Hyper-personalise data and hack search results: five lessons from the AI for Guest Experience webinar
AI is now part of the daily toolkit for hospitality. The takeaway from the AI for Guest Experience webinar, sponsored by Lolly: use tech to make the guest experience feel smoother and more human - not colder.
On the panel: Peter Moore (Lolly), Will Francis (digital marketing and AI), Dean Culpan (Miiro Templeton Garden; formerly South Place), and David Turner (Compass UK & Ireland). Here are the five lessons worth acting on this quarter.
1) Win the new search game
Google still dominates search, but guests are asking AI tools for recommendations - and those tools return a tiny shortlist. As Francis put it, "If you're not in the top three, you're invisible."
- Be everywhere: consistent SEO, up-to-date listings, and ongoing PR.
- Post fresh content: AI products prefer recent signals. Keep a steady rhythm on Reddit, TikTok, and Instagram. Quality, volume, and consistency matter.
- Localise at scale: Culpan ran an AI-led SEO strategy across 12 languages at South Place, adapting key phrases by market. Result: more impressions, more traffic, and booking conversion uplifts up to 120%.
2) Hyper-personalisation that feels human
Personalisation isn't "nice to have" - it removes friction. Moore and Turner highlighted the value of knowing preferences and flagged allergens before a guest needs to ask. That speeds decisions and reduces risk.
- Capture what matters: allergen profile, menu dislikes, room/location preferences, occasion, and booking context (leisure vs corporate).
- Use it with tact: Francis warned that guests still want a human feel with AI in the background. Culpan's rule: be helpful, not creepy.
- Train teams to close the loop: staff should confirm details naturally, correct errors fast, and adapt tone and pace to the guest.
3) Keep boundaries clear
Turner's line in the sand: solve real problems, don't over-familiarise. Personalisation should feel like service, not surveillance.
- Context first: two guests can like the same thing and still want very different interactions. Leisure may want stories; corporate may want speed.
- Set limits: avoid over-notifying, over-recommending, or guessing sensitive details. Let the guest lead.
4) Treat data like a privilege, not a right
Consent isn't a legal checkbox - it's an exchange of value. Explain how sharing data improves safety and convenience, then deliver on it.
- Ask at the right moment: tie consent to a clear benefit (e.g., pre-filtered menus for allergens, faster check-in).
- Collect the minimum, store securely, and set retention rules. Give guests control to view/edit preferences.
- Document your approach and train staff on process and language. Guests should hear the same message online and on-site.
- For guidance, review the UK Information Commissioner's Office resources on GDPR here.
5) Make AI do the heavy lifting behind the scenes
Moore introduced "Lena," a digital human that supports canteens and high-volume sites by scanning trays, forecasting workforce needs, and adjusting displays by taste. The point isn't novelty - it's outcomes.
- Dynamic menus: show what fits the guest's profile first. Reduce choice overload, increase average spend, and cut waste.
- Forecasting and prep: predict covers and items by daypart and weather; prep the right volumes and protect margin. Teams building these capabilities may find the AI Learning Path for Production Planners a useful practical resource.
- Start small: pick one site and one use case, baseline your KPIs (waste %, dwell time, average spend), run a four-week test, then scale.
Quick-start playbook for operators
- Week 1-2: Map the guest flow. List the top five friction points (search, booking, arrival, ordering, payment).
- Week 3-4: Implement a fresh-content cadence and multilingual SEO for your top three markets. Standardise NAP (name, address, phone) across listings.
- Week 5-6: Collect consented preferences at booking and pre-arrival. Add allergen prompts and auto-filtered menus.
- Week 7-8: Pilot one AI use case (menu personalisation or demand forecasting). Train staff on scripts and escalation paths.
- Week 9-10: Review results. Keep what moved the numbers, drop what didn't, and expand to the next site.
About the sponsor
At Lolly, the view is simple: AI should amplify human hospitality. Blending intelligent automation with service instincts creates memorable, seamless experiences. Events like this push the industry forward, and Lolly is committed to that conversation.
itslolly.com
Next up
Want to hear more about how AI is changing the workplace? Join the free one-hour webinar "AI: Building Trust in the Workplace" on 18 February. Secure your place early.
Upskill your team
If you're building practical AI capability for Hospitality and Events roles, explore curated training by job role - try the AI Learning Path for Production Planners or the AI Learning Path for Supply Chain Analysts.
Your membership also unlocks:
