Hospitality Intelligence
Turn the venue into a system that learns.
RAYN builds a living picture of the venue, the room, the guest journey and the market around it — then turns that evidence into questions, recommendations and operator decisions. Not one more dashboard. An intelligence layer underneath the platform.
One intelligence layer
The intelligence is not one dashboard. It is the layer beneath the whole platform.
Venue DNA explains the business. Seat intelligence explains the room. Guest memory adds preference. Revenue and operations explain performance. Market Lens adds outside context. RAYN Brain connects the evidence; Ask RAYN gives the operator a way to question it.
The loop
The full Brain loop. From signal to learning.
Sense. Understand. Recommend. Approve. Act. Measure. Learn. Ask RAYN is the conversational surface across that loop, so the operator can question what RAYN knows, why it believes it and what it is proposing next.
01
Venue DNA
A living picture of what this venue actually is.
RAYN keeps declared venue truth, approved brand context, house rules, operating knowledge, spatial facts and learned venue context in governed layers rather than flattening everything into one generic profile. The result is a venue that can be understood consistently by the systems around it.
- Declared truthOpening hours, policies, packages, zones, service rules and the facts the venue has explicitly recorded stay distinct from model inference.
- Curated identityApproved brand voice, positioning, vocabulary and service character can be stored with revision history instead of being rewritten differently in every channel.
- Learned venue knowledgeRevenue rhythm, demand patterns, floor structure, review themes and other observed signals can sit beside declared facts without pretending they are the same thing.
- Documents become usableOperator-supplied venue documents can be ingested into a searchable knowledge layer, so Ask RAYN can retrieve the relevant passage rather than inventing an answer.
02
Seat & Space Intelligence
Every table has a profile. Every guest has a preference.
RAYN does not treat every seat as interchangeable inventory. It models the characteristics of each space, translates how guests describe what they want, and sends those preferences into one canonical recommendation path so the widget, WhatsApp, Voice and agent surfaces do not invent different answers.
- Inventory DNAEach table or space can carry governed attributes plus scored axes for quiet, view, privacy, premium, family and shade, with indoor/outdoor held as a categorical fact rather than guessed from a score.
- Natural language becomes structure“Outside, quiet, with a view” can be translated into the same preference vocabulary the recommendation engine understands: terrace, quiet, nice view, window, romantic, family-friendly, lively, premium and more.
- One ranking authorityThe canonical zone-and-table recommendation path owns the ranking. Other channels can explain the match, but they do not silently create their own seat-scoring logic.
- Guest seat DNAVenue-specific spatial preferences can sit in guest memory, while selections and overrides create feedback about what that guest actually prefers rather than what the system merely assumed.
03
RAYN Brain
The decision loop underneath the platform.
RAYN Brain is the layer that turns signals into operator decisions. It senses what is happening, builds context, forms a recommendation, waits for approval where required, measures what happened next and feeds the result back into future reasoning.
- Sense the venueBookings, demand, conversations, payments, reviews, operating context and selected external evidence enter the intelligence layer with provenance.
- Understand before recommendingSignals are assembled into venue context rather than treated as isolated metrics, so a soft service, a high-interest zone and a review pattern can be reasoned about together.
- Recommend with evidenceA recommendation can carry its reasoning, confidence and evidence instead of appearing as an unexplained instruction from a black box.
- Measure and learnApproved actions and their outcomes can be measured back into the recommendation loop so successful patterns strengthen and weak ones lose weight.
The full Brain loop
Sense → Understand → Recommend → Approve → Act → Measure → Learn.
The Brain is not one model call and it is not one dashboard. It is the governed loop connecting evidence, recommendations, operator approval, execution and outcomes.
- Evidence first
Venue truth, spatial intelligence, guest journeys, revenue, reviews and market context remain attributable to their source.
BookingsSeat DNARevenueReviewsMarket - Human control where it matters
Commercial and sensitive actions can remain approval-gated rather than being released simply because a model proposed them.
- Outcomes become memory
What was approved, what happened afterwards and what proved ineffective can feed the next cycle instead of disappearing after the card is dismissed.
Illustrative — from signal to recommendation
Terrace interest is rising, Thursday conversion is weak and a particular guest segment historically prefers outdoor seating. RAYN can combine those signals into a targeted recommendation, show why it made the suggestion and leave the commercial decision with the operator.
04
Ask RAYN
The conversational interface to the venue’s intelligence.
Ask RAYN is not a generic chatbot placed on top of reports. It is designed to question the venue’s declared facts, learned knowledge, uploaded documents, live analytical tools and recommendation context — then return the answer in plain language with the evidence behind it.
- Ask the business naturally“What has changed this week?”, “Which zone is attracting attention but not converting?” or “What do we know about this guest’s seating preference?”
- Declared and learned knowledge stay separateAsk RAYN can distinguish what the venue has explicitly recorded from what RAYN has learned, inferred or retrieved from supporting evidence.
- Search the venue’s documentsUploaded PDFs and venue documents can be semantically searched through the knowledge layer when the answer lives in a source document rather than a database field.
- Move from answer to actionRecommendations, action cards and evidence can sit alongside the answer so the operator can decide what to do next rather than stopping at an observation.
05
Revenue & Operator Intelligence
Read the commercial picture in the context of the room.
RAYN connects revenue and operating performance back to the physical inventory being sold. The point is not simply to know covers or revenue. It is to understand which service, zone, table, time and operating condition produced the result — while there is still something the team can do about it.
- RevPASH, spatiallyRevenue per available seat-hour can be read by service, zone and table rather than only as a blended venue average.
- Demand and leakage togetherHigh-interest spaces, weak conversion, empty peak inventory and no-show exposure can be read against the same operating context.
- Live operator readoutBookings, covers, occupancy, demand and commercial performance can sit in one command view rather than being reconstructed after service from disconnected systems.
- Forecast, then decideForward-looking demand can inform yield or service recommendations, while the actual commercial decision remains with the operator.
07
Reports
The structured evidence underneath the intelligence.
Not every question needs an AI answer. RAYN also carries a broad reporting estate for the daily, weekly and longer-term read-back of the venue. The important design principle is that reports, Ask RAYN and recommendations should read from the same governed data rather than creating three competing versions of the truth.
- OperationsBookings, service and foundational operational reporting give the venue a consistent read-back of what happened.
- Revenue & behaviourRevenue, spatial signals, behavioural patterns, readiness, KPIs and analytics show how demand moved through the venue and where commercial performance changed.
- Market & reputationMarket Lens, competitor context, menu intelligence, social and reputation views add the outside-in picture around the venue’s own operating data.
- Value and proofValue Intelligence and proof-oriented views are designed to separate measured evidence from assumptions, so the operator can see what is actually supported.
06
Competitor Intelligence
Know where the venue sits in the market — without pretending the market is your own data.
Competitor Intelligence and Market Lens add outside context around the venue: public ratings, review themes, review velocity, pricing and package positioning, local benchmarks and other verified market signals. External evidence remains identifiable as external evidence rather than being blended silently into first-party truth.
- Competitive setPublic competitor signals can be organised into a consistent local comparison rather than checked manually one venue at a time.
- Market LensWhere enough comparable evidence exists, RAYN can frame menu, packages, demand and booking-journey performance against an anonymised cohort rather than exposing another venue’s private data.
- Themes, not just ratingsReview sentiment, review velocity and recurring themes can show where guests perceive meaningful differences in service, product or experience.
- Privacy before theatreBenchmark output can be withheld when the comparable cohort is too small or the result could reveal an individual venue, rather than forcing a number onto the screen.
- Your venue4.6
- Rooftop rival4.4
- Lounge rival4.2
- Bistro rival4.1
Review velocity rising. One rival is slipping on service — a door to win this month.
Illustrative competitive-set example.
Questions
RAYN Intelligence, answered.
Is this just another dashboard to read?
No. RAYN assembles the evidence and proposes a move with its reasoning and a confidence level — you approve or decline. The dashboards exist, but the point is the recommendation, not one more screen to interpret.
Where does the intelligence get its data?
From the RAYN evidence layer: bookings, demand, revenue, reviews, operating context and venue rules, with external market evidence kept identifiable when it is used. Ask RAYN, reports and recommendations are designed to read from the same governed sources.
Will it change prices or message guests on its own?
No. Sensitive and commercial actions — yield moves, deposits, guest messaging — wait for a human yes. RAYN drafts and evidences the move; a person approves it.
What happens when the data is thin?
It says so. Recommendations carry a confidence level and name the evidence behind them, and RAYN holds back rather than guessing when there is not enough to stand on.
See the intelligence layer on your venue.
A walkthrough takes fifteen minutes — Venue DNA, seat intelligence, RAYN Brain, Ask RAYN, reporting and the market context around your venue.
Book a 15-minute walkthrough