This is a recurring, neutral study of a question nobody had measured: when someone in Turkey asks an AI assistant which pots, pans or dinner sets to buy, which brands does it name?
We ask five large language models the same Turkish-language questions real people type, repeat each one five times, and record how often, how early, and across how many assistants each brand appears. The result is an AI Visibility Score — a measure of findability, not of quality. A brand ranks highly here because these systems have information about it and surface it readily. That is all it means.
Why this series exists
Three facts sit next to each other, and the gap between them is the reason for this research.
1. This is a real industry with real numbers. Turkey’s home and kitchen goods sector registered US$3.05bn of exports in 2025 on the exporters’-association basis — down 3.5% in current dollars from US$3.16bn in 2024, and roughly 1.12% of Turkey’s total merchandise exports (EVSİD/İDDMİB, February 2026; TİM export totals, January 2026). The United Kingdom, Germany, Italy, France and Spain took about a third of it. Employment is put at approximately 90,000 people across some 4,700 exporting companies (TİM/İDDMİB, 2023 — the report does not state the figure’s reference year). Production is genuinely clustered: ceramics, porcelain and glass account for around 60% of Kütahya’s provincial exports (Zafer Development Agency, TR33 Regional Plan 2024–2028), and Sürmene Bıçağı has held a registered geographical indication since 2018.
One caution we will repeat every edition: that export basket includes small electrical kitchen appliances and certain cooking appliances. It is not a clean measure of pots, plates, glasses and cutlery.
2. Nobody measures the category on its own. This is the fact that makes the series worth publishing. In the Ministry of Trade’s E-Ticaretin Görünümü report of 12 May 2026, kitchenware has no line of its own — it is bundled into “Ev, Bahçe, Mobilya ve Dekorasyon,” which generated TRY 215.57bn in 2025, about 8.8% of retail e-commerce. That headline grew 50.0% year on year, but in an economy with high inflation that is a nominal figure and not real growth, and the Ministry publishes no online-penetration rate for the category at all. There is likewise no ready-made TÜİK customs aggregate for the narrow basket of glassware, porcelain, metal cookware and cutlery, no consolidated SGK employment total (the sector spans a dozen NACE classes), and — as our own research pass confirmed — no credible public ranking of the market’s brands by share. Thirteen of the fourteen best-known names publish no brand-level Turkish revenue at all.
So a Turkish kitchenware brand cannot look up its market share. It cannot look up its category’s online penetration. Until now it could not look up its AI rank either.
3. Turkey uses AI assistants heavily — and not in the way the rest of the world does. By GWI’s Q2 2025 wave, 39.7% of Turkish internet users had used ChatGPT in the previous month, 11th of 54 markets against a surveyed-market average near 26.5% (DataReportal, Digital 2026, October 2025). Turkey’s own statistics office puts generative-AI use lower, at 19.2% of recent internet users aged 16–74 (TÜİK, October 2025) — a different population asked a different question, which is why the two figures differ and must not be averaged.
The distinctive Turkish fact, though, is which assistant. On Statcounter’s July 2026 referral data, ChatGPT accounts for 76.31% of Turkey’s AI-originated web referrals against 77.92% worldwide — slightly below the global average — while Gemini takes 18.18% in Turkey against 9.90% worldwide, nearly double-weighted.
That matters for this study more than it looks. Gemini is the assistant whose behaviour diverges most in our data: it searched the web on only 19.6% of its answers, named no marketplace at all, and produced a visibly different brand list from the models that retrieve. The assistant Turkey over-uses is the one that answers most from memory.
A correction to a figure this series published in July 2026
The gap. A substantial manufacturing sector. A shopping occasion — the çeyiz, the trousseau — that furnishes an entire household at once. A population that has adopted AI assistants quickly and leans unusually on the one that answers from memory. And no public measurement of which kitchen brands those assistants actually recommend. That is what this series publishes.
What the first edition found
The August 2026 baseline produced five findings worth reading in full.
| Finding | The number |
|---|---|
| Karaca leads the general ranking with the first perfect score in this series — it simultaneously has the highest mention rate, the earliest average position, and coverage by all five assistants. | 100.00 · named in 57.2% of 1,125 responses |
| The top three entities are one company. Karaca, Emsan and Jumbo are all Karaca Group brands — and only three of the group's thirteen-plus consumer brands appear at all. | 727 of 1,125 responses · 64.6% |
| Product segments have different leaders. Korkmaz leads cookware and whole-kitchen questions; Karaca leads tableware and cutlery. | 4 segments · 2 leaders |
| On questions that do not ask for Turkish brands, the shelf is Turkish-led but genuinely mixed — and more distinct foreign brands appear than Turkish ones, each named rarely. | 63.6% Turkish / 36.4% foreign · 54 Turkish vs 58 foreign entities |
| A quarter of all answers name a brand from outside the category — appliances and cabinetry — because mutfak means both 'kitchen' and 'kitchenware'. | 25.8% of responses · 51.3% on broad questions |
Headline findings from the August 2026 baseline edition. Full method, all ten ranked cuts and every limitation are in the edition itself.
→ Read the August 2026 edition
What a high score does not mean
A high AI Visibility Score means one thing: these systems currently have information about a brand and surface it readily when asked. It is not a statement that the brand is better made, better value, safer, more popular, better selling, or larger in market share.
This is worth more than a disclaimer here, because in this category the comparison cannot even be attempted.
- Quality is not measured. We record whether a brand is named and how early. We do not assess whether the recommendation was accurate, whether the description was positive, or whether the product performs. There is no sentiment layer and no product testing.
- Sales are not measured, by us or by anyone. The peer-reviewed evidence on AI-referred commerce is thin and cautious. The strongest study we found — 973 e-commerce sites over twelve months, published in Marketing Science in April 2026 — observed real transactions from AI referrals but found they converted better than paid social and worse than most traditional channels, on a very small traffic share. No published study estimates the incremental sales caused by a specific brand being named in an AI answer.
- Market share cannot be compared, because it does not exist in public. Our research pass looked for a credible public brand-share ranking of Turkish kitchenware or tableware with a disclosed market definition, brand-level shares, a stated measurement period and a reproducible method. It found none — no official series, no peer-reviewed study, no publicly inspectable commercial ranking. Nor is company revenue a workable substitute: of the fourteen most prominent names, only Kütahya Porselen publishes audited issuer-level accounts (TRY 4.22bn consolidated in 2025, about 2.0% lower in real terms than 2024). Karaca has only an approximate group-level figure reported externally as approaching US$1bn globally; Paşabahçe is visible only inside Şişecam’s glassware segment; Tefal only inside Groupe SEB’s €8.17bn worldwide total. Store counts are worse — sources variously count owned stores, franchises, dealers, shop-in-shops and sales points, and cannot be lined up.
So when a brand ranks first here, the honest reading is narrow and specific: it is the most findable, not the most bought. Claims of market leadership made by any company in this category should be treated as vendor claims unless they come with a named research provider, a market definition, a period and a percentage.
What we measure, and how
- Five models, queried directly. Claude Haiku 4.5, Gemini 3.6 Flash, GPT-5.6 Luna Pro, Perplexity Sonar and Grok 4.3 — via API, no system prompt, provider-default temperature, no locale set. The questions are in Turkish; the models infer the market from language alone.
- Real questions, repeated, and never rewritten. Each edition uses Turkish-language questions of the kind people actually type, asked five times each to capture run-to-run variation. Ambiguous questions are kept as asked — rewriting a question because its answers are inconvenient is selection on outcome.
- Entities discovered from the answers, then reviewed by hand. Candidates are extracted from the responses themselves, passed through a human review gate, and researched externally before any metric is computed. Origin is assigned by trademark ownership, so a foreign mark made or sold in Turkey under licence stays foreign.
- A three-component score. AI Visibility Score = 0.45 × Mention + 0.30 × Position + 0.25 × Breadth — how often a brand is named, how early it appears, and how many of the five assistants know it.
- Ten cuts, each on its own published scale. Every edition is read across product segments, origin scope (questions that demand Turkish brands versus questions that don’t), and behavioural types (discovery / attribute / use-case). Each cut has its own denominator and its own scale, fixed once and never recomputed. Turkish-versus-foreign comparisons are made only on origin-neutral questions — the only fair basis.
What we do not measure. Quality, accuracy, sentiment, price, durability or safety. A score is not an endorsement.
Editions
| Edition | Published | Scale | Headline |
|---|---|---|---|
| August 2026 — baseline | 3 August 2026 | 1,125 responses · 45 questions · 5 models · 144 entities tracked | Karaca leads at a perfect 100.00 — but Karaca, Emsan and Jumbo are one company covering 64.6% of all answers. Cookware and tableware have different leaders, and a quarter of responses name a brand from outside the category. |
Published editions. The study is repeated quarterly; trend analysis becomes possible from the second edition.
The wider context: is AI discovery real yet?
We think the honest answer is “real, measurable, growing from a small base — and smaller than the vendors say.” The evidence, weighed:
AI answers demonstrably suppress traditional clicks. The strongest independent measurement remains Pew Research Center, which observed 68,879 Google searches by 900 US adults in March 2025: when an AI summary appeared, users clicked a traditional result in only 8% of visits against 15% without one, and clicked a source inside the summary just 1% of the time. A July 2026 preprint using Comscore desktop clickstream data from 45,386 households found an outbound visit in only 5.2% of ChatGPT conversation sessions, against 31.1% of Google search sessions. Much of the time, the assistant answers and the click never happens.
Referral traffic is growing fast, but the conversion story is vendor-owned and it has moved. Adobe reported AI-referred US retail traffic up 138% year on year to May 2026, converting 54% better than non-AI traffic. That is worth reading against Adobe’s own earlier data: in February 2025 the same channel converted 9% worse than other traffic, and in July 2024, 43% worse. Similarweb separately estimated ChatGPT-referred retail visits converting at 11.4% against 5.3% for organic search. Adobe and Similarweb both sell analytics and AI-visibility products. The direction is credible; the magnitudes are not settled, and the base is small and undisclosed.
AI’s influence is larger than its clicks — probably. A June 2026 working paper matching opt-in users’ ChatGPT, Claude and Gemini conversations to their browsing found that when an assistant recommended an unfamiliar brand, the probability of subsequently Google-searching that brand rose 4.3 percentage points and of visiting the brand’s own site 2.4 points. Referral statistics therefore understate AI’s effect on which brands people go looking for. (Not peer-reviewed; the authors are affiliated with a company selling AI-visibility products.)
But the assistants are not yet good shopping advisers. Two 2025–26 benchmarks are sobering: one scored leading models at 11.22% and 3.92% on a composite shopping benchmark of 120 expert-curated tasks, documenting reliance on promotional misinformation; another found pass rates of 57–77% across 525 shopping missions, weakest on multi-turn dialogue and optional requirements.
And kitchenware’s position is genuinely unclear. Adobe’s data suggests home goods attract roughly three times apparel’s share of AI-originated traffic, consistent with the idea that specification-rich categories — coating, induction compatibility, dimensions, dishwasher safety — suit an assistant better than taste-driven ones. But the same data placed home goods among the weaker-converting AI-referred categories, and no behavioural study isolates kitchenware. There is no public evidence yet that AI kitchenware recommendations are more accurate or better-converting than any other category.
Which is precisely why a baseline matters now. The measurement is most valuable before the behaviour becomes mainstream. If AI-assisted kitchen shopping in Turkey grows the way general AI shopping has, the brands visible when it happens will not be the ones that start optimising afterwards.
What the research actually says about how models pick brands
We are careful not to sell certainty here, because the academic literature does not supply it.
There is no established model of how commercial assistants decide which brands to name. What exists is evidence for individual mechanisms, tested separately and on different models. Brand-origin bias is real but not simple: an EMNLP 2024 study found global brands disproportionately associated with positive attributes, while a FAccT 2026 study found geographic preference varies by model and that stating the user’s location moves rankings more than stating a brand’s origin — so there is no universal foreign-over-local rule. Position matters, conditionally: the well-known “lost in the middle” finding was qualified by EMNLP 2025 work showing that in realistic retrieval pipelines, sophisticated reordering did not beat random shuffling. Content optimisation is weaker than it is sold as: the founding GEO paper (KDD 2024) reported 30–40% visibility gains, but supplied sources to the model after retrieval rather than testing whether an edited page gets retrieved from the live web at all — and a NeurIPS 2025 benchmark found most conversational-SEO rewrites ineffective or negative, with gains shrinking as more competitors adopt them.
One gap is directly relevant to our own method. Our August edition found that grounded and ungrounded answers produce different brand lists, but declined to call it a retrieval effect, because the model chose when to search and chose by question type. We looked for anyone who had solved that identification problem. The closest is a vendor-published June 2026 study reporting that 80.2% of ChatGPT’s product recommendations changed when search was toggled — but it does not disclose the model version, interface or collection dates, so it cannot show that retrieval was the only thing that changed. No peer-reviewed study meeting the full standard was found.
What none of this establishes is which mechanism dominates in a live shopping answer, or that more AI mentions cause more sales — for which there is almost no independent evidence at all.
So this series does the one thing that can be done rigorously: measure the output. We do not claim to explain why a model names a brand. We record that it does, how often, and how early — and we publish the method so anyone can check us.
Data, caveats and sources
- Point-in-time. Each edition is a snapshot from a single collection window. Model versions and web indexes shift; trend claims become possible only from the second edition onward.
- Visibility is not quality, sales or market share. Repeated because it matters, and because in this category none of the three can be checked against public data.
- Market figures in this hub carry different confidence levels. Export, employment, e-commerce, marriage and demographic figures are official or official-derived (EVSİD/İDDMİB and TİM registrations, ETBİS/Ministry of Trade, TÜİK, SGK, development agencies). Sector size, çeyiz spending, AI conversion and referral statistics are commercial, industry-association or vendor-published estimates and are labelled as such where used. Turkish-lira growth rates should be read cautiously: in a high-inflation economy, nominal growth is not real growth.
- Different bases are not interchangeable. Exporters’-association registrations and TÜİK customs figures measure different things. The broad züccaciye sector figure of US$5.56bn for 2024 includes industrial kitchens, electrical goods and decorative products, and is not comparable with the narrower home-and-kitchen basket. We give both bases rather than picking one.
- Some things are simply not public, and we say so. There is no representative estimate of average çeyiz spending or of the kitchenware share within it — the most-cited figures are constructed editorial baskets, and two pages of the same source disagree with each other. There is no published statistic on how many couples buy a complete set from one brand. There is no monthly series isolating wedding-driven kitchenware sales. We flag these gaps rather than filling them with estimates.
- The çeyiz occasion is real, but “one brand for the whole kitchen” is not established. Peer-reviewed Turkish research shows an omnichannel pre-marriage research process — 70.6% research before buying, 52.5% inspect in store, 43.2% use the internet, 36.9% ask family — while industry testimony describes a shift away from fixed complete sets toward selective, product-by-product completion. When one brand appears in 84% of AI answers to broad trousseau questions, that is a fact about the assistants, not a measurement of what Turkish households buy.
Want the underlying data?
Frequently asked questions
Karaca leads the general ranking with a perfect score of 100.00, appearing in 57.2% of all 1,125 responses and named by all five assistants. Korkmaz (80.90) is second, followed by Kütahya Porselen (70.95), Porland (70.37) and Emsan (70.11). Leadership changes by segment: Korkmaz leads cookware and whole-kitchen questions, Karaca leads tableware and cutlery.
Not all of them. Karaca (#1), Emsan (#5) and Jumbo (#6) are all Karaca Group brands. Together they appear in 727 of 1,125 responses — 64.6% of the study — rising to 72.0% on origin-neutral questions.
It depends entirely on the segment. On origin-neutral questions, mentions split 63.6% Turkish to 36.4% foreign, and Tefal places second overall. But tableware and serving has just one foreign qualifier out of 21, while cutlery and storage has IKEA at #2 and Zwilling at #4.
83.3% of answers were web-grounded. Of the citations, 44.2% point to brands’ own websites, led by karaca.com, cited by 34.2% of all responses and by all five models. Editorial and community sources supply a further 17.9%.
No. The score measures only how often and how prominently a brand is named across the assistants. It does not assess quality, price, accuracy or sentiment, so a score is a measure of visibility, not an endorsement.
About Herm.io & disclosure
Herm.io is a consumer behaviour and marketing data company. We study how people discover and choose brands to help businesses reach the right customers. This report forms part of our public research and is conducted quarterly.
Disclosure & neutrality
The rankings and metrics in this report are entirely impartial and are based on the methodology described within the report. No brand can pay or provide sponsorship to be included in the report, improve its ranking or influence how it is described in the content.
A brand’s score in the report is not an endorsement or assessment of quality; it is solely a measure of its current visibility within AI models. No service or product offered by Herm.io is sold for the purpose of directly changing the results of this report or guaranteeing a ranking. To preserve objectivity, all references to Herm.io’s own domain are removed from the source data before analysis. This ensures that the company does not appear in, measure itself through or benefit from its own study.
Brands seeking to understand their position in the data can schedule a consultation for an impartial assessment of the findings. The consultation is advisory and free of charge.
Quarterly series · Baseline edition published 3 August 2026
Written by
Mert Can Elkaya
Contributor
I'm a product builder working at the intersection of product, fintech, and growth. From martech and venture capital to leading product at a proptech platform and co-founding a fintech startup, I help teams—and shoppers—make smarter, more confident decisions.
More from Mert