The buildout
Western economies are rebuilding domestic energy and resource capacity — gas processing, critical-minerals extraction, generation, transmission — and the deposits and sites are mostly remote. Remote capacity means rotational workforces: crews moved on schedules the operation sets, not schedules an airline sells. The demand is not only Western — energy capacity is being locked in globally this decade — and that carries a second consequence for Western operators: the systems that run critical workforce logistics increasingly need to sit in Western jurisdictions, on Western clouds.
This holds whichever energy future arrives. A renewables-heavy grid needs critical minerals mined in remote places and transmission built across them. An energy-security posture needs gas produced, processed, and shipped from home territory. Electrification and AI are driving the largest electricity demand growth in decades, which needs all of the above. Every path through the 2020s runs through remote-site construction, and remote sites run on rotational workforces. The thesis does not require taking a side in the energy debate; every side builds.
Each new site is a workforce-mobility problem that lasts the life of the asset. Construction crews first, then decades of operating rotations. The mobility spend is committed the day the project is sanctioned; the only question is how much of it is wasted.
The gap
Workforce mobility for the buildout runs on systems designed for something else. Corporate travel tools were built for office workers and expense compliance. Airline systems were built to sell the airline's own inventory at maximum yield. Rotational logistics sits in a no-man's land between the two: organisations hand it to carriers and charter brokers who have no systems for managing a client's demand against leased capacity — no demand-shaping, no visibility of budget leakage, no passenger experience when things go wrong.
The result is measurable. Comparable small-fleet operations average roughly 40% seat utilisation on capacity that is paid for whether or not anyone boards. The waste is not a procurement problem; it is a missing-software problem.
Disruption is the operating condition
In this vertical, disruption is not the exception — weather, route closures, medical evacuations, and pandemics are the normal operating environment of remote-site work. The COVID record is the proof case: scheduled commercial aviation collapsed by more than 90%, while resource-sector rotation traffic intensified — charter air bridges, essential-industry exemptions, relocated workforces. Organisations that controlled their own transport kept operating; organisations that depended on commercial schedules stopped.
When the system fragments, spend migrates from seat volume to coordination complexity. The coordination layer is what a platform monetizes — and black-swan interruptions to commercial travel push more organisations toward leased capacity each cycle, not fewer.
The thesis
The winner in this category is whoever builds demand-side management on infrastructure that holds under surge.
We have found no product that even advertises leased-travel management. The capability exists today only as a bespoke one-off — commissioned from a Big-Five consultancy at the cost and timeline that implies — or it does not exist at all, and the work falls back onto spreadsheets and an airline's call centre. UnityTrip occupies the space between that one-off build and the conventional travel SaaS, whose fixed roadmap cannot bend to a single company's, country's, or market's particular rules. It is configurable to the operation like the bespoke build, and operated as a product like the SaaS. That middle is the whole opportunity, and as far as we can find, no one else is building it.
The booking layer shift
We expect the booking interface itself to commoditize. Agentic booking assistants — AI agents acting for individual travellers — are already emerging for commercial travel, and their adoption case in the enterprise is direct: they replace team travel arrangers, a visible cost line. We expect rotational travel to follow.
When an agent books on a traveller's behalf, the scarce capability is not the interface. It is the layer the agent must call: who is entitled to which seat, in what priority, against which quota, with what consequence for a no-show — decided deterministically, identically every time, and explainable when the answer is no. UnityTrip already runs that layer in production. Our policies are machine-readable by design, the engine is deterministic by design, and an agent is just another client of it.
That is why this shift is our tailwind rather than our risk: we do not need to own the booking interface, and we intend to partner with agentic front-ends rather than compete with them. The interface can change hands; the allocation layer is where the operational truth lives.
Proof one: the demand-side machinery exists, and it is ours
Neither incumbent category — corporate-travel software nor airline systems — has machinery for shaping demand against leased capacity. UnityTrip's is in production: a priority matrix ranking passenger types and trip reasons, per-rule quotas and booking windows, penalty points scaled to departure proximity, displacement rules for full flights, and go-show automation that fills a freed seat minutes before departure. The measured result at a multinational LNG producer: no-shows down roughly 50%, utilisation lifted from the ~40% industry average to above 90%, and 80% of logistics previously considered too complex to automate now running as policy-governed workflow. The figures, with methodology, are published in our benchmarks.
The same machinery is public as a free tool: the Leased Travel Policy Builder turns an organisation's rules into a machine-readable policy a booking engine can execute. We published the category's demand-side logic before incumbents had a name for it.
Proof two: architecture built for the worst day
Look-to-book surges correlate exactly with disruption events. A cyclone closes a site, a roster blows up, and every affected worker and coordinator searches simultaneously — many times normal load, at precisely the moment the service must not degrade. Look-to-book ratios and their surge multiples are publicly documented across the travel industry, and agentic booking will push them higher still: agents shop exhaustively and never get tired. A booking system's worst load day is its client's worst operational day, and that day decides the renewal.
UnityTrip is event-sourced and natively distributed — Azure Container Apps over Cosmos DB, work queued and streamed rather than contended in a single database — designed to absorb surges of 10 to 1000 times normal activity. Several platforms claim distributed architecture; a useful test is to read their engineering job advertisements and see what their systems are actually made of.
We have watched the failure mode in production. One rogue automation inside a client's trusted network began hammering the platform: more than 430,000 requests in a single day, spiking past twenty times the normal baseline within hours. An architecture that relies on perimeter security faces a binary choice at that moment, because it cannot see which identity inside the trusted perimeter is responsible — block all of the client's traffic and stay up, or stay open and go down. UnityTrip did neither. Cosmos DB autoscale absorbed it: provisioned throughput of 21,000 request units a second — about 15,000 transactions a second — expandable by a third instantly, at negligible marginal cost and with tuning to avoid hotspots. And the platform identified the rogue identities in real time — down to the email address — so the offender could be isolated without blocking the client. Neither answer is available to a SQL monolith behind a perimeter. In an agentic world, where a misconfigured agent can do exactly what that rogue script did, per-identity observability and cheap surge headroom stop being architectural preferences and become the product.
Posture
UnityTrip is incorporated in New Zealand and runs on Microsoft Azure with regional data residency — critical workforce logistics sitting in a Western jurisdiction, on a Western cloud, as the buildout increasingly requires. A quiet asset for clients who care about vendor neutrality and data sovereignty, stated plainly rather than sold.
The architecture is also the cost structure. Infrastructure as code — BICEP, automated tenant provisioning — makes a new client configuration rather than an implementation project, and automation does the work that legacy SaaS carries as expensive, unwieldy headcount. UnityTrip is a lean layer by construction: low cost per transaction, low headcount per tenant, onboarding measured in weeks.
Because the platform is defined in code, each client chooses where to sit on the cost-and-resilience curve: a lean single-region footprint, or fault-tolerant, geo-replicated, active-active regions that cost more and fail over without interruption. The same codebase provisions either, and a client can move along the curve as its needs change. Across that range, infrastructure runs as much as one to two orders of magnitude below the SQL-backed platforms this category has historically required — a structural cost position, not a promotional discount, and the source of the margin that funds the company without outside capital.
The company is founder and customer funded and cash-flow positive on long-term contracts. We are not dependent on raising; we are selective about capital and partners who understand the structural position. The conversations we want are with people who read this page and recognised the shape of it.
Talk to us
If you think in these terms — the buildout, the gap, the surge problem, the agentic shift — we would like to hear from you.