TL;DR

• Personalisation is moving from “Dear <First Name>” to real-time, intent-based offers across web, app and email.
• Revenue teams are pairing demand forecasting with offer generation to protect margin during shoulder periods.
• Ops leaders are using AI to shrink ticket backlogs, speed up onboarding and cut waste in housekeeping and F&B.
• Destination managers are trialling AI for visitor flow, accessibility, and sustainability comms.
• Trust & safety (hallucinations, bias, privacy) is now a Board-level topic with clear guard-rails and audits.

1) From Segments to Signals: Personalisation That Actually Converts

Hospitality has talked about personalisation for a decade; AI is finally making it usable. Instead of broad segments (“families”, “couples”), teams are reading real-time signals—scroll depth on room pages, add-to-trip behaviours, past stay context, even weather at origin—and generating on-page copy, FAQs and bundles in milliseconds.

What good looks like

• On-site assistants that answer policy questions with live inventory data (late checkout, connecting rooms) and surface the right room type rather than dumping the rate grid.
• Email + web coordination: if a user engaged with “pet-friendly” content, the next hero block shows a pet package with transparent fees—no manual build required.
• Accessibility: automatic generation of plain-language pages from technical accessibility audits so guests can self-assess fit before booking.

Impact: higher look-to-book, fewer abandonment points, improved guest satisfaction scores—especially for first-time visitors.

2) Revenue Teams: Forecasts In, Offers Out

Dynamic pricing is old news; what’s new is combining demand forecasts with automated offer generation. When shoulder nights look soft, AI proposes value-adds (parking, breakfast, spa credit) that protect ADR rather than racing to the bottom on rate.

Playbook snapshot

1. Ingest daily pick-up, competitor rates, events, flight arrivals, and weather anomalies.
2. Predict unsold inventory by room class and length-of-stay gaps.
3. Generate marketing-safe offer copy and push to web/email with pre-approved brand tone.

Impact: 1–3 pts uplift in conversion on soft nights, cleaner inventory, and fewer ad-hoc discounts eroding brand position.

3) Operations: The Quiet Efficiency Revolution

The headline wins aren’t chatty robots—they’re boring, compounding time savers: auto-tagged guest messages, triaged tickets, and SOP copilots that turn “Where’s the crib?” into a one-click task with the right checklist.

Where it lands

• Housekeeping: AI reorders runs when early check-ins spike; supervisors review exceptions, not everything.
• F&B: menu descriptions, allergen matrices, and 86-lists updated across channels from one source of truth.
• Contact centres: AI answers the repetitive 60–70% (parking, pets, policy) and drafts replies for the rest.

Impact: shorter SLAs, higher first-contact resolution, and happier staff who spend more time on guest-facing moments.

4) Destinations: Managing Demand, Not Just Marketing It

Destination Marketing/Management Organisations (DMOs) are moving from promotion to orchestration. AI models blend historical arrivals, accommodation capacity, and mobility data to forecast hotspots, then nudge visitors towards alternative trails, museums or time slots.

Early patterns

• Smart wayfinding in visitor apps that adapts to live crowding, mobility needs and weather.
• Sustainability nudges: suggesting low-impact alternatives and surfacing public transport options with real-time headways.
• SME enablement: auto-generated Google Business Profile posts, events descriptions and itineraries for local operators.

Impact: better dispersion, improved resident sentiment, and stronger SME participation in the visitor economy.

5) Risk, Governance and “No-Surprises” AI

As AI touches pricing, content and guest comms, leaders are formalising guard-rails. The mature pattern is simple: use approved models for each job, log prompts and outputs, red-team the risky bits (pricing, safety info), and train staff on what AI can and cannot do.

Minimum viable governance

• Policy: clear do’s/don’ts (no PII in prompts, approved data sources, disclosure rules).
• Review: human oversight for anything that could affect safety, price, or legal compliance.
• Quality: automatic checks for hallucinations, broken links, and brand tone before publishing.

Impact: fewer compliance headaches and faster roll-outs because teams know the rules of the road.

Implementation Checklist (Copy-Paste Friendly)

• Pick one guest journey (e.g., late checkout) and automate end-to-end: discovery → policy answer → upsell → paym