750 responses, 5 models, two markets: the first AI-visibility snapshot of Turkey's travel sector

Vertical: Travel — tour agencies & car rental — Turkey
Method: A single point-in-time study of 750 responses — five large language models (GPT-4o-mini, Gemini, Claude, Perplexity, Grok), each asked 30 Turkish-language questions five times
Collection date: 12 June 2026

Key terms: AI visibility, AI Visibility Score, travel brands, Turkish tour agencies, car rental Turkey, Turkish (yerli) travel brands, LLM brand recommendations, ChatGPT/Gemini/Claude/Perplexity/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 five large language models name each travel brand. Because tour agencies and car rental are distinct markets, they are ranked separately. 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 travel 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, cheaper or more reliable. 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 service quality, price or trustworthiness. Readers and travellers 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 travel sector splits into two very different markets — tour agencies, surfaced almost entirely by Turkish brands and led by Jolly Tur, and car rental, led by international chains with Garenta the strongest Turkish name.

• Jolly Tur leads tour agencies; Avis leads car rental. Jolly Tur appears in 70% of all tour-agency responses across all five models, topping that market with a 93.2 score and a clear 12.8-point gap to Setur (80.4). In car rental, Avis tops the table at 93.2 (51% of responses).
• The two markets look opposite on origin. The tour-agency top five — Jolly Tur, Setur, ETS Tur, Prontotour, MNG Turizm — are all Turkish. The car-rental top five is led by international chains (Avis, Sixt, then Garenta as the top Turkish brand at #3, Enterprise, Budget).
• 52 brands qualified for ranking. Of 177 tracked brands, 21 tour agencies and 31 car-rental firms cleared the 5% mention threshold within their market. Average scores: 49.1 (tour) and 52.1 (car rental).
• The models see the same markets at different breadths. Grok has the widest repertoire (9.5 distinct brands per answer); GPT-4o-mini and Claude Haiku are the narrowest (4.2 each). Unlike some markets, the top of each leaderboard is broadly agreed across models.
• Search was near-universal. About 98% of all answers were web-grounded (every model except Gemini searched on ~100% of answers; Gemini on ~91%). With almost no “memory-only” answers, this edition does not include a web-search-vs-memory comparison; see Chapter 5.
• The open market skews strongly Turkish. Across the 20 origin-unrestricted questions (500 responses), mentions split 81.4% Turkish / 18.6% foreign. Foreign visibility is concentrated almost entirely on the car-rental shelf (Avis, Sixt, Budget, Enterprise); tour agencies remain overwhelmingly Turkish even when origin is not restricted.
• Sources are brand-owned, not fragmented. Nearly half — 48% — of cited sources are brands’ own websites (jollytur.com, setur.com.tr, avis.com.tr lead), a very different picture from fragmented consumer categories. OTAs/aggregators and a long tail of editorial and forum content make up the rest.

Why it matters. Travellers are shifting discovery from search engines to AI assistants. Whether a tour operator or car-rental firm 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 travel sector. A reminder: the numbers below describe visibility and information availability — not which brands are best.

2. Methodology

One-sentence takeaway: Five large language models were asked 30 Turkish questions five times each — with no system prompt — to produce 750 responses; brands were identified by automated extraction plus human validation and ranked with a three-component (45/30/25) score, computed separately for the two markets.

2.1 Scope and models

The study queried five large language models directly via API with identical questions. 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.

Scale: 30 questions × 5 repeats × 5 models = 750 responses, all completed. Single run, 12 June 2026. The questions cover two markets: tour agencies (18 questions → 450 responses) and car rental (12 questions → 300 responses). Final brand dictionary: 177 brands.

2.2 Web-search configuration

Web search was offered as an enabled tool to all five models; each decided for itself whether to use it (Perplexity’s Sonar is search-grounded by design). Whether a response drew on web search or the model’s own knowledge was inferred by whether the provider returned any citation/source: a citation means “web-search,” none means “own knowledge.” This is a reasonable proxy, not direct proof of the model’s internal process. In this run ~98% of answers were grounded (see Chapter 5).

2.3 Questions

The 30 questions were generated in a two-step process and then split across two markets and three behavioural categories (discovery / attribute / use_case):

1. Seeding. We first asked LLMs to generate the questions and keywords people in Turkey might plausibly ask an assistant about booking tours and renting cars.
2. Validation against real demand. We took those seeds to Google, YouTube and Ekşi Sözlük to find related real-world queries, and used Google’s “People also ask,” alsoasked.com and answerthepublic.com for additional question and keyword variants. We then merged the lists and selected the top 15 per market by occurrence.

Of the 30, 10 are “constrained” (they explicitly ask for Turkish / yerli providers) and 20 are “open” (no origin restriction). The Turkish-vs-foreign comparison in Chapter 7 rests solely on the 20 open questions — the only fair basis for comparing Turkish and foreign brands.

2.4 Brand extraction and validation

Brands were extracted in two stages. (1) Automated extraction: emphasised brand names and citation domains were collected, and spelling/format variants merged into a single canonical name. (2) Human validation: the candidate list was reviewed by hand — real brands confirmed, non-brands (category words, generic phrases, vehicle models, places) filtered out, aliases merged, and each brand labelled Turkish/foreign. Final dictionary: 177 brands. Per study scope, aggregators and online travel agencies (Enuygun, Yolcu360, Tatilsepeti, Tatilbudur, Gruppal) are included as brands.

Origin rule. A brand is Turkish if founded and headquartered in Turkey; foreign if of foreign origin, even when widely operated in Turkey (e.g. Avis, Sixt, Budget and Enterprise operate locally but are foreign-origin chains).

Definitions.

• Mention: a validated brand (or alias) appears in the answer text — counted once per response (presence, not frequency), matched word-boundary aware, case- and diacritic-insensitive, longest-alias-first so a shorter name never double-counts inside a longer one.
• Position (MRR): mean reciprocal of the brand’s first-appearance rank within each answer (first = 1.0; second = 0.5; …).
• Breadth: number of the 5 models (0–5) that mention the brand within that market.

2.5 AI Visibility Score (0–100)

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

Each component is scaled so the leading qualified brand in each market = 100, then combined with the weights. Qualification (≥5% rule): only brands mentioned in at least 5% of their market’s responses are ranked (21 tour, 31 car rental). This threshold limits the risk that a rare brand which happens to appear first in a few answers inflates the position component. Unfiltered metrics for all 177 brands appear in Chapter 11.

2.6 Neutrality and self-exclusion

This is a market-wide, neutral study with no focus brand. To prevent any conflict of interest, all citations to Herm.io’s own domains were excluded from the source data before analysis, so the company’s own content neither appears among the most-cited domains nor influences any figure.

3. Overall Visibility Leaderboards

One-sentence takeaway: Each market has a clear front-runner with a wide gap to second — Jolly Tur (93.2) among tour agencies, Avis (93.2) in car rental — but the two markets differ sharply on origin.

As a reminder, these rank visibility, not service quality, price or reliability.

3.1 Tour agencies — 21 qualified brands

Average score of the 21 qualified tour agencies: 49.1.

Reading. Jolly Tur maxes out the mention and breadth components and tops the table at 93.2 — it appears in 70% of tour-agency answers, more than two in three. Setur (80.4), ETS Tur (76.8), Prontotour (75.3) and MNG Turizm (63.6) complete a top five that is entirely Turkish. A handful of brands clear the 5% bar mainly on position and breadth rather than frequency (e.g. Jetset, Vizesiz Rotalar at ~5–6% mentions); their ranks should be read cautiously (see Chapter 10).

3.2 Car rental — 31 qualified brands

Average score of the 31 qualified car-rental brands: 52.1.

Reading. The car-rental market is led by international chains: Avis (93.2), Sixt (77.6), then Garenta (76.7) as the most visible Turkish brand at #3, followed by Enterprise (73.2) and Budget (70.8). Turkish firms hold a strong mid-field — Garenta, Enuygun, Yolcu360, Rent Go, Ziraat Filo — but the very top is global. This is the mirror image of the tour-agency market, where Turkish brands own the top outright.

4. Differences Between Models

One-sentence takeaway: All five models broadly agree at the top of each market, but they differ markedly in breadth — Grok lists nearly 9.5 brands per answer, GPT-4o-mini and Claude Haiku closer to 4.

4.1 Per-model behaviour summary

Web search was near-universal: four of the five models searched on essentially every answer; Gemini on ~91%. The clearest difference is breadth: Grok lists ~9.5 brands per answer, while GPT-4o-mini and Claude Haiku name ~4.2 — a more selective shelf. Perplexity (5.9) and Gemini (6.1) sit in between.

4.2 Each model’s most-named brands

The tables below give each model’s six most-mentioned brands across its 150 responses (rate = % of that model’s answers naming the brand). Jolly Tur leads in every model — a notable consensus at the top.

The practical reading: which brands appear is fairly consistent at the very top, but how many brands an answer contains depends heavily on the model. A long-tail brand’s chance of being named is much higher in Grok than in the more selective models.

5. Search vs. Memory: Why This Edition Omits the Split

One-sentence takeaway: Almost every answer in this run used web search, so there is no meaningful “memory-only” segment to compare against — itself a finding.

In the cosmetics edition, answers split into web-grounded and memory-only segments that produced different leaders. That split does not exist in this travel run. About 98% of the 750 answers returned at least one citation or search result; the “own-knowledge” segment is roughly a dozen responses — far too small to support a comparison. Perplexity is search-grounded by design, and GPT-4o-mini, Claude Haiku and Grok each searched on ~100% of answers, with Gemini at ~91%.

The takeaway is that, for Turkish travel questions, these assistants reach for live web content almost every time. That makes a current, well-structured web presence especially closely tied to visibility here (see Chapter 8). We will keep measuring the grounding rate each edition; if a memory-only segment grows, the comparison will return.

6. Category Ownership

One-sentence takeaway: Within each market the leaders are largely stable across question types, but use-case questions reshuffle the order — Garenta, for instance, leads car-rental use-cases even though Avis leads overall.

Each market’s questions are split across three behavioural types: discovery (general “best/recommended”), attribute (specific qualities — affordable, cancellation-friendly, visa-free, electric fleet) and use_case (need-driven — first trip abroad, honeymoon, family car, long-term rental). The tables show the most visible brands per type (rate = relative to that type’s responses within the market).

Tour agencies

Car rental

Reading. Tour agencies are remarkably stable — Jolly Tur tops all three types. Car rental shifts more: the international chains dominate discovery and attribute questions, but Garenta leads use-case questions (family trips, long-term, out-of-town rentals), where a strong local-operations story surfaces Turkish brands more often.

7. Open Market: Turkish vs. Foreign

One-sentence takeaway: Across origin-neutral questions, 81.4% of mentions go to Turkish brands and 18.6% to foreign — but that headline hides a split market: tour agencies are almost entirely Turkish, while foreign brands cluster on the car-rental shelf.

This chapter relies only on the 20 open questions with no origin restriction (500 responses) — the only fair basis for comparing Turkish and foreign brands.

7.1 Origin share

The Turkish skew is much stronger than in some consumer categories — and it is driven by the tour-agency market, where Turkish operators dominate even without any origin restriction.

7.2 Open-market top 15 brands (any origin)

7.3 The pattern: Turkish leads tours, foreign competes in cars

The individual ranking reveals the structure the aggregate hides. The most visible open-market brands are Turkish tour agencies — Jolly Tur, ETS Tur, Prontotour, Setur, MNG Turizm. The only foreign brands in the top tier are car-rental chains — Avis, Sixt, Budget, Enterprise — which sit just below the Turkish tour operators. So when a traveller asks an origin-neutral question, the answer is dominated by Turkish tour brands, with foreign visibility almost entirely confined to car rental. Whether that car-rental balance shifts is exactly the kind of movement a quarterly baseline is built to track.

8. The Discovery Ecosystem: Where AI Learns About Brands

One-sentence takeaway: Unlike fragmented consumer categories, AI’s picture of Turkish travel brands rests heavily on brands’ own websites — nearly half of all cited sources — alongside OTAs and a long tail of editorial and forum content.

When an assistant answers with web search, it draws on the content available to it at that moment. We collected the source addresses returned by the models, reduced them to domains, and ranked them by how many distinct answers cited each — a map of where these systems read about travel brands. Across the run, 484 distinct domains were cited.

8.1 The 9 most-cited domains

The most-cited domains are overwhelmingly brand-owned — jollytur.com, setur.com.tr, avis.com.tr, prontotour.com, mngturizm.com — interleaved with editorial listicles (onedio.com, oteltavsiyeleri.com), the OTA enuygun.com, and the forum ekşisözlük.

8.2 Source-type mix

8.3 The brand-owned finding

The most striking pattern is concentration on official sites: 48% of citations point to brands’ own domains, with another 8.7% to OTAs/aggregators. This is very different from a fragmented long tail — here, when an assistant looks something up, it most often lands on the brand’s own website or a major travel platform. The practical reading (a description, not advice): in this sector, a current, machine-readable brand site appears closely associated with AI visibility, more so than in categories where editorial and retail content dominate.

9. What the Patterns Suggest

One-sentence takeaway: Highly visible travel brands tend to share three things — coverage across all five models, a strong own-site web presence, and clear leadership within their own market and question type.

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

1) Breadth is a baseline condition. The top brands in both markets appear in all five models. Brands seen by only one or two models sit lower largely for this reason — broad model coverage is what the most visible brands share.

2) Mention and position are different things. Some brands are named often but late; others are named rarely but early. A few clear the 5% bar mainly on position and breadth (Jetset, Vizesiz Rotalar in tour); “named often” and “named early” are independent signals.

3) Web presence tracks with the dominant scenario. With ~98% of answers grounded in web search, and brand-owned sites the single largest source type, a current and structured web presence is the factor most closely associated with visibility in this sector.

4) The two markets are genuinely different. Tour-agency visibility is Turkish-dominated and stable across question types; car rental is led by international chains and reshuffles on use-case questions. A brand’s context — which market, which question type — matters as much as its overall rank.

5) Visibility is split by market, not won everywhere. No brand leads both markets. Within each, niches (use-case questions, attribute questions) surface different brands than the discovery headline.

How a brand can locate itself in the data. Using the dataset (available on request), a brand can read four things in order: whether it appears in all five models; how often it is mentioned within its market; how early it is named when mentioned; and which model or question type it is weakest in. These readings show where a visibility gap sits — without saying anything about whether its service is good.

10. Limitations and Notes

One-sentence takeaway: This is a single, point-in-time snapshot of two Turkish travel markets; read the figures within the limitations below.

• Single, point-in-time measurement. Collected in one run on 12 June 2026 — a snapshot, not a trend. This is the first (baseline) edition; quarterly editions will enable trend analysis.
• Small-sample position inflation. A few brands clear the 5% bar but rank high mainly on position and breadth despite thin mentions (e.g. Jetset, Vizesiz Rotalar in the tour market). Read their ranks cautiously.
• Small score gaps are not meaningful. In the packed middle of each leaderboard, differences of a point or two should not be over-interpreted; only larger gaps (such as each leader’s wide margin) are robust.
• Visibility is not quality. This report measures only mention and position. It does not assess whether a brand was described positively, whether a recommendation was accurate, or whether a service is good. A score is not an endorsement.
• Models are probabilistic. The same question can produce different answers; five repeats reduce but do not eliminate this.
• Web search depends on momentary conditions — the content indexed and served at measurement time. “Web search vs. own knowledge” is a proxy based on whether a citation accompanied the answer.
• Precision over recall. Brand matching was cautious to avoid false positives; brands mentioned indirectly may be undercounted.
• Source-type mix is heuristic. Domain types were auto-classified (brand-owned detected by matching the domain to a brand); treat the mix as indicative. Source coverage also varies by model.
• Origin classification involves judgment. Some operators are foreign-owned but run locally in Turkey (e.g. Avis, Budget); the Turkish/foreign label is debatable in such cases.
• Aggregators/OTAs are included as brands per study scope; they compete differently from direct providers.
• Output-length limits and model versions differ and date quickly; findings are specific to the versions in Chapter 2 as of June 2026.

11. Appendix & Data

11.1 All questions (30) and their categories

Distribution: 10 discovery, 10 attribute, 10 use_case; 10 constrained (Turkish-only) and 20 open (origin-unrestricted).

11.2 All tracked tour-agency brands

11.3 All tracked car-rental brands

11.4 Data availability

The study tracked 177 brands; of these, 52 cleared the 5% threshold within their market (21 tour agencies, 31 car-rental firms). To support scrutiny and reproduction, the underlying data is available on request: response-level brand mentions, both qualified leaderboards, full unfiltered brand metrics, per-model and per-category breakdowns, the open-market breakdown, the top source domains, and the methodology and data dictionary.

All figures reflect a single point-in-time run (12 June 2026) and are subject to the limitations in Chapter 10. This is the first (baseline) measurement; the study will be repeated quarterly.

11.5 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 quarterly study.

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, 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. Brands that want to understand their position in the data are welcome to book a call for a neutral walkthrough; this is advisory and free.