Nootropics & Cognitive AI Visibility Index 2026

Mind Lab Pro led Nootropics / Cognitive visibility in the measured answers, appearing in 37.8% of 286 returned answers. Its 95% interval was 32.3% to 43.5%. Thorne ranked second at 33.9%, but the intervals overlap, so this sample does not establish that Mind Lab Pro is more visible than Thorne. This Nootropics / Cognitive report covers focus, memory, Lion's Mane, and brain fog prompts in the Supplements AI Visibility Index 2026.

The measured results show a category shaped by engine disagreement. Mind Lab Pro appeared across all four engines, while several brands depended on one engine for much of their visibility. Brands that ignore these differences risk reading one engine's output as a category-wide result.

Which brands lead the leaderboard?

Mind Lab Pro ranked first with a 37.8% appearance rate. Thorne followed at 33.9%, Onnit ranked third at 30.8%, and Qualia ranked fourth at 22.7%. Appearance rate uses answers returned as its denominator.

Mind Lab Pro also appeared early when named, with an average first mention position of 2.6. Thorne's average first mention position was 4.5, Onnit's was 7.8, and Qualia's was 5.3. A high appearance rate and an earlier first mention measure different parts of visibility.

This metric indicates brand visibility: it is the share of the 286 returned AI answers that named each brand.

The 95% intervals overlap across the first five ranked brands: Mind Lab Pro, Thorne, Onnit, Qualia, and Life Extension. The differences among those brands are within the margin of this sample. The leaderboard is useful for locating the measured field, but the overlapping intervals limit rank-to-rank claims inside that group.

The leading measured brands had different engine patterns. Each profile uses answers returned for appearance rates and answers with retrieval evidence for owned-domain citation rates.

Where do the engines disagree?

The engines disagree sharply for several brands, so a brand's category result cannot be inferred from a single system. Perplexity, ChatGPT, Gemini, and Claude use different models and retrieval tools. A gap can reflect system design rather than a stable difference in a brand's wider visibility.

Engine comparisons remain denominator-consistent because each cell uses that engine's own answer count.

ChatGPT named Thorne in 61.4% of its 70 returned answers, while Claude named it in 12.5% of its 72. ChatGPT also named Life Extension and NOW Foods in 47.1% of its 70 answers. Claude named Life Extension in 4.2% of 72 answers and did not name NOW Foods in 72 answers. Onnit had a split pattern too: 45.7% on ChatGPT and 44.4% on Gemini, against 11.1% on Claude.

Perplexity produced concentrated visibility for several names. It named NooCube in 34.7% of 72 answers, compared with 12.9% on ChatGPT, 11.1% on Gemini, and 15.3% on Claude. It named Innerbody Labs in 45.8% of 72 answers, while the other three engines did not name Innerbody Labs in their returned answers. Perplexity also named Focus in 40.3% of 72 answers, while Claude did not name it in 72 answers.

Gemini had its own concentration points. It named Nootropics Depot in 31.9% of 72 answers, compared with 1.4% on Perplexity, 24.3% on ChatGPT, and 4.2% on Claude. Gemini named Thesis in 27.8% of 72 answers, while ChatGPT named Thesis in 2.9% and Claude in 5.6%. Gemini also named Citicoline in 22.2% and Bacopa monnieri in 19.4%, while Claude did not name either in 72 answers.

Mind Lab Pro is the most even leader among the top brands, appearing from 26.4% on Gemini to 45.8% on Perplexity. Qualia is also comparatively even, ranging from 16.7% on Claude to 26.4% on Perplexity. The contrast with Thorne, Onnit, Life Extension, NOW Foods, NooCube, and Nootropics Depot shows that aggregate rank can conceal different engine dependencies.

Which brands lead each demand cluster?

The demand clusters separate broad cognitive supplement questions from more specific needs. The cluster results identify where a brand was most often named within the measured prompts, rather than proving consumer preference or product performance.

This view shows demand-cluster variation by regrouping the same answers according to the cluster from which each prompt was sampled; each cluster rate uses its own answer count.

The chart groups clusters under their brief headings. General buyer-intent clusters show which brands appear in broad selection questions. Sub-category and specific-need clusters show which brands surface when the request narrows to focus, memory, Lion's Mane, brain fog, adaptogens, or ADHD-adjacent non-prescription alternatives. These are different kinds of demand, so leadership in one group does not establish leadership in the other.

Which sites receive citations in the answers?

Innerbody Labs had the highest owned-domain citation rate among citation-eligible brands: 64% of 286 answers with retrieval evidence. Mind Lab Pro followed at 39.5%. Owned-domain citation rates use retrieval-evidenced answers as their denominator, not all answers returned.

This measures first-party sourcing: the share of the 286 answers with retrieval evidence that cited a URL on the brand's own website.

The citation chart separates being named from having a brand's own site cited. Life Extension appeared in 22% of returned answers and had a 14.7% owned-domain citation rate among retrieval-evidenced answers. Thorne appeared in 33.9% of returned answers and had a 10.5% owned-domain citation rate. Thesis, Pure Encapsulations, Neuriva, and Hunter Focus each had a 0% owned-domain citation rate among retrieval-evidenced answers, despite appearing in the returned answers.

Where the answers come from

Publisher citations were concentrated across a mix of brand-owned sites, health publishers, consumer guides, retailers, and research sources. Innerbody.com appeared in 64% of 286 answers with retrieval evidence. Healthline.com appeared in 34.3%, while Fortune.com and Forbes.com each appeared in 32.2%.

This identifies the publishers used as evidence, with each counted once per answer rather than once per link.

Mindlabpro.com was cited in 39.2% of retrieval-evidenced answers, and lifeextension.com in 14.7%. ConsumerLab appeared in 11.9%, Amazon in 14.3%, GoodRx in 13.3%, and PMC appeared in 10.5%. The measured citations therefore include owned sites and third-party publishers. This tally does not evaluate the editorial quality, certification status, or clinical evidence behind any source.

Who is winning and who is at risk?

Who is winning?

Mind Lab Pro is winning on measured aggregate visibility: it ranked first at 37.8%, had an average first mention position of 2.6, appeared on all four engines, and had a 39.5% owned-domain citation rate among retrieval-evidenced answers. Its interval overlaps several nearby ranks, so the result does not establish separation from those brands.

Thorne is winning on ChatGPT exposure, where it appeared in 61.4% of 70 returned answers, the strongest rate among its engine results. It ranked second at 33.9% overall and appeared on all four engines. Life Extension had a smaller overall appearance rate of 22%, but its owned-domain citation rate was 14.7% among retrieval-evidenced answers.

Who is at risk?

NOW Foods is at risk of engine concentration. It appeared in 47.1% of ChatGPT's 70 returned answers but was absent from Claude's 72 answers, with 8.3% on Gemini. Its 18.2% overall appearance rate masks that split.

Innerbody Labs is at risk of single-engine dependence. It was named in 45.8% of Perplexity's 72 returned answers and in none of the 70 ChatGPT, 72 Gemini, or 72 Claude returned answers. Its 64% owned-domain citation rate among retrieval-evidenced answers is high, but its cross-engine consistency was 25%.

Thesis, Pure Encapsulations, Neuriva, and Hunter Focus are at risk on citations. Each was citation eligible and had a 0% owned-domain citation rate among the 286 answers with retrieval evidence. Their measured appearance rates were 12.9%, 11.5%, 7.7%, and 7.3%, respectively.

What structural findings does the tally show?

The category has a visible top group, but the rank order inside it is uncertain. Mind Lab Pro's 32.3% to 43.5% interval overlaps Thorne's 28.7% to 39.6%, Onnit's 25.7% to 36.3%, Qualia's 18.3% to 27.9%, and Life Extension's 17.6% to 27.2%. The data supports identifying the leading measured brands, not declaring a settled order among those overlapping intervals.

Engine convergence is uneven. Mind Lab Pro, Thorne, Onnit, Qualia, Life Extension, Performance Lab, NooCube, Nordic Naturals, Nootropics Depot, Jarrow Formulas, Thesis, Pure Encapsulations, and Hunter Focus had 100% cross-engine consistency. Yet the per-engine appearance rates often differ widely. A brand can be present across all engines but receive most of its mentions from one system.

Owned domains and third-party publishers both carry citations. Innerbody.com, mindlabpro.com, and lifeextension.com are among the cited domains, alongside Healthline, Fortune, Forbes, ConsumerLab, Amazon, GoodRx, and PMC. The measurement does not establish a correlation between certification signals and citation share because the tally does not contain certification-status fields for the brands.

Among named brands requested in the brief, Four Sigmatic had a 5.6% appearance rate, Nordic Naturals had 15.4%, and Onnit had 30.8%. The tally does not measure Qualia Mind as a separate leaderboard brand. It records Qualia at rank 4 and includes Qualia Mind among its rolled-up names.

What changed during the measurement window?

The measurement records repeated responses, not a before-and-after series. It does not show category shifts during a measurement window. It does show repeat consistency differences, including 61.5% for Innerbody Labs, 50% for Neuriva, 42.6% for Mind Lab Pro, and 4% for Pure Encapsulations.

What should brands test next?

Brands should use the measured weaknesses as a test plan, then repeat the same category measurement to check whether the gap narrows.

• Check engine dependence for NOW Foods. Its rate was 47.1% in ChatGPT's 70 returned answers and 0% in Claude's 72. Publish product and ingredient pages that state the product name, intended cognitive use, ingredients, and limitations in plain language. Re-measure its appearance rate by engine and its cross-engine consistency.
• Reduce single-engine dependence for Innerbody Labs. It appeared in 45.8% of Perplexity's 72 returned answers and 0% in the other engines' returned answers. Review the cited Innerbody pages and publish equivalent factual pages that make product identity and claims easy to verify. Re-measure whether ChatGPT, Gemini, and Claude name Innerbody Labs.
• Build owned citation coverage for Thesis, Pure Encapsulations, Neuriva, and Hunter Focus. Each had a 0% owned-domain citation rate among 286 retrieval-evidenced answers. Publish pages that identify the product, ingredients, dosing information, safety information, and source material. Re-measure owned-domain citation rate using retrieval-evidenced answers.
• Improve answer placement for Onnit. It ranked third at 30.8% but had an average first mention position of 7.8. Create pages that answer narrow focus, memory, Lion's Mane, and brain fog questions before product detail. Re-measure average first mention position and appearance rate.
• Test Gemini-specific coverage for Nootropics Depot. Gemini named it in 31.9% of 72 returned answers, while Perplexity named it in 1.4%. Compare the pages cited or surfaced around its product names and ingredient terms, then repeat the engine-level measurement.
• Test Claude coverage for brands with low Claude rates. Thorne appeared in 12.5% of Claude's 72 returned answers, Onnit in 11.1%, Life Extension in 4.2%, and NOW Foods in 0%. Add clear source pages and repeat the same Claude measurement alongside the other engines.

What does this mean for brands in the category?

Interpretation: nootropics and cognitive supplement visibility is not one contest. A brand can rank well in aggregate and still depend on one engine. A brand can also be frequently cited by its own domain without broad cross-engine naming, as Innerbody Labs shows in this tally.

Interpretation: brands need separate measurement for being named, appearing early in an answer, and having their own site cited. The tally records distinct outcomes for each. Treating them as the same measure can conceal a weakness that appears when a consumer uses a different engine.

How was this measured?

The study provides a substantial cross-engine test: 24 stratified prompts were tested three times on Perplexity, ChatGPT, Gemini, and Claude. It produced 286 returned answers from 288 attempts. Appearance rates use returned answers as the denominator. Citation and source rates use 286 answers with retrieval evidence as the denominator.

The study measures these prompts on September 30, 2026, not every cognitive supplement query or consumer preference. Models and retrieval tools differ, brand names were rolled up where recorded, and overlapping 95% intervals limit rank comparisons. The report does not measure clinical efficacy, safety, market share, sales, or certification status.

How this study was measured

• Sampling was controlled in advance: 24 prompts were allocated across the demand clusters before any questions were asked.
• Cross-engine comparability was maintained by submitting the same questions to Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic), 3 trial(s) each.
• Return coverage was high: 286 answers returned of 288 attempted, 286 with retrieval evidence.
• Brand attribution was standardized: product lines rolled up to parent brands using a registry fixed before scoring.
• Uncertainty and evidence scope were stated explicitly: appearance rate reported with a 95% Wilson interval; citation metrics counted only on answers with evidence.

Charts and data tables

These names are not category brands, so they are measured consistently but excluded from the brand leaderboard.

Methodology appendix

The appendix supports auditability because it is generated directly from the measurement record for this study. It is included so any brand named here can reproduce or challenge the result.

What was measured

• Category: Nootropics / Cognitive
• Measurement date: 2026-09-30
• Depth: flagship: 24 prompts, 3 trial(s) per prompt per engine, 4 engines
• Answers attempted: 288
• Answers returned (appearance denominator): 286
• Answers with retrieval evidence (citation denominator): 286 (99.3% of attempted)
• Answers that failed outright: 0
• Analysis version: study-methodology-v2 · entity registry: entity-registry-2026-08-27

Claim limit: Variation can be observed both across prompts and across repeat generations because flagship depth covers 24 stratified prompts measured three times per engine.

How the prompts were sampled

Demand coverage is fixed rather than selected by the prompt writer: 14 buyer-intent need(s) from the client's brief are asked verbatim, one prompt each, grouped under: Nootropics / Cognitive; then , then a documented set of generic commercial demand clusters (general best, everyday mainstream choice, price/value, primary use case, beginner, retail channel, gifting/occasion, specific need, comparison, trust, ownership attribution, quality tier, durability, specialist, emerging) receives the remaining allocation round-robin. The prompt writer supplies wording inside these cells only and does not choose the category's demand structure.

Cross-engine gaps should be interpreted cautiously: each engine uses a different model, a different retrieval tool and a different answer format, so a gap between two engines can reflect system design as much as brand visibility.

How each metric is defined

• Appearance rate : This captures naming visibility as the share of returned answers that named the brand. It is measured on every answer the engine returned, whether or not it searched the web, so an engine's retrieval behaviour does not silently change the denominator.
• Macro appearance rate : This gives every prompt equal weight by taking the mean of the per-prompt appearance rates, regardless of how many engines answered it.
• 95% interval : This is the Wilson score interval on the appearance rate. Where two brands' intervals overlap, this study does not support a claim that one is more visible than the other.
• Owned-domain citation rate : This narrowly assesses use of the brand's own site as a source, as the share of answers with retrieval evidence that cited a URL on the brand's own domain; it is not a measure of total mentions.
• Share of all citations : This expresses the brand's own domain as a share of all attributed publisher citations in the study.
• Not measured : This means the brand's official domain could not be established, so no citation result exists. It is never reported as 0%.
• 0% : This means the metric could be evaluated and the event did not occur.
• Average position : This reports the mean position of the brand's first mention in the answer's ordered list of named brands.
• Cross-engine consistency : This shows the share of engines that named the brand at least once.
• Repeat consistency : This shows the share of repeated prompt/engine cells where the brand appeared in every trial. Lower values mean the engine's answer is unstable rather than the brand being absent.

Evidence rules

• Retrieval evidence is counted only when the engine returned a retrieval receipt (search annotations or grounding records). A URL the model merely typed into its prose is not treated as evidence that it searched.
• Publisher attribution follows the retrieval receipt: sources are assigned to the publisher behind each receipt. Intermediary and redirect hosts are never reported as publishers.
• Attribution was complete: publisher identity could not be established for 0 retrieval receipt(s); those are disclosed here rather than attributed or dropped.
• Source totals avoid link inflation: a publisher counts once per answer in which it was cited, not once per URL.

Entity handling

Brand-level results use a pre-specified attribution rule: product lines and model names roll up to their parent brand through a registry fixed before scoring. Nothing is merged on the basis of this study's own counts alone.

No roll-ups were applied in this study.

These names require review rather than inference: 40 further name(s) picked up in the answers were not confidently resolvable to a brand in this category and are held for human classification before the next study; they carry no metrics here.

Limits of these findings

• The findings are time-specific: this is a point-in-time measurement of what the named systems returned for these exact prompts on 2026-09-30. Engine behaviour changes without notice.
• Results are bounded by the sample: the prompt set is a stratified sample of category demand, not a census. Results describe these prompts and should not be read as representative of all searches in the category.
• Overlapping intervals limit comparative conclusions: differences between brands are not statistically distinguishable at this sample size.
• Omission does not establish absence: a brand not listed here was not measured as absent, it simply was not named in these answers.
• Appearance indicates visibility only, not an endorsement, a quality judgement or a market-share estimate.
• Citation coverage is necessarily limited: 5 of 26 ranked brand(s) had no official domain established before scoring, so they carry appearance results only and are excluded from citation metrics and source tables rather than being reported as zero.