How AI Search Is Recommending Online Therapy

Buyer discovery in online therapy is shifting from search engine results and paid advertising to AI-generated recommendations. When a consumer asks an AI assistant for the best therapy platform or compares pricing between services, the AI does not return a list of paid placements. It returns a synthesized shortlist built from available public sources. This changes where buyer decisions are formed and which platforms capture consideration at the moment intent is highest.

The LLM Authority Index benchmark for June 2026 reveals a market undergoing shortlist compression. Across 1,252 observations spanning discovery, comparison, and decision-stage prompts, two platforms dominate AI recommendations while a second tier of challengers competes for the remaining shortlist positions. Talkspace leads with the highest recommendation coverage and strongest shortlist presence across buyer stages. BetterHelp holds the second position with the strongest rank-one rate but trails in overall recommendation volume. Several well-known brands appear in AI responses at moderate rates but fail to convert that presence into ranked recommendations, creating a significant gap between visibility and commercial influence. CiteWorks Studio interprets this benchmark to show what the data means for competitive positioning in AI-led discovery.

Methodology

1. Market studied: Online therapy platforms, including direct-to-consumer therapy services, telehealth counseling platforms, and mental health support services marketed to general consumers.
2. Brands/entities included: BetterHelp, Talkspace, Grow Therapy, Brightside Health, Online-Therapy.com, 7 Cups, Amwell, Cerebral, Regain, and Calmerry. This universe covers the major publicly visible online therapy platforms but is not a full market census.
3. Data collection date/window: June 2026, based on a snapshot of AI platform outputs during the reporting month.
4. AI platforms tested: ChatGPT, Copilot, Gemini, Google AI Mode, Google AI Overviews, and Perplexity.
5. Number of prompts tested: Prompt count was not provided in the supplied dataset. A total of 1,252 observations were analyzed across all platforms and prompt clusters.
6. Prompt categories: Three public high-intent clusters were analyzed for this report: Best Online Therapy Platforms (consideration stage), Online Therapy Platform Comparisons (evaluation stage), and Online Therapy Pricing and Cost (decision stage). The full LLM Authority Index report includes ten clusters.
7. Definition of a mention: A mention means the company appeared in an AI-generated response, regardless of sentiment, framing, or rank position.
8. Definition of a valid recommendation: A valid recommendation is a positive, shortlist-quality or ranked recommendation that earns recommendation credit. Neutral references, cautionary mentions, comparison anchors, and listed-only appearances do not qualify as valid recommendations. This distinction is the core analytical framework applied throughout this report.
9. Ranking/scoring metrics used: Valid recommendation coverage, top-three rate, rank-one rate, top-ten rate, average recommended rank, net sentiment score, positive visibility rate, neutral visibility rate, negative visibility rate, modeled monthly AI authority value, modeled monthly AI recommendation value, modeled monthly AI visibility assist value, and captured share of AI opportunity.
10. Limitations: This is a point-in-time benchmark. AI outputs can change with model updates, source changes, and prompt variations. Modeled values are estimates based on commercial intent signals and are not revenue, pipeline, or booked sales. This report is not a full audit or full market census. Equal platform weights are applied across all six tested AI systems. The three prompt clusters described publicly represent a subset of the full ten-cluster benchmark.

Key Findings

Recommendation power concentrates in two platforms. Talkspace and BetterHelp together capture over 20 percent of the total modeled AI opportunity value of $21.9 million per month. Talkspace leads with a 33.3 percent valid recommendation coverage rate and a modeled monthly AI authority value of $2.32 million. BetterHelp follows with 22.8 percent recommendation coverage and $2.09 million in modeled authority value. Every other platform in the category falls below 20 percent recommendation coverage, with most below 10 percent, indicating a highly compressed shortlist structure.

Visibility does not equal recommendation credit. Several platforms with significant brand recognition appear in AI responses but fail to earn shortlist positions. Amwell appears in 12.2 percent of all observations but earns a valid recommendation in only 4.7 percent of cases. Cerebral shows a similar pattern, appearing in 11.6 percent of observations but achieving only 4.0 percent recommendation coverage. These platforms are present in AI responses. They are not being selected for buyer shortlists. The gap between presence and recommendation represents a structural commercial disadvantage.

Challengers show strong sentiment but weak rank positioning. Grow Therapy and Brightside Health achieve high net sentiment scores of 0.68 and 0.79 respectively, indicating positive framing when AI systems reference them. However, Brightside Health's average recommended rank of 3.5 and top-three rate of 7.2 percent suggest it is frequently listed but not prioritized. Grow Therapy performs better with a 16.2 percent top-three rate but still trails the leaders significantly in both recommendation coverage and modeled value.

The pricing and cost cluster carries the highest commercial weight. The decision-stage cluster applies a 1.5x buyer stage multiplier, making it the highest-intent prompt category in the benchmark. BetterHelp leads this cluster with a 23.8 percent top-three rate and 16.3 percent rank-one rate. Grow Therapy and Brightside Health both show 22.7 percent recommendation coverage in this cluster, suggesting they compete effectively when cost and pricing are the buyer's primary concern.

Platform-level variation creates uneven recommendation exposure. Talkspace performs consistently across all six AI platforms, with its strongest recommendation coverage on Google AI Mode at 41.2 percent and Perplexity at 39.6 percent. BetterHelp achieves its highest rank-one rate on Perplexity at 33.0 percent. Grow Therapy's recommendation coverage varies from 5.7 percent on Perplexity to 31.9 percent on Google AI Mode, indicating platform-specific source dependencies that may be limiting its overall recommendation reach.

What Changed in the Market

Buyers searching for online therapy are no longer moving only from Google results to brand websites. They are also asking AI systems to compare providers, explain reputation, summarize pricing, surface alternatives, and recommend shortlists. This shift matters because therapy is a trust-heavy category where legitimacy, clinical credibility, verified reviews, and third-party validation carry significant weight in purchase decisions. AI systems that synthesize public sources to produce recommendations are functioning as de facto shortlist builders for a category where buyer hesitation is high and information asymmetry is significant.

The benchmark shows that recommendation power in online therapy is not evenly distributed. Talkspace appears in 70.5 percent of all observations and earns a valid recommendation in 33.3 percent of cases. BetterHelp appears in 59.5 percent of observations with 22.8 percent recommendation coverage. Every other platform in the category falls below 20 percent recommendation coverage. This creates a two-tier market where the leaders capture the majority of AI-driven buyer attention and the remaining platforms compete for a shrinking share of shortlist positions.

For a category as sensitive as mental health, framing quality matters beyond commercial positioning. AI systems that reference complaints, regulatory issues, or negative press are less likely to recommend a platform with confidence. Platforms with clean, positive, and clinically grounded public source profiles are more likely to earn recommendation credit across discovery, comparison, and decision-stage prompts. The benchmark's sentiment scores and negative visibility rates reflect this dynamic directly.

The three public prompt clusters analyzed here represent the most common buyer intents: identifying available platforms, comparing services, and evaluating cost. In each of these moments, the AI acts as a shortlist curator. Buyers do not see the full competitive field. They see the platforms that AI systems surface with confidence. Being mentioned somewhere in an AI response is not the same as being placed on that shortlist.

What the Benchmark Found

Recommendation Leaders

Talkspace is the recommendation leader in the online therapy category by the benchmark's primary metrics. It appears in 70.5 percent of all observations and earns a valid recommendation in 33.3 percent of cases, the highest recommendation coverage in the dataset. Its top-three rate of 30.3 percent and rank-one rate of 13.0 percent demonstrate consistent shortlist dominance. Talkspace achieves an average recommended rank of 1.9, meaning when it is recommended, it typically appears in the first or second position. Its modeled monthly AI authority value of $2.32 million represents 10.6 percent of the total category opportunity.

BetterHelp holds the second position and leads the category on one important metric: rank-one rate at 14.8 percent. Its average recommended rank of 1.47 is the best in the category, meaning when BetterHelp is recommended, it tends to appear at the top of the shortlist. However, its 22.8 percent recommendation coverage trails Talkspace significantly, and its net sentiment score of 0.42 is lower than several challengers. BetterHelp's modeled monthly AI authority value of $2.09 million represents 9.5 percent of the category opportunity. The benchmark identifies BetterHelp as a rank-one leader with narrower recommendation breadth than Talkspace.

Challenger Tier

Grow Therapy has established the strongest position in the challenger tier. It appears in 31.9 percent of observations with 18.9 percent recommendation coverage. Its net sentiment score of 0.68 is the second highest in the category, indicating strong positive framing in AI responses. Grow Therapy achieves a top-three rate of 16.2 percent and a rank-one rate of 5.5 percent. Its modeled monthly AI authority value of $498,636 represents 2.3 percent of the category opportunity. The gap between Grow Therapy and the two leaders is substantial in recommendation coverage but less severe in sentiment quality.

Brightside Health holds the highest net sentiment score in the category at 0.79. When AI systems reference Brightside Health, the framing is strongly positive. It appears in 22.8 percent of observations with 16.4 percent recommendation coverage. However, its average recommended rank of 3.5 and top-three rate of 7.2 percent indicate it is frequently listed but not prioritized. Its modeled monthly AI authority value of $356,867 represents 1.6 percent of the category opportunity. The benchmark positions Brightside Health as a specialist option with strong framing quality but limited shortlist priority.

Online-Therapy.com and Regain each hold positions in the lower challenger range with recommendation coverage below 10 percent and average recommended ranks near or above 3.5. Calmerry and 7 Cups show limited recommendation coverage and modeled authority value below $200,000 per month, placing them in the under-cited challenger segment.

Visible but Under-Recommended

Amwell appears in 12.2 percent of observations but earns a valid recommendation in only 4.7 percent of cases. Its top-three rate is 1.9 percent, and its average recommended rank of 4.0 places it at the bottom of most AI-generated shortlists when it appears. Its modeled monthly AI authority value of $89,455 represents 0.4 percent of the category opportunity. Amwell has visible presence but commercially weak recommendation positioning.

Cerebral follows a similar pattern with 11.6 percent observation presence and only 4.0 percent recommendation coverage. Its top-three rate of 1.6 percent and average recommended rank of 3.7 suggest it is named in AI responses but not advanced into shortlists. Its modeled monthly AI authority value of $84,547 represents 0.4 percent of the category opportunity.

Platform-Specific Patterns

Talkspace performs consistently across all six AI platforms, with its strongest recommendation coverage on Google AI Mode at 41.2 percent and Perplexity at 39.6 percent. BetterHelp achieves its highest rank-one rate on Perplexity at 33.0 percent, suggesting strong source support on that platform. Grow Therapy's recommendation coverage ranges from 5.7 percent on Perplexity to 31.9 percent on Google AI Mode, a variation that suggests platform-specific source gaps rather than a uniform competitive disadvantage.

Prompt-Cluster Patterns

In the Best Online Therapy Platforms cluster, Talkspace leads with a 29.3 percent top-three rate and 15.6 percent rank-one rate. Grow Therapy performs well here with a 17.1 percent top-three rate and 7.2 percent rank-one rate. In the Online Therapy Platform Comparisons cluster, Talkspace again leads with a 29.3 percent top-three rate and 12.6 percent rank-one rate, while Grow Therapy drops to 13.7 percent top-three, suggesting its comparison-stage positioning is less developed than its discovery-stage presence. In the Online Therapy Pricing and Cost cluster, BetterHelp leads with a 23.8 percent top-three rate and 16.3 percent rank-one rate, with Grow Therapy and Brightside Health both at 22.7 percent recommendation coverage, indicating stronger competitive parity at the decision stage.

Why Visibility Is Not Enough

A brand can appear in AI answers and still fail to win the buyer shortlist. The benchmark data makes this distinction concrete. Raw mention presence measures how often a company is referenced in AI responses. Valid recommendation coverage measures how often a company is actually recommended or shortlisted. These are not the same signal, and conflating them produces a false picture of competitive strength.

Amwell and Cerebral illustrate this gap clearly. Amwell appears in more than one in ten AI responses but earns shortlist credit in fewer than one in twenty. Cerebral follows the same pattern. Both platforms have some presence in the AI information layer. Neither is being selected for the buyer shortlist at a commercially meaningful rate. The gap between observation presence and recommendation coverage is where buyer opportunity is won or lost.

Top-three placement and rank-one placement carry more commercial weight than raw mention count. Talkspace's 30.3 percent top-three rate drives its $2.32 million modeled monthly authority value. Brightside Health's high net sentiment score of 0.79 does not compensate for its 7.2 percent top-three rate. Positive framing without rank priority does not translate into recommendation-stage influence. BetterHelp's 1.47 average recommended rank is the best in the category, yet its lower recommendation coverage frequency means it reaches fewer buyers overall than Talkspace, even when its rank position is stronger when it appears.

Sentiment and framing quality matter, but they are not sufficient on their own. A neutral or mixed mention, even without negative content, is not recommendation credit. BetterHelp's 1.6 percent negative visibility rate, the highest in the category, introduces framing risk that may limit its recommendation breadth even as its rank-one rate leads the field.

Modeled monthly AI authority value is a benchmark estimate. It reflects the relative commercial weight of recommendation positions based on prompt volume, buyer intent signals, stage multipliers, and rank position. It is not revenue, pipeline, or booked sales. It is a comparative signal that shows where recommendation value is concentrating in the category and which platforms are capturing or missing that concentration.

The Citation Layer

AI systems synthesize answers from available public sources. The citation architecture that shapes online therapy recommendations appears to draw from several distinct source types. Official brand sites provide foundational service information including pricing structures, therapist credentialing, and platform features. Editorial reviews and comparison articles from health publications and consumer guides help AI systems distinguish between platforms and form ranked opinions. Clinical directories and professional references add credibility signals that matter in a regulated, trust-sensitive category. Review platforms and user forums, including Reddit communities, contribute sentiment signals and real-world experience accounts that shape framing quality. Insurance and telehealth ecosystem pages may support specific claims about coverage acceptance and clinical scope.

Platforms with higher recommendation coverage appear to benefit from more extensive and more authoritative public evidence layers. Talkspace and BetterHelp have accumulated substantial review content, comparison references, clinical mentions, and brand authority signals across a wide range of source types. This gives AI systems more retrievable material to synthesize and more confidence to recommend. Brightside Health's high net sentiment score of 0.79 suggests that when sources do reference the platform, they frame it positively. The challenge is that those positive sources may not yet be numerous or widely distributed enough to drive consistent high-frequency recommendations across all six AI platforms.

Grow Therapy's variation between platforms, from 5.7 percent recommendation coverage on Perplexity to 31.9 percent on Google AI Mode, may indicate that its public evidence layer is stronger on some platforms' source networks than others. This kind of platform-specific variation often reflects differences in which sources each AI system prioritizes when synthesizing answers.

Platforms that appear in authoritative comparison guides, clinical directories, verified review platforms, and well-cited editorial content are more likely to be recommended than those that rely primarily on brand advertising or direct traffic. The gap between presence and recommendation for platforms like Amwell and Cerebral may reflect a public source footprint that AI systems can acknowledge factually but cannot endorse with confidence. The source layer shapes not just whether a platform is mentioned but whether it is trusted enough to be recommended.

What Brands Need to Fix

Weak valid recommendation coverage is the primary problem for the majority of platforms in this category. The gap between observation presence and recommendation credit is the most commercially actionable signal in the benchmark. Platforms appearing in AI responses without earning shortlist positions need to understand which source gaps or framing issues are preventing recommendation credit.

Low top-three and rank-one presence limits commercial exposure even for platforms with moderate recommendation coverage. Average recommended rank is a meaningful indicator. Platforms ranked 3.5 or lower on average are frequently listed but rarely lead the buyer's shortlist. Improving rank position requires a stronger source signal that positions the platform as a primary recommendation rather than an available alternative.

Inconsistent prompt-cluster coverage creates gaps in the buyer journey. Platforms that perform well at the discovery stage but drop at the comparison or pricing stage are losing buyers at the moment intent is highest. Consistent recommendation coverage across all buyer stages is necessary to capture the full purchase journey.

Neutral or mixed framing in public sources reduces recommendation confidence. BetterHelp's net sentiment score of 0.42 and 1.6 percent negative visibility rate indicate framing risk. Platforms with any negative visibility need to understand which sources are generating that framing and whether those sources are reshaping AI recommendations.

Thin or uneven source footprints prevent AI systems from recommending platforms with confidence. Clinical references, verified reviews, comparison content, editorial coverage, and authoritative third-party citations all contribute to the public evidence layer that AI systems draw from. Platforms that lack depth across these source types may be systematically under-recommended regardless of brand awareness or advertising investment.

Inconsistent entity information across the public web reduces the reliability of AI-synthesized answers. Pricing details, therapist qualifications, insurance acceptance, platform features, and clinical scope need to be described consistently across owned and third-party sources so AI systems can form accurate and confident recommendations.

Weak review and comparison visibility is a structural gap for platforms that operate primarily through direct marketing. Review platform presence, comparison article inclusion, and consumer guide citations are part of the citation architecture that shapes AI shortlists. Platforms absent from these source types are operating with a thinner evidence base than competitors who have invested in them.

How CiteWorks Studio Helps

1. Map AI recommendation visibility. Track prompts, platforms, company presence, valid recommendations, top-three and rank-one performance, framing quality, and citation sources across the online therapy category and against specific competitors.
2. Identify the sources shaping AI answers. Find the editorial, review, clinical, directory, forum, and owned sources that influence brand framing and recommendation positioning on each AI platform.
3. Build the citation architecture plan. Strengthen the public evidence layer so AI systems have more accurate, consistent, and persuasive source material to synthesize when buyers ask about therapy platforms.

Commercial Takeaway

The online therapy category is experiencing shortlist compression driven by AI recommendation patterns. Two platforms capture the majority of recommendation value, and the remaining eight platforms compete for a structurally smaller share of AI-driven buyer attention. This compression is likely to intensify as AI systems become more embedded in consumer health decisions and buyers increasingly rely on AI-generated shortlists rather than unstructured search results.

Competitor displacement is already visible in the benchmark data. Platforms with strong brand recognition but weak recommendation coverage are being passed over at the moment buyers form their shortlists. The gap between the two leaders and the rest of the field is not marginal. It spans an order of magnitude in recommendation coverage and modeled benchmark value. Platforms outside the top two are not simply losing share. They are losing the moment where buyer decisions are made.

The opportunity is to improve recommendation-stage visibility, not merely chase AI mentions. Platforms that invest in the public evidence layers that AI systems trust, including clinical references, verified reviews, comparison content, and authoritative citations across multiple source types, are more likely to earn higher recommendation rates and stronger rank positions. Platforms that rely on advertising reach or direct traffic without building that evidence layer will find themselves increasingly invisible to buyers who are asking AI systems to make the first selection.

Find Out Where You Stand in AI Recommendations

The benchmark shows the market shape for the online therapy category. A company-specific analysis would show which prompts each platform wins or loses, which AI platforms are under-recognizing the brand, which source layers are shaping recommendations, and where changes to the public evidence layer may improve AI shortlist eligibility.

CiteWorks Studio can show where your brand appears in AI responses, where competitors are being recommended instead, which prompt clusters carry the most commercial risk, which sources appear to be shaping AI answers, and what needs to change to improve recommendation-stage visibility. Request an AI Visibility Audit, an AI Company Discovery Report, or a Citation Architecture Review to see your brand's specific position in the benchmark landscape.

Benchmark Source

This analysis is based on the 2026 AI Market Discovery Index for Online Therapy, published by LLM Authority Index. The benchmark dataset and public report were supplied for this category analysis. Read the full benchmark report at the LLM Authority Index.