About this report
How to read the figures in this report - please read before citing anything This report contains two kinds of number and they are labelled differently throughout. MEASURED figures are published statistics from named sources, cited inline and listed in full in the source register at the end. MODELLED figures are our own calculations, built from stated inputs and stated assumptions, and are always introduced as modelled. They are illustrative arithmetic, not observed findings. No figure in this report comes from a primary survey, because no primary survey was conducted. Any figure you quote externally should carry the same distinction. |
This report covers the four largest travel and hospitality markets in Western Europe by operator count: the United Kingdom, France, Spain and Italy. It examines demand and seasonality, distribution channel structure, the communications infrastructure position, and the regulatory constraints that bear on how operators handle customer contact. Companion reports cover each market individually in greater depth.
Why the modelled sections exist
Several of the questions that matter most to a travel operator have no published answer anywhere. Nobody collects national statistics on how many booking enquiries go unanswered, or what proportion of inbound contact to a hotel is a booking enquiry rather than a supplier call. Those figures do not exist in any statistical office.
Rather than leave the questions unaddressed or imply a precision we do not have, we model them. Each model states its inputs, names the source of every input that is measured, states every assumption explicitly, and presents a range rather than a point estimate. The purpose is to give operators a defensible structure they can populate with their own data - not to tell them what their number is.
What would make this report stronger
Primary research. A structured survey of operators across these markets would replace the modelled sections with measured ones. Until that exists, the modelled sections should be read as a framework rather than as evidence.
1. Executive summary
Four findings, of which the first three are measured and the fourth is modelled.
· Demand is growing, but the growth is moving. Demand across all four markets is growing but decelerating, and the deceleration is uneven between hotel and non-hotel accommodation. Italy's non-hotel sector grew 14.7% in Q1 2026 against 3.9% for hotels. France's Q1 2026 saw hotels up 2.5% while other collective accommodation fell 1.7%. The competitive pressure on hotels is not coming from other hotels.
· Distribution is polarised, not average. Direct bookings represented 50.9% of European hotel distribution against 28.8% for OTAs, but the distribution is heavily polarised - roughly one hotel in five takes more than half its business through OTAs. The average conceals two very different populations.
· The infrastructure position is not comparable across markets. The four markets are at four entirely different points in the copper and PSTN transition. The UK faces a hard deadline of 31 January 2027 with approximately 1.9 million lines outstanding. Spain finished in May 2025. France and Italy are mid-transition on geographic schedules running to 2030 and beyond. A single European infrastructure strategy is not available.
· The contact layer is unmeasured. No statistical office in any of these markets measures unanswered booking enquiry. Section 6 models it transparently. The modelled ranges are large, which is itself the finding: operators cannot currently answer a question that materially affects their revenue.
2. Market demand and seasonality
2.1 The measured position
All four markets recorded growth into 2026, at different rates and with different internal composition.
Market | Most recent measured position | Source |
Spain | Overnight stays +1.1% year on year in April 2026, with 58.5% of bed-places filled (0.9% below April 2025). January 2026 stays +3.4%, occupancy 48.0%. Hotel price index +4.3% year on year in January. | INE, provisional |
Italy | Q1 2026: 23m arrivals and 71.6m overnight stays, up 4.2% and 7.5%. Non-hotel accommodation +14.7% against hotels +3.9%. Foreign visitors 54.6% of stays, up 12.3% against 2.2% domestic. | ISTAT |
France | Q1 2026 occupancy excluding campsites +1.0%, with hotels +2.5% and other collective accommodation -1.7%. Summer 2025 reached 257.8m nights (+3.7%), hotels 90.4m (+4.4%). Winter 2026 87.6m nights. | INSEE |
UK | Hospitality employs approximately 3.6m people and contributes around GBP 96bn annually, across 176,685 businesses of which 99.6% are SMEs. | UKHospitality; House of Commons Library |
2.2 What the composition tells you
Two patterns are visible across markets and neither is about total volume.
The first is that non-hotel accommodation is growing considerably faster than hotels wherever the statistics separate them. Italy's Q1 2026 gap was 14.7% against 3.9%. France's Q1 2026 showed the opposite sign but the same divergence - hotels up 2.5%, other collective accommodation down 1.7% - which suggests the French non-hotel segment is behaving differently again, and that generalising across markets is unwise.
The second is the internationalisation of demand. Italy's foreign visitor stays grew 12.3% in Q1 2026 against 2.2% domestic. Where growth is disproportionately international, the operational consequence is linguistic: contact volume grows fastest in the languages an operation is least likely to staff overnight, and the growth arrives concentrated in the weeks of the source market's holidays rather than spread across the year.
The operational reading of the demand data Aggregate growth of one to seven per cent sounds manageable and is not the operational problem. The problem is that the growth is disproportionately foreign, disproportionately concentrated in specific weeks, and disproportionately outside the hotel sector. An operation resourced against last year's total will be adequate in aggregate and short in exactly the hours the new demand arrives. |
3. Distribution: the channel picture
3.1 The measured position
European hotel distribution recorded direct bookings at 50.9% and OTAs at 28.8%, with the remainder across GDS, wholesale and corporate channels.
The average is the least useful number in that sentence. The same data shows 28% of hotels with OTA shares between 30% and 49%, and roughly one hotel in five above 50%. That is not a distribution around a mean; it is two populations - operators who have built a direct channel and operators who have not - and a European average sits in the empty space between them.
3.2 Where the contact layer sits in this
Assisted enquiry - voice, chat and any channel where a person is helped to buy - is not separately reported in any of the standard distribution datasets. It is generally absorbed into 'direct' or into a residual category.
This matters because it makes the channel invisible to the function that manages distribution. An operator can see that direct is 50% and OTA is 29% without being able to see how much of that direct business arrived by telephone, what it converted at, or what it cost to serve. Decisions about direct booking strategy are consequently being made with one of the direct channels missing from the analysis.
3.3 Investment context
European hotel investment reached EUR 14.65 billion in 2025, with France, the UK, Spain and Germany accounting for EUR 9.61 billion - approximately 66% of transaction volume. 63% of European hoteliers anticipated improving conditions in the 2025 European Accommodation Barometer.
The relevance to the contact layer is ownership churn. Transaction volume at this level means a material number of properties changing hands, and properties changing hands is the mechanism by which estates become technologically fragmented - each acquisition arriving with its own telephony contract, its own PMS version and its own local installer relationship.
4. The infrastructure position
The four markets are at four different points in the copper and PSTN transition, and the differences are large enough that a single European migration strategy is not available.
Market | Position | Deadline character |
United Kingdom | Openreach PSTN and WLR switch-off confirmed for 31 January 2027, stated as final after the earlier move from December 2025. Approximately 1.9 million lines still to migrate as at mid-2026. | Hard national deadline, approximately 18 months out |
France | Commercial closure of copper services completed 31 January 2026. Technical closure proceeding lot by lot to around 2030, with 763 communes in the first technical batch. | Geographic, site-specific, no single national date |
Italy | TIM began decommissioning in May 2024 with the first 62 all-copper exchanges. Approximately 6,700 targeted by 2028, with the wider FiberCop plan running to 2030-31. | Geographic and phased, regionally uneven |
Spain | Complete. Telefonica closed retail copper access in 2024 and shut the final 661 exchanges in May 2025. | No deadline remains |
What this means for a multi-market operator
A group operating across all four markets has one market with a hard deadline eighteen months away, two markets on geographic schedules where exposure depends on which commune or exchange each property sits behind, and one market where the transition is finished.
The practical consequence is that migration planning has to be done per property rather than per country, and that the urgency messaging that is legitimate in the UK is inaccurate in Spain. Operators being told by suppliers that a European deadline applies should ask which one.
5. The regulatory position affecting customer contact
Four regimes bear directly on how operators handle contact, and three of them changed in the last eighteen months.
· EU AI Act. Article 50 transparency obligations became applicable on 2 August 2026, requiring that people interacting with an AI system are told so at the point of contact. Penalties reach EUR 15 million or 3% of total worldwide annual turnover. The Digital Omnibus on AI, adopted in June 2026, deferred high-risk obligations for Annex III systems to 2 December 2027 and Annex I to 2 August 2028 - but expressly excluded Article 50 from that deferral. Any operator running a chatbot or voicebot in the EU is in scope now.
· European Accessibility Act. Applicable since 28 June 2025, covering e-commerce and passenger transport services. Guest-facing conversational interfaces form part of that service and are frequently missed because accessibility is treated as a web obligation.
· PCI-DSS. Version 4.0.1 future-dated requirements became mandatory on 31 March 2025. Relevant wherever card data is taken by telephone, which in travel is common.
· Workplace monitoring. Recording and quality monitoring engage employment law as well as data protection, and the requirements differ sharply. Italy requires a collective agreement or authorisation from the labour inspectorate under Article 4 of the Workers' Statute. France requires CSE consultation. Spain requires prior explicit information to workers and their representatives under LOPDGDD Articles 89-90. The UK has no equivalent process step.
The finding operators most often miss Of the four markets, three have an employment-law process step with elapsed time attached before call recording can go live. Italy's authorisation route in particular can outlast the technical deployment. A four-market rollout planned on UK assumptions will stall. |
6. Modelled: the scale of unmeasured enquiry
This section is a model, not a finding Everything in section 6 is our own calculation. No statistical office measures unanswered booking enquiry in any of these markets, and we did not survey operators. What follows states its inputs, names which are measured and which are assumed, and presents ranges. Treat it as a structure to populate with your own data. |
6.1 What is measured and what is not
Input | Status | Basis |
Number of accommodation establishments | MEASURED | National statistics offices and sector analyses - see country reports |
Occupancy and overnight stays | MEASURED | INE, ISTAT, INSEE |
Inbound contact volume per property | NOT MEASURED | No published source in any of the four markets |
Proportion of inbound that is booking enquiry | NOT MEASURED | No published source |
Unanswered and abandoned rate | NOT MEASURED | No published source |
Conversion on assisted enquiry | NOT MEASURED | No published source |
Average booking value | PARTIALLY MEASURED | Price indices published; segment-level booking values are not |
Four of the seven inputs required to size this question have no published source anywhere in Europe. That is the substantive finding of this section, and it holds regardless of what any model built on top of it produces.
6.2 The model structure
For any individual operator the calculation is five steps. We publish the structure rather than a European total, because a European total built on four assumed inputs would imply a precision that does not exist.
1. Total inbound contact volume over a defined period, by hour. Source: your own contact platform.
2. Less repeat attempts by the same caller, giving distinct contacts. Source: your platform, if it can distinguish caller identity. Many legacy systems cannot, which is itself informative.
3. Multiplied by the proportion that are booking enquiries rather than supplier, service or internal contact. Source: manual coding of a representative sample over a fortnight. There is no benchmark worth using here.
4. Multiplied by a conversion assumption, anchored to your measured conversion on handled enquiries and stated as a range rather than a point.
5. Multiplied by segmented average booking value. Source: your own revenue reporting.
6.3 Why we are not publishing a European figure
It would be straightforward to multiply an assumed contact volume by an assumed enquiry proportion by an assumed conversion rate across 16,778 Spanish hotels and produce a large, quotable number. That number would be arithmetic performed on four guesses, and it would be wrong in ways nobody could audit.
The honest position is that the sector does not know the answer, that the question is material, and that the only reliable route to a figure is an operator measuring their own. Section 6.2 is that route.
6.4 What operators can measure this quarter
· Inbound versus answered, by hour, for a representative fortnight. Available from most modern platforms and absent from most legacy ones.
· Repeat-caller rate - distinct numbers against total attempts. This single figure changes the size of the problem materially and almost nobody has it.
· Enquiry proportion, by manual coding. Two weeks of one person's time.
· Property-level inbound volume, where central reporting exists but property reporting does not. This is where the escalation path inverts and it is invisible in group reporting.
7. What this means for operators
· Plan against composition, not totals. Growth is arriving disproportionately from foreign markets and outside the hotel sector. Resourcing against last year's aggregate will be adequate in total and short in the specific hours and languages the new demand arrives.
· Do not benchmark against the European distribution average. A European average of 50.9% direct conceals two populations. Benchmark against operators of your type and size, not against the continent.
· Treat migration as four separate programmes. The UK has eighteen months and a hard date. Spain has finished. France and Italy depend on the specific commune or exchange. Plan per property.
· Start the recording compliance work before procurement completes. Three of the four markets have an employment-law process step before recording can go live, and Italy's can outlast the technical build. Start those in parallel with procurement.
· Measure what nobody else is measuring. Four of the seven inputs needed to size enquiry leakage have no published source. The operators who will act on this are the ones who measure their own position first.
Source register
Every measured figure in this report, with its source. Verify currency before external citation - tourism statistics are revised, and provisional figures are marked as such by the issuing office.
Scope | Statement | Source | As at |
EU | 20.0% of EU enterprises with 10 or more employees used AI technologies in 2025, up from 13.5% in 2024 - a rise of 6.5 percentage points in a single year. | Eurostat, '20% of EU enterprises use AI technologies', December 2025 (dataset isoc_eb_ai) | 2025 |
EU | Adoption is highly uneven across the EU. Highest: Denmark 42.0%, Finland 37.8%, Sweden 35.0%, Belgium 34.5%, Netherlands 33.2%. Lowest: Romania 5.2%, Poland 8.4%, Bulgaria 8.5%. | Eurostat, December 2025 (dataset isoc_eb_ai) | 2025 |
EU | EU AI Act Article 50 transparency obligations became applicable on 2 August 2026, requiring that people interacting with an AI system are informed of that fact, with penalties up to EUR 15 million or 3% of total worldwide annual turnover. | European Commission, transparency obligations under Article 50 of the AI Act | 2 Aug 2026 |
EU | The Digital Omnibus on AI, adopted by the European Parliament on 16 June 2026 and the Council on 29 June 2026, deferred high-risk obligations for standalone Annex III systems to 2 December 2027 and for AI embedded in regulated products under Annex I to 2 August 2028. Article 50 was expressly excluded from the deferral. | Gibson Dunn, Freshfields and DLA Piper published analyses of the Digital Omnibus on AI, 2026 | Jul 2026 |
EU | The European Accessibility Act (Directive (EU) 2019/882) has applied since 28 June 2025 and covers e-commerce and passenger transport services. | Directive (EU) 2019/882 | 28 Jun 2025 |
EU | PCI-DSS v4.0.1 future-dated requirements became mandatory on 31 March 2025. | PCI Security Standards Council | 31 Mar 2025 |
EU | Direct bookings accounted for 50.9% of European hotel distribution in 2023, with OTAs at 28.8%. Distribution is highly polarised: 28% of hotels report OTA shares of 30-49%, and roughly one hotel in five exceeds 50%. | D-EDGE Hotel Distribution Report, Europe | 2023-24 |
EU | European hotel investment reached EUR 14.65 billion in 2025, of which France, the UK, Spain and Germany accounted for EUR 9.61 billion - approximately 66% of transaction volume. | Global Asset Solutions, European Hotel Transactions Report 2025 | 2025 |
EU | 63% of European hoteliers anticipated improving conditions in the 2025 European Accommodation Barometer. | European Accommodation Barometer 2025 | 2025 |
EU | ITU-T Recommendation G.114 sets 150 ms one-way latency as the threshold for acceptable conversational voice quality. Industry-standard supporting thresholds are jitter under 30 ms and packet loss under 1%. | ITU-T Recommendation G.114 | current |
UK | Hospitality is the third largest employer in the UK, with approximately 3.6 million people working in the sector, and contributes around GBP 96 billion annually to the economy. | UKHospitality, Facts and Stats | 2025-26 |
UK | There were 176,685 hospitality businesses in the UK in March 2025, of which 99.6% were small or medium-sized enterprises. | House of Commons Library Research Briefing CBP-10333, February 2026 | Mar 2025 |
UK | ONS recorded approximately 69,000 vacancies in the accommodation and food service series in March 2026, down from an annual average of 79,000 across 2025 and a 2022 peak above 177,000. | ONS vacancy series, accommodation and food service activities | Mar 2026 |
UK | UKHospitality reported in January 2026 that the sector employed 8,784 fewer people in December 2025 than in November 2025, and 20,014 fewer than in September 2025. Trade bodies estimate 150,000 to 200,000 hospitality jobs are at risk during 2026. | UKHospitality, January 2026 | Jan 2026 |
UK | The Openreach PSTN and WLR switch-off is confirmed for 31 January 2027, stated as final following the earlier move from December 2025. Approximately 1.9 million lines were still to migrate as at mid-2026. | Openreach; ISPreview, June 2026 | Jun 2026 |
Spain | Spain had 16,778 hotel establishments in 2025. Independent hotels accounted for 57.65% of the market, reflecting a large base of family-owned properties. | Mordor Intelligence, Hospitality Industry in Spain | 2025 |
Spain | Spain expects nearly 775 new establishments and approximately 52,000 additional rooms between 2025 and 2028. | Mordor Intelligence; Spanish hotel pipeline analyses | 2025-28 |
Spain | Overnight stays in Spanish hotel establishments rose 3.4% year on year in January 2026. 48.0% of available bed-places were filled, 1.1% more than January 2025. The hotel price index rose 4.3% year on year. | INE, Coyuntura Turistica Hotelera (EOH/IPH/IRSH), January 2026, provisional | Jan 2026 |
Spain | In February 2026, 53.5% of available bed-places were filled, 0.1% below February 2025. Across the first two months of 2026 overnight stays grew 2.2% year on year. The Canary Islands recorded the highest occupancy at 78.4%. | INE, Coyuntura Turistica Hotelera, February 2026, provisional | Feb 2026 |
Spain | Overnight stays rose 1.1% year on year in April 2026, with 58.5% of available bed-places filled - 0.9% below April 2025. | INE, Coyuntura Turistica Hotelera, April 2026, provisional | Apr 2026 |
Spain | Telefonica closed retail copper access in 2024 and shut the final 661 copper exchanges in May 2025, completing the copper switch-off in Spain. | Telefonica; CNMC blog, May 2025 | May 2025 |
Spain | Spanish employee monitoring is governed by Articles 89 and 90 of the LOPDGDD alongside GDPR, requiring that workers and their representatives are informed in advance and explicitly about monitoring measures. | Ley Organica 3/2018 (LOPDGDD), Articles 89-90 | current |
Italy | Italian accommodation establishments recorded 23 million arrivals and 71.6 million overnight stays in Q1 2026, up 4.2% and 7.5% respectively on Q1 2025. | ISTAT, I flussi turistici - I trimestre 2026 | Q1 2026 |
Italy | Growth was concentrated outside hotels: non-hotel establishments grew 14.7% in overnight stays against 3.9% for the hotel sector. | ISTAT, I flussi turistici - I trimestre 2026 | Q1 2026 |
Italy | Foreign visitors accounted for 54.6% of Q1 2026 overnight stays. Foreign stays rose 12.3% year on year against 2.2% for domestic visitors. | ISTAT, I flussi turistici - I trimestre 2026 | Q1 2026 |
Italy | Nights spent in Italian hotels reached 288 million in 2025, and over 535 million including non-hotel accommodation - close to a tenfold increase on 1924. | ISTAT, Dal Grand tour al turismo di massa | 2025 |
Italy | TIM began decommissioning the copper network in May 2024 with the first 62 all-copper exchanges. Approximately 6,700 exchanges are targeted by 2028, with the wider FiberCop plan running to 2030-31. | TIM Group press release, 24 May 2024; FiberCop decommissioning plan | 2024-31 |
Italy | Article 4 of the Italian Workers' Statute (Law 300/1970 as amended) governs systems from which remote monitoring of workers may derive. Systems capable of monitoring worker activity, including call recording and quality monitoring, generally require a collective agreement with union representatives or authorisation from the territorially competent office of the Ispettorato Nazionale del Lavoro. | Law 300/1970, Article 4, as amended | current |
Italy | Italy has 32,943 hotels, 1.1 million rooms and 2.3 million beds, making it the largest hotel capacity in Europe. | Federalberghi, X Rapporto sul sistema alberghiero e turistico ricettivo in Italia | 2025 |
Italy | Three-star hotels and residenze turistico-alberghiere are the largest category at 18,182 establishments, 55.2% of supply. There are 6,639 four-star and 771 five-star hotels; one and two-star properties account for 2,339 and 5,002 respectively. | Federalberghi, X Rapporto | 2025 |
Italy | The average Italian hotel has 69.3 beds. Properties in the south and islands are larger, averaging 91.4 beds. | Federalberghi, X Rapporto | 2025 |
Italy | 61% of the 283.9 million tourist presences recorded in Italy in 2024 were in hotels, up 3.1% year on year. | Federalberghi, X Rapporto | 2024 |
France | French collective tourist accommodation recorded 257.8 million overnight stays in summer 2025, up 3.7% on summer 2024 - an increase of 9.1 million nights. Hotels grew faster than the market at 4.4%, reaching 90.4 million nights. | INSEE, Saison touristique d'ete 2025, Insee Focus 363 | Summer 2025 |
France | Q3 2025 occupancy rose 2.9% year on year across all accommodation types: hotels 4.3%, campsites 2.2%, other collective accommodation 2.4%. | INSEE, Informations rapides 286 | Q3 2025 |
France | Q4 2025 occupancy in collective tourist accommodation excluding campsites was 2.2% above Q4 2024. | INSEE, Informations rapides 39 | Q4 2025 |
France | Q1 2026 occupancy excluding campsites was 1.0% above Q1 2025, with hotels up 2.5% but other collective accommodation down 1.7%. | INSEE, Informations rapides 120 | Q1 2026 |
France | The 2026 winter season (December 2025 to March 2026) recorded 87.6 million overnight stays in hotels and other collective accommodation, 1.6 million more than winter 2025. | INSEE, Saison touristique d'hiver 2026, Insee Focus 382 | Winter 2026 |
France | Commercial closure of copper services in France completed on 31 January 2026. Technical closure is proceeding lot by lot to around 2030, with 763 communes in the first technical batch. | Arcep, copper closure dossier, January 2026; Orange Business all-IP schedule | Jan 2026 |
France | French employee monitoring requires information and consultation of the Comite Social et Economique (CSE) before implementation, in addition to CNIL requirements and individual employee notification. | French Labour Code; CNIL guidance | current |
France | France had 16,291 classified and unclassified hotel establishments, following a net contraction of 319 addresses during 2025. | INSEE, Parc et frequentation des hotels, annual data 2025; Coach Omnium panorama | 2025 |
France | Independent hotels represent approximately 81% of the French hotel park, around 13,000 establishments, with chains accounting for the remainder. | Coach Omnium, panorama de l'hotellerie en France | 2025 |
France | Chain hotels in France recorded 109.2 million overnight stays in 2025. | INSEE / Coach Omnium | 2025 |
France | The 2025 contraction in the French hotel park fell disproportionately on small hotels in rural and highly seasonal locations, and on properties in small and medium-sized cities with weak hotel markets. | Coach Omnium panorama | 2025 |
Sources consulted but not cited for specific figures
· Eurostat, Digital economy and society statistics - enterprises
· UNWTO / UN Tourism barometers
· European Accommodation Barometer
· Ofcom, Arcep, CNMC and AGCOM sector reporting
· Openreach migration communications
· European Commission, Shaping Europe's Digital Future
A note on provisional data
INE and ISTAT release provisional monthly and quarterly figures that are subsequently revised. Where a figure in this report is drawn from a provisional release, the source register says so. INSEE quarterly series are also subject to revision. None of the revisions typically alters the direction of the trends described here, but the specific percentages may move.