Skip to main content
Lintel/Platform
One product · four layers · end to end

The platform,
finished as one product.

Modern hotel software is more capable than ever, but it is often built in separate modules, products, and add-ons. Lintel was designed as a single system, so reservations, revenue, housekeeping, and guest history all work from the same foundation.

8Surfaces · one app
1Codebase · one log-in
Marketplace add-ons
Design principles

Six beliefs about hotel software.

The rules we wrote at the beginning. Every screen, every endpoint, every release is checked against them.

One product, not a marketplace.

No app store, no certified-partner pyramid, no plug-in tax.

Nota hub for someone else's plugins.

Restraint over decoration.

Typography and palette do the work. No drop shadows pretending to be importance.

Notcharts that bounce when they load.

Every override, in the open.

If the system suggests a rate and a human changes it, the change is named, dated, and reversible.

Nota black box that just makes numbers happen.

The data is yours.

Read-only export of the whole record, on demand, in formats your accountants already speak.

Nota "data export" enterprise SKU.

Pricing is published.

Per-room, per-month, with everything included. The contract is short enough to read at dinner.

Nota quote disguised as a conversation.

Founders on call.

For the founding cohort, a private line to the team that built it. Response measured in minutes.

Nota chatbot wearing a person's name.
Layer i. Surfaces

Eight screens.
One vocabulary.

Eight surfaces, each finished to the same standard. Learning the second surface takes ten minutes — every shortcut from the first one still works.

  • 01Shared component library — the same calendar widget appears in Reservations, Revenue, and Group Bookings.
  • 02One auth, one permission system — a single role grants access across every surface.
  • 03Cross-surface deeplinks — every entity has a permanent URL that opens it in any surface.
Tour all eight surfaces
Layer ii. Intelligence

The system
watches its own house.

A continuous loop of sensing, scoring, recommending, and writing back. Not a chatbot. A small, opinionated set of agents that read every event, every minute.

  • 01Continuous, not on-demand — the loop runs every 90 seconds across pricing, pickup, parity, guest signals, and operations.
  • 02Confidence intervals, always — every recommendation is a range with a posterior, not a number.
  • 03One-click write-back — accept a suggestion and it cascades; the audit trail records who said yes.
Intelligence loop · livecycle 14.32.04
step 01
Sense
Every event since the last cycle — bookings, pickup, parity probes.
~340ms
↓ pass
step 02
Score
Run the 9-layer rate model, anomaly detectors, and pickup forecast.
~1.2s
↓ pass
step 03
Recommend
Six or fewer decisions, each with a range and confidence.
~80ms
↓ pass
step 04
Write back · on consent
If accepted, cascade to every surface; if overridden, log it.
audit
Cadence · every 90sLatency p99 · 2.4sCycles today · 936
Layer iii. Backbone

Every event in the house,
one stream.

Every action — a guest checking in, a rate being changed, a room being cleaned — is published as a typed event into a single, durable, replayable stream.

  • 0142 typed event categories covering reservations, rates, guest, payments, ops, and parity probes.
  • 02Replayable from any point — rebuild the entire state of the hotel at any timestamp.
  • 03Yours, on request — daily exports of the raw stream to your warehouse. No data hostage.
Event stream · Maison Beaumont · last 90stail -f
14.32.04reservationbooking.confirmed · res_8e2c1f · 3 nt · room 24booking.com
14.31.58raterate.applied · deluxe_king · £438 · auto · conf 0.91revenue.engine
14.31.41housekeepingroom.cleaned · room_318 · 22m driftstaff.maria
14.31.22paymentfolio.charged · res_4a91c · £312.00 · authorisedstripe
14.30.58guestprofile.merged · guest_e_laurent · 3 records → 1guest.intel
14.30.41rateparity.probed · expedia · ok · drift 0bpschannel.mgr
14.30.14reservationbooking.modified · res_3a18b · arrival → 16 Maydirect
14.29.51housekeepingroom.inspected · room_215 · passstaff.amir
14.29.32guestpreference.added · no_nuts · guest_e_laurentfront.desk
14.29.04raterate.applied · courtyard_dbl · £312 · auto · conf 0.88revenue.engine
14.28.41reservationbooking.confirmed · res_b4f7e · 2 nt · suite 4direct
Events today · 14,318Replay window · 730 days
Layer iv. Integrations

Open at the edges,
finished in the middle.

Lintel speaks fluently to the systems that matter, with first-party connectors maintained by us — and a clean, documented API for everything else.

Channel
Booking.com
First-party · < 200ms
Channel
Expedia
First-party · < 200ms
Channel
Airbnb
First-party
GDS
Sabre · Amadeus
First-party
Payments
Stripe
Native · PSD2
Payments
Adyen
Native · multi-acquirer
Accounting
Xero · QuickBooks · Sage
Daily journal sync
Reviews
Google · TripAdvisor
Inbound & outbound
Comms
Postmark · Twilio · WhatsApp
First-party templates
Terminal
Square · SumUp
Tap-to-pay supported
ID & KYC
Onfido · Veriff
Optional at check-in
Doors & keys
Salto · Assa Abloy · 4Suites
Mobile key supported
For the engineers
Open API. Everything talks.

REST and webhooks for every resource — guests, reservations, rates, folios, events. Versioned, documented, rate-limited honestly.

# Subscribe to rate changes
POST /v1/webhooks
{
  "events": ["rate.applied"],
  "url": "https://ops.example.com/hooks"
}
# → 201 created · whk_b9c4ae
For your data team
Raw export, on a schedule.

Daily, full-fidelity export of the event stream to your S3, BigQuery, or Snowflake — no SKU, no separate contract.

# Configure warehouse export
lintel export --to snowflake \
              --cadence daily \
              --retain 730

# Snowflake · lintel_events · live
# 42 tables · 14,318 rows · last 24h
Table stakes

What we don't call features,
but consider non-negotiable.

Every Lintel tier ships with these. They are not add-ons, premium tiers, or marketplace plugins.

Security
SOC 2, GDPR, PCI-DSS

Independently audited. Encryption at rest and in transit.

SOC 2 Type II · renewed Mar '26
Reliability
99.97% uptime SLA

Multi-region failover. Status page transparent to the minute.

Uptime last 90d · 99.991%
Performance
p99 < 400ms · global

Every surface measured against the same budget.

p99 last 30d · 312ms
Support
Founders on call

A private Slack with the team that built it. Response in minutes.

Median first response · 7m
Apply to the 2026 cohort

The whole platform.
One conversation.

Thirty minutes with a founder — walked through your property, your stack, your particulars. The platform on screen, and the questions you actually want answered.