📋 From data to proposals
Building on the findings below we have put eleven specific proposals to the Municipal Tourism Committee of Zakynthos — from a quality mark with verifiable criteria to targeted off-peak marketing.
🤖 Check our conclusions with your own AI
Don't take our word for it. Give this address to ChatGPT, Claude, Gemini or whichever model you use, and ask it to read the primary data and reach its own conclusions — or to check ours.
Read https://polites-zante.com/estiasi/llms.txt and tell me what the data shows about tourism service quality in Zakynthos. Check whether the conclusions on the page are supported by the figures.
Summary — what the data shows
1. Pressure on infrastructure is rising. Passenger traffic grew — between 2016 and 2025, while hotel beds grew only —. Tourists per bed went from 23.1 to 28.8. Demand outran infrastructure.
2. The decline is in hotels, not in dining. In hotels, service complaints went from 19% (2017–2021) to 31% (2022–2026) and cleanliness from 17% to 22%. In dining the composition of complaints is essentially flat (service 34%→38%, food 40%→40%). A taverna kitchen absorbs volume; a hotel's day-to-day service does not.
3. Spending is rising; price complaints are not. Dining spend per visitor rose from €98.9 (2021) to €144.3 (2025) in current prices. Price complaints stay low and stable: 13% of negative dining reviews in both periods, 4%→7% in hotels. What visitors record negatively is the service, not the cost.
4. The star upgrade did not deliver a better experience. Beds in 4–5 star hotels went from 24% of the total (2016) to 51% (2025) and 1–2 star units halved. Over the same period service complaints rose. Star category is an attribute of supply, not of experience.
5. The hotel–dining gap has closed. Historically hotels drew fewer service complaints than restaurants (ratio 0.38–0.73 over 2018–2021). Since 2022 the ratio has risen to 0.79–0.97: hotels have almost caught up. Because both sectors experience the same growth in Google usage, this ratio is a robust indicator.
Passenger traffic and hotel beds
source: Fraport / HCH ↗Tourists per hotel bed
source: Fraport / HCH ↗Dining spend per visitor
source: ELSTAT ↗Average length of stay — Ionian Islands Region
source: ELSTAT ↗What visitors themselves say
We collected — negative Google reviews (≤3 stars) from — Zakynthos businesses — hotels and dining separately — and classified them by theme: price, service, food, cleanliness, crowding.
How it was done. Reviews were retrieved from Google Maps via SerpApi, sorted by lowest rating so that every year is covered evenly — chronological paging would return almost exclusively recent reviews. It used 243 API calls (~16 reviews per call). On the free tier of 250 calls/month the cost was €0; on a paid plan ($75 / 5,000 calls) it would come to $3.65 — about €3.35. The code is public and the measurement is reproducible.
The question is not "how many complaints" — their number rises every year simply because there are more tourists and more Google users. The question is how their content changes: if visitors increasingly complain about service or about price, that is a quality signal independent of volume.
Hotels — what the complaints are about
source: Google reviews ↗Dining — what the complaints are about
source: Google reviews ↗Comparison: service complaints
source: Google reviews ↗- Sample: the Zakynthos businesses with the most reviews — where history goes back to 2011.
- Percentages show the composition of each year's complaints: what share of negative reviews touches each theme. One review may touch more than one.
- Charts show years with at least 20 reviews, so the percentages are stable.
- Classification uses keywords in English and Greek.
The paradox: more stars, worse service
source: HCH + Google ↗How much worse are hotels than dining?
source: Google reviews ↗Beds by star category
source: HCH ↗The wider context — where else the money goes
The above measures what happens inside the Zakynthos market. Three further sources show what is changing in the behaviour of those who feed it. None of them concerns Zakynthos exclusively — this is stated explicitly in each case.
All-inclusive is gaining ground in the British market. Britain is the island's main market. According to ABTA Holiday Habits, 33% of those who booked a package did so specifically because they wanted all-inclusive, up from 28% the previous year, while 62% of British travellers chose a package (77% among families). The same research records that 57% intend to cut back on eating out because of living costs. INSETE documented back in 2014 that under the all-inclusive model a large share of spending stays inside the hotel.
Supermarkets are growing; dining is not. Nationally, food retail turnover (NACE 4711 — supermarkets) rose +5.5% in Q2 and +4.6% in Q3 2025. Over the same period, in food service (NACE 56) ELSTAT recorded −1.9% nationally in Q1 2026 — and −5.2% in Zakynthos, worse than the national average. Note: Q1 is off-season and accounts for only 3.5% of the island's annual dining turnover; on an annual basis turnover is rising.
Ionian households spend below the average. In ELSTAT's 2024 Household Budget Survey the Ionian Islands Region sits at 92.0 against a national average of 100. Spending on "restaurants, cafés and hotels" rose +6.5% nationally. ELSTAT publishes this category at national level; for the Zakynthos picture we have prepared a template questionnaire.
Detailed figures
| Year | Passengers | Beds | Tour./bed | Dining (€) | €/visitor | Food businesses | €/business | Stay (Ionian) |
|---|
💬 This study is live — tell us what you think
This study does not stop here. It is updated with new figures every month and expanded on the basis of citizens' suggestions. Several of the sections you see were added because someone asked for a particular angle to be examined.
Have another angle to suggest? A comparison with Kefalonia or Corfu, data by area of the island, the relationship with property prices, staffing levels — anything you think is missing, write to us. If public data exists, we will look into it.
Sources
- Passenger traffic: Fraport Greece — Αεροδρόμιο Ζακύνθου «Δ. Σολωμός», Air Traffic Statistics (μηνιαία δελτία, το δωδέκατο κάθε έτους είναι αθροιστικό).
- Hotel capacity: Ξενοδοχειακό Επιμελητήριο Ελλάδος, Δυναμικότητες — «Ανά Περιφέρεια · Νομό · Νησί», επεξεργασία ΙΤΕΠ.
- Turnover: ΕΛΣΤΑΤ, «Εξέλιξη κύκλου εργασιών επιχειρήσεων παροχής καταλύματος και υπηρεσιών εστίασης», ετήσιοι πίνακες ανά Περιφερειακή Ενότητα (ΚΑΔ 55 και 56, NACE Αναθ. 2).
- Retail trade (supermarkets, NACE 4711): ΕΛΣΤΑΤ, «Εξέλιξη κύκλου εργασιών επιχειρήσεων λιανικού εμπορίου», τριμηνιαία. Δημοσιεύεται ανά Περιφέρεια — όχι ανά νησί.
- Household spending: ΕΛΣΤΑΤ, Έρευνα Οικογενειακών Προϋπολογισμών 2024 (δείγμα 6.198 νοικοκυριών πανελλαδικά).
- British market: ABTA Holiday Habits και ΙΝΣΕΤΕ, «Τα ξενοδοχεία All Inclusive στον ελληνικό τουρισμό» (2014). Αφορούν τη βρετανική αγορά και την Ελλάδα συνολικά, όχι τη Ζάκυνθο.
- Reviews: δημόσιες αξιολογήσεις Google Maps, συλλογή μέσω SerpApi. Δείγμα επιχειρήσεων της Ζακύνθου με τις περισσότερες κριτικές. Αποθηκεύονται μόνο συγκεντρωτικά στοιχεία ανά έτος και τα ονόματα των επιχειρήσεων — όχι το κείμενο ή οι συντάκτες των κριτικών, σύμφωνα με τους όρους χρήσης του Google Maps Platform. Αναλυτικά: ξενοδοχεία · εστίαση.