540 responses, 4 models, 31 brands: the first AI-visibility snapshot of Turkey's cosmetics market

Vertical: Cosmetics, skincare, makeup & personal care — Turkey Method: A single point-in-time study of 540 responses — four large language models (Gemini, ChatGPT, Claude, Grok), each asked 27 Turkish-language questions five times Collection date: 24–25 May 2026 Author: Mert Can Elkaya · Herm.io AI Visibility Database

Key terms: AI visibility, AI Visibility Score, cosmetic brands, Turkish cosmetics market, Turkish (yerli) cosmetic brands, LLM brand recommendations, ChatGPT/Gemini/Claude/Grok brand recommendation, Generative Engine Optimization (GEO).

This report reproduces the real Turkish-language questions consumers ask AI assistants and measures how often, in what order, and across how many models four large language models name each cosmetic brand. It is a single point-in-time study and should be read within the limitations in Chapter 10.

How to read this report — visibility is not quality

This report measures one thing: how often, how prominently, and across how many AI assistants a brand is named when people ask about cosmetics in Turkey. A high score means these systems currently have a lot of information available about a brand and surface it readily — not that the brand is better, safer, healthier or more effective. A low score, or a zero, means a brand is currently less visible to these systems — not that it is inferior. AI visibility reflects information availability and discovery, not product quality. Readers and consumers should not treat an AI recommendation — or this report — as a verdict on whether a brand is “good” or “bad.” This study does not measure the accuracy, quality, or sentiment of any recommendation.

1. Executive Summary

One-sentence takeaway: AI visibility in Turkey’s cosmetics market is concentrated in a handful of brands; the leader, The Purest Solutions, surfaces in more than half of all responses and across all four models, while the makeup and dermocosmetic segments are surfaced by different brands.

• The Purest Solutions is the runaway leader. It appeared in 54.4% of all 540 responses, surfaced in all four models, and tops the table with a 96.3 Visibility Score. There is a clear gap to second-placed Golden Rose (85.6).
• The whole top five are established, widely-known houses. Golden Rose (85.6), Flormar (78.6), Pastel (75.3) and Farmasi (68.3) complete the top five; all appear in all four models.
• 31 Turkish brands qualified for ranking. Of 138 tracked brands (100 Turkish, 38 foreign), only 31 cleared the 5% mention threshold. The qualified brands’ average score is 52.2 — meaning the list thins out quickly past the middle.
• The models see the same market differently. Grok has the widest repertoire (~9.7 distinct Turkish brands per answer); Claude the narrowest (~5.7). More strikingly: Claude is the only model that surfaces Farmasi most.
• Web search changes the leaders. In cited (web-search-grounded) answers, The Purest Solutions leads; in un-cited answers drawing on the model’s own memory, Farmasi rises to the top. This exposes a “new-generation digital brands vs. legacy memory brands” divide. (Note: the memory segment is small — 58 responses — so this finding is indicative, not definitive.)
• The open market is balanced between Turkish and foreign. Across the 5 origin-neutral questions, mentions split 50.9% foreign / 49.1% Turkish; yet the top four individual brands are all Turkish (Flormar, Pastel, Golden Rose, The Purest Solutions). Foreign brands are more visible on the dermocosmetic shelf. (This split rests on a small base — 5 questions, 100 responses — and should be read as a signal to watch across editions.)
• Categories are surfaced by different brands. Makeup discovery is led by Pastel/Flormar/Golden Rose; clean/vegan attributes by The Purest Solutions; skincare/haircare use-cases by Dermoskin/Cosmed/Maruderm/Bioxcin.
• Sources are fragmented. About 46% of the sources models cited form a long tail of hundreds of small sites, followed by e-commerce/retail (15.9%) and editorial beauty media. The most-cited individual domains: oggusto.com, hopi.com.tr, dermoeczanem.com.

Why it matters. Consumers are shifting product discovery from search engines to AI assistants. Whether a brand appears in those assistants’ answers is becoming a discovery channel in its own right. This report captures the first snapshot (baseline) of this new “visibility shelf” for Turkey’s cosmetics market: which brands stand out, why the models diverge, and how a brand can read its own position. A reminder: the numbers below describe visibility and information availability — not which brands are best.

2. Methodology

One-sentence takeaway: Four large language models were asked 27 Turkish questions five times each — with no system prompt — to produce 540 responses; brands were identified by automated extraction plus human validation and ranked with a three-component (45/30/25) score.

2.1 Scope and models

The study queried four large language models directly via API with identical questions. Each model was called through its provider’s official API; all models received only the user question — no system prompt was used, temperature was not set manually (provider default), and no region/locale parameter was defined. Because the questions are in Turkish, models infer the Turkish market from language alone; an explicit locale was deliberately not set.

Scale: 27 questions × 5 repeats × 4 models = 540 responses, all completed. Single run, 24–25 May 2026. Tracked brands: 138 (100 Turkish, 38 foreign). Across all answers, 4,111 brand mentions were captured in 521 of the 540 responses.

2.2 Web-search configuration

Web search was offered as an enabled tool to all four models; but each model decided for itself whether to use it when answering a given question. Search was neither forced nor disabled. Consequently, differences in “web-search usage rate” across models reflect each model’s own behavior, not an external constraint.

Whether a response drew on web search or on the model’s own knowledge was determined by whether the provider returned any citation/source: if a citation came back, “web-search”; if none, “own knowledge.” This is a reasonable proxy, not direct proof of the model’s internal process.

2.3 Questions

The 27 questions were built from real search queries, keyword research and user research, and distributed equally across three behavioral categories (9 each):

• find (discovery): general recommendation questions of the “best / popular / standout brands” type.
• attribute: questions by specific qualities — vegan, cruelty-free, affordable, natural-ingredient, high-quality, etc.
• use_case: need-driven questions — oily/sensitive/dry/mature skin, acne, men’s care, oily hair, etc.

Of the 27, 22 are “constrained” (they explicitly ask for Turkish / yerli brands) and 5 are “open” (they ask for the best/best-selling brands available in Turkey, with no origin restriction). The Turkish-vs-foreign comparison is based solely on these 5 “open market” questions (100 responses) — the only question group without an origin restriction, and thus the only fair basis for comparison.

2.4 Brand extraction and validation

Brands were extracted in two stages. (1) Automated extraction: emphasized brand names in responses and brand labels in structured API sources were collected; spelling/format variants were merged into a single canonical name. (2) Human validation: the automated candidate list was manually reviewed by the team; real brands were confirmed, non-brands (content/category/UI words, ingredients, retailers) were filtered out, aliases merged, and each brand labeled Turkish/foreign. Final dictionary: 138 brands.

Origin classification rule. A brand is labeled Turkish if it was founded and is headquartered in Turkey; it is labeled foreign if it is of foreign origin, even when manufactured or widely sold in Turkey. Borderline cases — brands with Turkish roots later acquired by foreign owners, or foreign brands with large Turkish operations — were decided case by case and are inherently debatable; these edge cases are noted in Chapter 10.

Definitions.

• Mention: A brand is counted once for a response if its name (or a validated alias) appears in the answer text (regardless of how many times it occurs). It is binary per response: mentioned / not.
• Position (mention rank): The order of a brand’s first appearance within a single answer (1 = first brand named). Lower is better.
• MRR (mean reciprocal rank): The average of 1/rank (first = 1.0; second = 0.5; third = 0.33…). A position metric independent of list length.
• Breadth (models covering): The number of models that mention the brand at least once (0–4).

2.5 AI Visibility Score (0–100)

The AI Visibility Score is a weighted blend of three normalized components:

Score = 0.45 × Mention + 0.30 × Position + 0.25 × Breadth

Each component is scaled so the leading qualified brand = 100, then combined with the weights. The score blends a brand’s frequency, position and reach.

Why these weights. Mention frequency carries the most weight because being named at all is the primary signal of visibility; position is second because a brand named early in an answer is more likely to reach the reader; breadth is third because appearing across multiple models indicates a more robust, less model-specific presence. The weights are a deliberate editorial choice. At the top of the leaderboard the gaps are large enough to be robust to reasonable reweighting; in the tightly-packed middle of the table, small score differences (a point or two) should not be over-interpreted.

Qualification (≥5% rule): Only Turkish brands mentioned in at least 5% of the 540 responses are ranked (31 brands). This threshold prevents rare brands that happen to appear first in a handful of answers from inflating the position component. Unfiltered metrics for all 138 brands are available on request (see Data availability, Chapter 11).

2.6 Neutrality and self-exclusion

To prevent any conflict of interest, all citations to Herm.io’s own domains were removed from the dataset before analysis. Any AI response that referenced a Herm.io page had that citation stripped, so the company’s own content could neither appear among the most-cited domains nor influence the source-type mix or any brand’s visibility figures. Herm.io is the publisher of this report; excluding its own footprint ensures the company does not measure, cite, or benefit from itself.

3. Overall Visibility Leaderboard

One-sentence takeaway: The Purest Solutions leads by an undisputed 10.7-point margin; the top ten mix established makeup houses with the mid-tier dermocosmetic field, and the 31-brand list quickly tapers into a long tail.

The list below ranks all 31 Turkish brands that cleared the 5% threshold, by AI Visibility Score. As a reminder, this ranks visibility, not brand quality.

Average score of the 31 qualified brands: 52.2.

3.1 Tier narrative

The runaway leader (96.3). The Purest Solutions maxes out both the mention (100/100 component) and breadth components, with a very strong position component (87.8) too. It appears in 54% of all responses — more than one in two answers. The 10.7-point gap to second place is a break not seen anywhere else on the list.

The established makeup houses (ranks 2–5). Golden Rose, Flormar, Pastel and Farmasi — Turkey’s entrenched color-cosmetics brands. A telling nuance: Golden Rose’s position component is 100 (i.e., when it is mentioned it is usually named near the top), but its total mention frequency is lower than the leader’s, so it sits second. Pastel, by contrast, has high mentions (38.7%) but a weaker position (avg. 4.0), placing it fourth.

The dermocosmetic mid-tier (ranks 6–11). Dermoskin, Cosmed, Maruderm, Gülsha, Rosece and Bioxcin — the backbone of skincare and pharmacy cosmetics. Bioxcin is especially notable: despite a low mention rate (11.8%), it ranks 11th because when it is mentioned it appears very high (position component 75.8) — a niche but high-prominence positioning.

The long tail (ranks 12–31). Scores fall from 53.6 to 23.6 here. Most brands still appear in all four models (breadth 100), but both their mention and position components decline. At the very bottom, Beaulis (3 models) and Bionnex (2 models) sit lowest because their breadth component also drops. A lower rank here means lower current visibility to AI systems — not lower brand quality — and many capable brands sit below the 5% threshold entirely.

4. Differences Between Models

One-sentence takeaway: The four models see the same market at different breadths and with different leaders; Grok offers the widest list, while Claude both gives the narrowest list and is the only model that elevates Farmasi to #1.

4.1 Per-model behavior summary

Each model produced 135 responses. Web-search usage varies markedly: Grok searched on every answer (100%), while Claude used no search at all in roughly a quarter of its answers (74.8%). This difference was not externally imposed; each model decided on its own.

On breadth, Grok is clearly ahead: on average 9.7 distinct Turkish brands per answer and 79 distinct Turkish brands in total. Claude is the most conservative list-maker: 5.7 Turkish brands per answer, 58 in total. On foreign-brand variety, Claude (31) and ChatGPT (29) lead — i.e., Claude gives a smaller but more global mix.

4.2 Each model’s top 8 brands

The tables below give each model’s 8 most-mentioned Turkish brands across its 135 responses (rate = % of answers mentioning the brand).

Gemini — discovery-led, dermocosmetic-heavy:

ChatGPT — balanced makeup + dermocosmetic, high mention rates:

Claude — a different leader: Farmasi at #1:

Grok — the widest and most niche-friendly list:

4.3 The disagreement story

Two findings stand out. First, Claude’s different leader. Three of the four models place The Purest Solutions at the top, but in Claude the leader is Farmasi (54.1%), with The Purest Solutions only fourth (40.0%). The likely reason is that Claude has the lowest web-search rate (74.8%): it relies on its own memory more often, and in memory-based answers Farmasi rises to the top (see Chapter 5).

Second, Grok’s breadth elevates niche brands. Grok’s top 8 features brands that lag in other models at very high rates — Rosece (52.6%), Gülsha (51.9%) and Misbahçe (41.5%). Because Grok searches on every answer and lists ~9.7 brands on average, the chance of long-tail brands being captured rises markedly in this model. The practical reading: where a brand is visible depends heavily on which model is asked.

5. Web Search or Model Memory?

One-sentence takeaway: Web-grounded answers elevate new-generation digital brands (especially The Purest Solutions), while answers from the model’s own memory elevate legacy memory brands (Farmasi above all) — though the memory segment is small and this pattern should be confirmed in future editions.

The 540 responses were split into two segments: web-search answers carrying citations (482 responses) and own-knowledge answers with no citations (58 responses). Turkish-brand mention rates in the two segments reveal which brands surface “via search” versus “via memory.” Because the own-knowledge segment is small (58 responses), the figures in that column are indicative rather than conclusive.

5.1 The leaders swap places

The difference is stark. In web-search answers, The Purest Solutions is the clear leader (60.2%). But in the own-knowledge segment the same brand falls to 6.9% (just 4 of 58 answers), and Farmasi (44.8%) takes the top. This pattern is consistent with the Claude finding in Chapter 4: because Claude is the least search-reliant model, Farmasi naturally emerges as its leader in a memory-weighted world — indeed, Claude is the only model that makes Farmasi #1. With only 58 memory-based responses, the exact rates carry wide uncertainty; the direction of the effect is the takeaway, not the precise percentages.

5.2 “Legacy vs. new-generation” brands

This split offers a useful reading of how brands become visible:

• New-generation / digital-native brands (favored by search): Brands like The Purest Solutions have a strong footprint in current web content (e-commerce listings, beauty media, their own sites). When search is on, these brands jump forward. Their visibility is tied to the live content the models index at that moment.
• Legacy / memory brands (favored by memory): Established, high-awareness brands like Farmasi, Atelier Rebul and Eyüp Sabri Tuncer are strongly represented in models’ training data, so they are recalled even without search. Their visibility is less dependent on the web’s current state.

What this suggests: because web-grounded answers make up the majority (482/540), a current, well-structured web presence is closely associated with AI visibility. In the minority of scenarios where models fall back on memory, established brand awareness appears to carry more weight. This is an observed association in the data, not a proven cause-and-effect.

6. Category Ownership

One-sentence takeaway: There is no single “winner”: makeup discovery is led by Pastel/Flormar/Golden Rose, attribute/clean questions by The Purest Solutions, and skincare/haircare use-cases by the dermocosmetic brands (Dermoskin, Cosmed, Maruderm, Bioxcin).

Each category comprises 9 questions and 180 responses (9 questions × 4 models × 5 repeats). The table below shows the most visible 8 Turkish brands per category (rate = relative to the 180 responses in the category).

Reading: In the find category, makeup brands lead — Pastel appears in 65% of discovery questions. In attribute and use_case, The Purest Solutions takes the top; in use-cases specifically, the dermocosmetic brands (Dermoskin, Cosmed, Maruderm, Bioxcin) fill the top five.

6.1 Who is most visible for specific attributes and use-cases?

The table below drills into individual themed questions to identify the most visible Turkish brands within each niche (each theme = 18–20 responses, so individual cells carry meaningful uncertainty; rate = relative to that question’s responses).

Takeaways (these describe which brands AI surfaces most for each theme, not which products perform best):

• In vegan & cruelty-free attributes, Pastel and Note Cosmetics are the most visible; on the clean side, The Purest Solutions is very strong (in 94% of the cruelty-free question).
• When affordability comes up, Golden Rose appears in nearly every answer (100%) — the most surfaced reference point for value.
• The natural-ingredient niche surfaces an entirely different group: Bade Naturel, Homemade Aromaterapi and Rosece — a space where the big makeup houses are largely absent.
• Oily-hair questions are dominated by the Bioblas/Bioxcin/Otacı trio; haircare is a separate space where the overall leaders are weak.
• In skincare use-cases (acne, sensitive, mature skin), The Purest Solutions and the dermocosmetic brands share the table; for anti-aging specifically, Maruderm is most visible.

7. Open Market: Turkish vs. Foreign

One-sentence takeaway: In origin-neutral questions, total mentions split almost evenly between Turkish and foreign (49.1% to 50.9%); yet the four most visible brands are entirely Turkish, while foreign brands are more visible on the dermocosmetic shelf. This chapter rests on a small base (5 questions, 100 responses) and should be read as an early signal.

This chapter relies only on the 5 open questions with no origin restriction (100 responses) — the only question group where we can fairly compare Turkish and foreign brands. With 100 responses, the figures below are directional; the Turkish-vs-foreign balance is one of the headline numbers we will track most closely across future quarterly editions.

7.1 Origin share

The total is close to balanced; but this near-even split masks the individual visibility of brands.

7.2 Open-market top 15 brands (any origin)

7.3 The “Turkish brands lead makeup, foreign leads derma” pattern

The individual ranking reveals a pattern that the aggregate balance hides: the four most visible brands are all Turkish (Flormar, Pastel, Golden Rose, The Purest Solutions), and three of them are makeup brands. Turkish brands are clearly more visible in color-cosmetics/makeup discovery.

By contrast, foreign brands cluster on the dermocosmetic (pharmacy-skincare) shelf: La Roche-Posay, Bioderma, The Ordinary, CeraVe and Avène sit in slots 5–14 of the open market. So when models are asked “the best skincare brands I can buy in Turkey,” Turkish dermocosmetic brands (Dermoskin, Cosmed, etc.) appear less often than global names in an origin-unrestricted context.

The reading for brands: Turkish makeup brands hold a strong open-market visibility position; Turkish dermocosmetic brands are currently less visible than their global rivals in origin-neutral questions. Whether this gap widens or narrows is exactly the kind of movement a quarterly baseline is designed to track.

8. The Discovery Ecosystem: Where AI Learns About Brands

One-sentence takeaway: AI’s picture of cosmetic brands is shaped not by a single authority but by hundreds of small beauty sites, e-commerce/retail listings and brands’ own sites together — the source landscape is highly fragmented.

When an AI assistant answers with web search, it draws on the content available to it at that moment. In web-search answers, the source addresses returned by the models were collected, reduced to domains, and ranked by how many distinct answers cited them — a map of where these systems are reading about cosmetic brands. Note: the machine-readability of source data varies by model; this analysis is most complete for Gemini, ChatGPT and Grok, while Claude exposes fewer machine-readable sources (see Chapter 10).

8.1 The 9 most-cited domains

The three most-cited domains — oggusto.com, hopi.com.tr, dermoeczanem.com — are, respectively, a beauty-media outlet, a retail/loyalty platform and a dermocosmetic e-commerce site. This trio captures the character of the source landscape: editorial content + retail listings + vertical e-commerce.

8.2 Source-type mix

8.3 The fragmentation finding

The most striking finding is fragmentation: 46.1% of all citations come from a long tail of ~395 small sites, none of which is individually dominant. No single mega-source determines AI’s perception; instead, many small beauty sites, blogs and listing pages add up to form the picture. The next three large blocks — e-commerce/retail (15.9%), editorial beauty media (13.4%) and brand-owned sites (11.7%) — together make up roughly 41% of citations.

What this means for understanding brand discovery. The brands that are most visible tend to appear across all three legs of this ecosystem rather than relying on one: a presence across many Turkish beauty sites and editorial lists; listings on major retail/e-commerce platforms (Trendyol, Hepsiburada, Watsons, dermocosmetic e-commerce); and a structured, current website (thepurestsolutions.com, maruderm.com and dermoskin.com.tr all appear in the top 15). This is a description of an observed pattern, not advice on how to rank.

9. What the Patterns Suggest

One-sentence takeaway: Highly visible brands tend to share three things — broad model coverage, a strong and current web presence, and a clear category/attribute association.

This chapter is a neutral reading of the data. It describes patterns associated with visibility; it is not advice, a service, or a product recommendation, and visibility remains separate from quality.

1) Breadth is a baseline condition. All of the top 29 brands appear in all four models (breadth component 100). Beaulis (3 models) and Bionnex (2 models) at the bottom of the list fall to the lower ranks largely for this reason. Observation: appearing in only one or two models caps a brand’s total score; broad coverage across the major models is what the most visible brands share.

2) Mention and position are different things. Pastel has high mentions (38.7%) but a low position (avg. 4.0); Golden Rose and Bioxcin are mentioned less often but appear near the top when they do (position components 100 and 75.8). Two distinct questions sit behind a brand’s visibility: “Is it named often?” (mention) and “Is it named early?” (position). They behave independently.

3) Web presence tracks with the majority scenario. 482 of 540 answers relied on web search, and in that segment the most visible brands are those with current, structured web content (The Purest Solutions). Established awareness (Farmasi) leads only in the minority no-search scenarios. Observation: in this dataset, a current, machine-readable web presence is the factor most strongly associated with AI visibility.

4) Visibility is won in niches. No brand is most visible everywhere. There are clear gaps — natural ingredients (Bade Naturel, Homemade Aromaterapi), oily hair (Bioblas, Bioxcin) and anti-aging (Maruderm) — where the overall leaders are absent. Consistent visibility in a narrow niche can produce a higher position score than a broad but shallow presence.

5) The source footprint is multi-channel. The fragmented source landscape (long tail 46%) shows that no single channel drives visibility. The footprint of the most visible brands spreads across many small beauty sites + major retail listings + their own structured site.

How a brand can locate itself in the data. Using the dataset (available on request), a brand can read four things in order: (a) whether it appears in all four models; (b) how frequently it is mentioned; (c) how high it ranks when mentioned; and (d) which model or category it is weakest in. Together, these four readings show where a brand’s visibility gap sits — without saying anything about whether its products are good.

10. Limitations and Notes

One-sentence takeaway: This report is a single, one-market, point-in-time snapshot; the figures should be read within the limitations below.

• Single, point-in-time measurement. Data were collected in a single run on 24–25 May 2026; it is a snapshot of that moment, not a trend or time series. This is the first (baseline) measurement; subsequent quarterly editions will enable trend analysis.
• Small sub-samples carry real uncertainty. Several headline findings rest on small bases — the Turkish-vs-foreign comparison on 5 open questions (100 responses) and the web-search-vs-memory split on 58 own-knowledge responses. Treat these as directional signals to be confirmed across editions, not settled facts.
• Small score gaps are not meaningful. In the tightly-packed middle of the leaderboard, differences of a point or two should not be over-interpreted; only larger gaps (such as the leader’s 10.7-point margin) are robust.
• Visibility is not quality. This report measures only mention and position. It does not assess whether a brand was described positively or negatively, whether a recommendation was accurate, or whether a product is effective. A score is not an endorsement.
• Models are probabilistic. The same question can produce different answers; 5 repeats reduce this variability but do not eliminate it.
• Web search depends on momentary conditions. Search-based answers depend on the content indexed and served at the time of measurement; different sources may return at a different moment.
• “Web search vs. own knowledge” is a proxy. This distinction relies on whether a citation accompanied the answer; it is not definitive proof of the model’s internal process.
• Precision over recall. Brand matching was done cautiously to avoid false positives; brands mentioned indirectly or implicitly may be somewhat undercounted.
• Source coverage varies by model. Because some models share fewer machine-readable sources, the source analysis does not reflect the entirety of all answers.
• Origin classification involves judgment. Some brands’ Turkish/foreign label is debatable (e.g., brands manufactured in Turkey but of foreign origin, or Turkish-founded brands later acquired by foreign owners).
• Output-length limits differ by model. Models’ maximum output token limits are not the same; this can partly affect how many brands fit in an answer.
• Model versions date quickly. The findings are specific to the model versions listed in Chapter 2 as of late May 2026; provider updates may shift results in later editions.

11. Appendix & Data

11.1 All questions (27) and their categories

Distribution: 9 find, 9 attribute, 9 use_case; 22 constrained (Turkish-only) and 5 open (origin-unrestricted: Q4, Q6, Q11, Q12, Q18).

11.2 All tracked brands (138)

The table below lists all 138 brands tracked in the study (100 Turkish, 38 foreign) and marks the 31 brands that cleared the 5% mention threshold to qualify for ranking with a green tag. The list is ordered by mention count, descending.

11.3 Data availability

The study tracked 138 brands (100 Turkish, 38 foreign); of these, 31 Turkish brands cleared the 5% mention threshold to qualify for ranking. To support scrutiny and reproduction, the underlying data behind this report is available on request: the response-level brand mentions, the qualified-brand leaderboard, the full unfiltered brand metrics, the per-model and per-category breakdowns, the web-search-vs-memory and open-market breakdowns, the top source domains, and the methodology and data dictionary.

All figures reflect a single point-in-time run (24–25 May 2026) and are subject to the limitations in Chapter 10. This is the first (baseline) measurement; the study will be repeated quarterly.

11.4 About Herm.io & disclosure

Herm.io is a consumer-behavior and marketing-data company. We study how people discover and choose brands so that brands can reach the right customers. This report is part of our public research and is a recurring study, published quarterly.

Disclosure & neutrality. Herm.io does not sell SEO or GEO (search/AI-ranking) services, and this report does not recommend any. No brand paid to be included, ranked, or described in this report, and a brand’s score is not an endorsement or a judgment of its quality — it is a measure of visibility only. To keep the analysis objective, all citations to Herm.io’s own domains were excluded from the source data, so the company does not appear in, measure, or benefit from its own study (see Section 2.6). Brands that want to understand their position in the data are welcome to book a call for a neutral walkthrough of the findings; this is advisory and free.

*Report period: May 2026 · Edition 1 (baseline) of a quarterly series