01Brief capture
Brief Capture
A client message — voice or text — arrives as an open brief, not a rigid form.
Romance, wine estates, soft pacing, western routing.
Voice + text
The problem
Itinerary design is slow, manual, and non-transactionable.
01
Weeks of work
The current process: brief intake, manual property selection, route planning, narrative writing, pricing. Each step is a separate human task. A 17-day itinerary can take 20+ hours of consultant time before it reaches the client.
02
Brochures, not bookings
WETU's itinerary builder produces a visual brochure. No checkout. No payment. No contract. The client receives a link to a wetu.com subdomain, then wires money through a third party. The operator never owns the transaction.
03
One document, many audiences
A travel consultant sends the same brochure to the client, the operations team, and the lodge manager. Each audience needs different information. The client wants imagery and narrative. The ops team needs contacts and dietary logs. The lodge needs arrival times. None gets the right detail.
The solution
Natural language in. Designed itineraries out.
Murmurgo accepts voice briefs, written briefs, formal client briefs, or detailed emails — natural language, not forms. The more context, the better the design. We encourage stream-of-consciousness voice notes because they tap into deeper desires — emotion, visualization, dream — that forms never capture.
01
Forms → conversation
Traditional software asks you to fill in fields. Murmurgo asks you to describe the trip. The AI extracts structure from natural language — dates, budget, accessibility, emotional intent — and feeds it into the design pipeline.
02
Brochures → transactions
Every itinerary is a Shopify product with a price and a checkout. Not a PDF. Not a link to a subdomain. A transactable object on your own domain.
03
One design → four surfaces
The same canonical data renders differently for the client, the sales team, the operations staff, and the traveller. Change one source, all four update.
What it is
Murmurgo is three things working together.
01
Distilled expertise.
The work of an itinerary designer — sequencing, costing, supplier selection, constraint detection — encoded as a governed pipeline of skills that runs on current-generation language models. Built specifically for African itinerary design; applied the same way every time.
02
A curated data layer.
Places, supplier content, photos, panoramas and documentation — assembled and ranked for itinerary design.
03
A live language model.
A model works with your client in natural language inside that governed pipeline — refining the brief, surfacing options, answering questions. The expertise and the data constrain what the model can say.
Itineraries that book
A Shopify product — priced, currency-correct, transactable on your domain.
The output of the pipeline is a Shopify product — priced, in your traveller's currency, and transactable on your domain.
Two checkout paths run from one store. Consumer-card payment for the direct traveller, with Shop Pay converting up to 1.72× higher than standard guest checkout (Shopify's own data). And trade-account checkout — company profiles, payment terms, volume pricing — for agents and operators buying on behalf of clients.
OTAs are part of the distribution landscape. This isn't a story against them. It's on-domain infrastructure — pricing, payment, compliance and data stay with you.
The pipeline
Brief → canonical → four surfaces.
A single client brief becomes a structured itinerary, then four audience-specific renders. The canonical record is one source of truth — customer view, operator dossier, operations run-sheet, and pre-departure pack all render from it, with no manual reformatting.
The pipeline · brief to four surfaces
Brief in. Four audience-specific surfaces out.
Distilled expertise runs as a governed sequence of skills. A curated data layer narrows against the brief. A consultant confirms. One canonical record renders to four live Shopify surfaces — same data, four audiences, no manual reformatting. Switch itineraries to see how the same pipeline shapes two very different trips.
A governed sequence — applied the same way, every time
Stages lift in sequence as the pipeline runs. Reduced-motion shows the completed state immediately.
Capturing brief
02Constraint detection
Constraint Detection
Dates, budget, pace, accessibility, emotional intent — structured into context.
Dates17 nights
PriorityRomantic routing
ConstraintNo long drives
03Validation gates
Validation Gates
The Plekify data layer narrows through region, dates, and fit before ranking begins.
450,899
Plekify data layer · places indexed
04Rank + route
Rank + Route
Candidates scored against the brief and sequenced into a coherent shortlist.
Ranked against the brief
98.6 fit
05Human review
Validation
A consultant reviews the shortlist, checks special conditions, confirms publishable.
Consultant confirms routing pace and private dining feasibility.
06Rendered outputs
Rendered Outputs
One validated itinerary fans out into four live surfaces — same data, four audiences.
WETU enrichment
applied to the selected properties
Photos · 360° panoramas · descriptions · documentation — applied to each selected property
Rendered outputs
One design. Four surfaces. Two live examples.
The pipeline produces one canonical itinerary. That design fans out into four audience-specific views — each a Shopify product with real pricing and checkout. Switch between two production reference itineraries to see the same four-surface system rendered for different briefs.
Mosaic Lagoon Lodge · Bushmans River Hotel · Garth's Place · ...
Client-facing
Itinerary Presentation
A three-pane workspace: journey timeline, route maps, day-by-day narrative, inspector panel with 14 templates, and a booking console with per-person pricing.
Pre-Departure Pack
Client-facing pre-trip document: trip summary, packing advisory, health/visa guidelines, emergency contacts, departure plan, readiness checks.
A three-pane workspace: journey timeline, route maps, day-by-day narrative, inspector panel with 14 templates, and a booking console with per-person pricing.
Pre-Departure Pack
Client-facing pre-trip document: trip summary, packing advisory, health/visa guidelines, emergency contacts, departure plan, readiness checks.
Supplier-facing
Operator Dossier
Confidential internal document: executive summary, QMD evidence, route feasibility, compliance, commercial structure. Marked CONFIDENTIAL.
Operations Brief
Printable run-sheet for guides and drivers: property contacts, routing, vehicle allocation, dietary log, accessibility, emergency services.
Confidential internal document: executive summary, QMD evidence, route feasibility, compliance, commercial structure. Marked CONFIDENTIAL.
Operations Brief
Printable run-sheet for guides and drivers: property contacts, routing, vehicle allocation, dietary log, accessibility, emergency services.
Thulamela Royal Lodge · ...
Client-facing
Itinerary Presentation
A three-pane workspace: journey timeline, route maps, day-by-day narrative, inspector panel with 14 templates, and a booking console with per-person pricing.
Pre-Departure Pack
Client-facing pre-trip document: trip summary, packing advisory, health/visa guidelines, emergency contacts, departure plan, readiness checks.
A three-pane workspace: journey timeline, route maps, day-by-day narrative, inspector panel with 14 templates, and a booking console with per-person pricing.
Pre-Departure Pack
Client-facing pre-trip document: trip summary, packing advisory, health/visa guidelines, emergency contacts, departure plan, readiness checks.
Supplier-facing
Operator Dossier
Confidential internal document: executive summary, QMD evidence, route feasibility, compliance, commercial structure. Marked CONFIDENTIAL.
Operations Brief
Printable run-sheet for guides and drivers: property contacts, routing, vehicle allocation, dietary log, accessibility, emergency services.
Confidential internal document: executive summary, QMD evidence, route feasibility, compliance, commercial structure. Marked CONFIDENTIAL.
Operations Brief
Printable run-sheet for guides and drivers: property contacts, routing, vehicle allocation, dietary log, accessibility, emergency services.
The four surfaces
Each surface is a real Shopify product view, from one canonical record.
Same data, four audiences — no manual reformatting.
| Surface | Audience | Purpose |
|---|
Two live reference itineraries — full-price, real-client products on a live Shopify storefront.
The four surfaces
One canonical design. Four rendered surfaces.
Every itinerary produces four audience-specific views from a single data source. No duplication. No divergence. Change the canonical JSON, all four update.
Itineraries, enhanced
A WETU itinerary, enhanced and made bookable on Shopify.
AI encodes commentary, deep-links, and operational detail into your WETU itinerary, then publishes it as one Shopify product — four audience surfaces, one canonical record, transactable.
Reference itinerary
Hoffmann Anniversary
The traveller
Itinerary Presentation
The bookable, priced, visual itinerary — the storefront product the traveller sees and buys.
View →
The agent / DMC
Operator Dossier
Supplier terms, costing, margin, constraint evidence. Confidential — not client-facing.
View →
The traveller, pre-trip
Pre-Departure Pack
Final pack — documents, packing, climate, emergency.
View →
The data layer
Four sources, layered.
The data layer · curated narrowing
A governed data layer, narrowed to the brief.
Place data, supplier source, photos and documentation — assembled and quality-ordered by people who build African itineraries. Not a scraped index. The same governed universe narrows differently for every brief, and lands on the real properties in the real itinerary.
The shape of curation
Each tier is a real constraint applied to the Plekify data layer. The funnel narrows from 450,899 places (Plekify data layer) to the real shortlist — five for Hoffmann, three for Osei. WETU is the enrichment layer applied to the finished shortlist, not the discovery universe.
Place data.
Google Maps as the outer bound — name, location, rating, review volume, photos. The broadest discovery layer for African places.
Property enrichment.
Supplier-sourced descriptions, facilities, rooms, panoramas and documentation. Deep, structured, human-maintained content.
Quality-ordered discovery.
Places and properties are ranked by rating and review volume — the same ranking a senior planner applies mentally.
WETU enrichment — you don't start from scratch.
If you're a WETU customer, Murmurgo reads the itineraries you already have — past trips, sample packages — and enriches them: transactable, multi-surface, ready for checkout. Your existing library becomes a storefront. WETU is a partner and an input, not a system to rip out.
Data infrastructure
The broadest property dataset in African travel.
Discovery layer
Google Maps is the global outer bound of place data — every business, every location, every review. We have indexed that universe across 38 African countries: nearly half a million places, over a million reviews, structured and searchable. This is the largest selection of property data available for African itinerary design. It is not limited to properties that are customers of any single platform. Every property exists. Every property is discoverable.
Enrichment layer
For a finished itinerary — the document a client receives — richer content matters. WETU provides exactly this: high-quality photographs, 3D panoramas, detailed property descriptions, and real travel documentation sourced directly from the properties themselves. Businesses that are already WETU customers can enrich their itineraries significantly, with each property and activity gaining access to WETU's full multimedia library.
Better together
Murmurgo offers a competing product to WETU's itinerary builder. But it works better together with WETU. The discovery layer finds every property; the enrichment layer makes the final itinerary visually rich and client-ready. Operators who use both get the broadest discovery and the deepest content — comprehensive itinerary design with full multimedia, not limited to any single platform's customer base.
Why Shopify
The stack Murmurgo runs on.
And the one it replaces: WordPress plus a payment link, with the traveller leaving your domain to pay.
Agentic commerce leads.
Shopify's Catalog API structures product data and syndicates it to the AI surfaces travellers already use — ChatGPT, Copilot, Google AI Mode, Meta. Shopify reports AI searches powered by Catalog convert at 2× the rate of those using scraped data (their own claim; methodology not disclosed).
Source shopify.com · Spring '26 EditionB2B on more plans.
Company profiles, payment terms, volume pricing and up to three B2B catalogues now on Basic, Grow and Advanced — not Plus-only.
Source changelog.shopify.comMulti-currency, including ZAR.
Sell to USD, EUR and GBP travellers with locally displayed, duty-and-tax-aware pricing from one store.
Source help.shopify.com · Managed MarketsRedesigned checkout.
Tighter layout; payment methods dynamically ordered to convert.
Source shopify.com · Spring '26 EditionSouth African compliance, built in.
Cart and Checkout Validation Functions now validate billing address and PO number server-side (API 2026-04). That's the agent-vs-principal / commercial-status rule enforced at the checkout layer — not bolted on.
Source shopify.dev · API 2026-04Standard storefront events.
A shopify:-namespaced event layer lets apps and AI agents observe and drive commerce interactions without DOM scraping.
Source shopify.dev · Standard eventsThe commerce layer
Itineraries are products. Shopify sells products.
Every itinerary is a Shopify product with a price, a variant, and a checkout. Not a brochure. Not a PDF. A transactable commerce object on your own domain.
And the foundation the checkout layer inherits — by default, on every store.
PCI-DSS Level 1
by default
99.9%
platform uptime
$92.0B
GMV, Q3 2025
Pricing and the pilot
Three things to know up front.
Transparent ranges, not a black box. Exact figures are confirmed at scoping.
Implementation is founder-led.
A scoped, fixed-fee setup — not a self-serve sign-up. Pilots are priced to clear procurement quickly.
Platform fees are transparent.
Per-property recurring from roughly R4,500/month. Groups and consortia are quoted per member, with a capped take-rate on itinerary GMV — Murmurgo does better when you do. Final terms are confirmed at scoping.
The pilot is structured, not discounted.
A paid, scoped engagement with a 90-day opt-out and pre-signed annual contract. Pilot fees credit against year one on hitting agreed success metrics. In exchange: a named case study and a reference call.
Property collection (3–8 properties)
~R4,500 / property / month + capped take-rate
- One brand, one storefront.
- Base subscription + modest take-rate + cap.
- Full four-surface rendering + Shopify checkout.
- WETU library read and enriched in place.
DMC consortium (5–15 operators)
~R15,000 / member / month + capped take-rate
- Per-member licence + lower take-rate + higher cap.
- Full Tourplan / Xero / Sage integration.
- Up to three B2B catalogues for trade segments.
- Unlimited view-only seats for sub-agents and suppliers.
Enterprise (group-wide)
Custom consortium pricing · Talk to us
- Custom from a higher base; consortium per-member pricing.
- Group-wide rollout and governance.
- Dedicated integration scope.
- Named case study + reference call.
A direct booking is, on Kalibri Labs' figures, around 12.5% more profitable than the OTA equivalent. Across a season's bookings, that retained margin more than covers the annual platform fee — one new itinerary sale makes the case obvious.
Apply
The pilot is curated — two to three customers in the first cohort.
Ideally one property group and one DMC. We read every application.
Pre-empting the three questions we hear most
Switching and data loss
Your WETU library is read, not migrated — you keep WETU. Existing itineraries are enriched in place; nothing is deleted.
Integration
The pipeline writes to Shopify's standard product, metafield and Functions APIs. Tourplan, Xero and Sage sync via documented endpoints.
Commercial status and VAT
Billing address and PO validation run server-side at checkout. The agent-vs-principal rule is enforced at the checkout layer, not in a spreadsheet.
Trust signals
Beta access
Apply for beta access.
We are selecting one pilot client to implement the full pipeline: AI itinerary design, four-surface rendering, Shopify checkout, and WETU content integration. Bespoke implementation by the founder.
We review every application personally. You will hear from Matthew within 48 hours.