Lirevon
ServicesWorkIndustriesPricingAboutContact
العربية
Book a free audit
Loading
Lirevon

A creative + AI studio in Lahore, building brands, websites and AI consoles for the Gulf.

Lahore, Pakistan

Studio

About
Work
Pricing
Contact

Services

Brand & logo
WordPress
Web apps
AI consoles
Mobile apps
Content & video

Industries

Clinics
Restaurants
Real estate
Retail
Professional services

Resources

  • Blog
  • Tools
  • Automations

Explore

Work
Solutions
Answers
Locations
Glossary
Explore All

Start a project

  • hello@lirevon.com
  • WhatsApp
  • Book a call

© 2026 Lirevon. All rights reserved.

PrivacyTerms
LinkedInInstagramBehance

Work / Sulafa Boutique Hotel

All work

AI chatbots

Sulafa Boutique Hotel

Client:Boutique hotel (48 rooms) · DohaYear:2025Industry:HospitalityDuration:7 weeks
  • ai-chatbot
  • hotel
  • qatar

Overview

Bilingual EN/AR WhatsApp concierge covering pre-arrival requests, room-service ordering, local-activity booking, and late-checkout negotiation. Upsell revenue per occupied room grows AED 94; Google rating rises from 4.3 to 4.8.

The challenge

Sulafa's small front-desk team of three was fielding identical pre-arrival WhatsApp messages from every guest: airport transfer, room preferences, early check-in, prayer-time requests, and halal dining options. The repetitive workload left staff with little capacity to handle genuine service recovery moments. Management also wanted a way to surface paid upgrades and local experiences without feeling pushy to GCC guests who value discretion.


The solution

Our approach

Lirevon designed a three-phase conversation flow: pre-arrival personalisation (24 h before check-in), in-stay concierge (room service, local bookings, housekeeping), and post-stay review nudge. Upsell offers are triggered only on positive sentiment signals and limited to one offer per conversation turn. A Stripe payment link is embedded inline for instant upgrade confirmation. The bot escalates to a human agent if sentiment drops below a threshold or a complaint keyword is detected.


Outcomes

What we delivered

Upsell revenue per occupied room increased by AED 94, adding AED 54,000 to annual revenue at 60% occupancy

Google rating improved from 4.3 to 4.8 stars within three months of launch

Front-desk WhatsApp handling time reduced by 74%, giving staff capacity for genuine service recovery

Pre-arrival request fulfilment accuracy reached 99.1%, eliminating room-preference mismatches on arrival


Key metrics

+AED 94Upsell revenue/occupied room
4.3 → 4.8 ★Google rating
−74%Front-desk WA handling
99.1%Pre-arrival accuracy

Tech stack

  • Next.js 15
  • WhatsApp Cloud API
  • Gemini
  • Supabase
  • Stripe

Services

  • AI Chatbot Development
  • WhatsApp Integration
  • Upsell Flow Design
  • Payment Link Embedding
  • Sentiment-Based Escalation

Client testimonial

“Guests now arrive knowing exactly what to expect. The bot remembered one guest asked for a prayer-direction card and had it waiting in the room — that kind of detail wins loyalty.”

— نورة الكواري، مديرة العمليات


Related work

More in this category

Multi-specialty dental clinic (5 chairs) · Riyadh

Noor Dental Centre

Bilingual EN/AR WhatsApp bot grounded in live appointment slots and CCHI-approved insurance panel; triages pain severity, verifies coverage, and books same-day emergency slots. After-hours self-booking climbs from zero to 34 appointments per week.

  • ai-chatbot
  • clinic
  • saudi

Result:After-hours bookings 0 → 34/wk; insurance verification 12min → 55s

View project

Islamic investment advisory platform · Kuwait City

Wahed Wealth

Bilingual EN/AR Sharia-compliant portfolio advisor grounded in CMAs guidelines, live sukuk yields, and halal equity screener. Average onboarding time drops from 40 minutes to 7 minutes; AUM per advisor grows 2.8×.

  • ai-chatbot
  • fintech
  • kuwait

Result:Onboarding 40 min → 7 min; AUM per advisor 2.8×

View project

Restaurant · Karachi

Tasty Bites

WhatsApp ordering agent grounded in live menu; never invents items. Transforms late-evening orderings from 61% to 97% accuracy.

  • messaging
  • ordering
  • restaurant

Result:Response 6m → 9s; late-evening orders 61% → 97%

View project

Let's build something that ranks — and runs itself.

Book a free 30-minute audit, or message us on WhatsApp. We'll send a fixed-price proposal in two days.

Book a free 30-min auditWhatsApp us