rate rose 3.5 points to 7.8%. LEARN Behavioral recorded 9 valid recommendations, 8 top-three placements, and 4 rank-one results in October 2026.

The distinction to notice: LEARN Behavioral appears in a similar share of qualified observations to several peers but converts a higher proportion of those appearances into top-three and rank-one placement. Its coverage of 8.8% sits below its raw mention presence of 22.6%, which is the same visible-but-not-recommended gap that characterises most of the category. Its sentiment score eased from 0.6 to 0.5 even as placements improved.

Highest-priority diagnostic: Which prompts produced the four rank-one placements, and whether they cluster in a surface where the brand previously received no top-slot credit.

BlueSprig Pediatrics and the small-count tier

BlueSprig Pediatrics held coverage flat at 13.7% from July 2026 to October 2026, with a prior-to-current gain of 3.0 points from 10.7% in September 2026 that did not cross the threshold. Its top-three rate rose 6.2 points to 11.8% while raw presence eased 4.5 points to 14.7%. It recorded 14 valid recommendations in October 2026, up from 13 in September 2026.

Below it sit three brands whose absolute counts are small enough that individual observations carry outsized weight. Autism Learning Partners recorded 8 valid recommendations at 7.8% coverage, up 0.4 points from September 2026 but down 0.9 points from baseline. Acorn Health and Centria Autism both recorded 3 valid recommendations at 2.9% coverage, each down 1.4 points from July 2026. Helping Hands recorded 1 valid recommendation at 1.0% coverage.

The distinction to notice: for these brands, coverage percentages and recommendation counts should be read together. A 1.4-point move at Acorn Health and Centria Autism reflects a difference of roughly two observations, and Helping Hands' movement from 3.3% in September 2026 to 1.0% in October 2026 reflects a change of three valid recommendations.

Highest-priority diagnostic: For each brand in this tier, which specific prompts produced its valid recommendations, and whether those prompts repeat month to month.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters were captured in the qualified observations, and which were missing?
• What commercial questions about cost and head-to-head comparisons can this benchmark not yet answer for ABA therapy families?

All 102 qualified observations in October 2026 fell into the Brand Recommendation cluster, matching the pattern in July 2026 and every month since. The benchmark therefore measures which provider AI systems recommend when asked directly, and how consistently that recommendation holds across surfaces.

The public benchmark cannot yet answer pricing, value, or head-to-head comparison questions, because no qualified observations fell into those clusters in any month of the series. This is a meaningful blind spot for a category where families weigh insurance coverage, waitlist length, and session intensity before choosing a provider. The inability to state what AI systems say about cost or about one named provider versus another leaves those commercial questions unmeasured by this benchmark.

Brand Opportunity Summary

Questions This Section Answers

• Which diagnostic question should each ABA therapy provider prioritize based on its October 2026 signal?
• Which brands show a gap between presence and recommendation placement that warrants further investigation?

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations capturing the 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
• AI Visibility Industry Market Methodology
• AI Visibility Industry Market Metrics
• AI Visibility Industry Market Standards

Report-Specific Interpretation Notes

• The qualified set (102 observations in October 2026) differs from the raw collection (677 prompts); all percentages reference the qualified set.
• Small counts apply to several brands: Trumpet Behavioral Health and Helping Hands (1 valid recommendation each), Acorn Health and Centria Autism (3 each), and Autism Learning Partners (8) require caution in interpreting movement.
• Directional analysis identifies changes worth investigating; it does not by itself establish cause.
• The benchmark tests six canonical AI surface families; all six met the qualification bar in October 2026 (ChatGPT, Copilot, Gemini, Perplexity, AI Mode, AI Overviews).
• The qualified denominator contracted across the series, from 161 observations in July 2026 to 102 in October 2026, so percentage movements reflect both brand-level changes and the smaller base.
• The four critical or high-severity response inconsistencies reported above concern ownership and compensation questions and are not part of the recommendation coverage metric.

Next Step

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

The aggregate percentages only reveal the surface. Beneath the numbers sit questions the public benchmark cannot answer: which high-intent prompts a brand wins or loses, which competitor takes the recommendation when a brand is not chosen, what attributes AI associates with each provider, and which external sources shape those answers. For a category like ABA therapy providers, where families are making high-stakes care decisions and where AI systems are already returning contradictory ownership and compensation claims, those details determine whether a recommendation converts into an inquiry.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. The benchmark shows where attention is warranted; the audit explains why and what to do about it.

Request an AI visibility audit