Corp Travel …So Why Can't I Just Book My Own Trip?

From the Travel Tech Podcast newsletter for episode 30, Why Can't I Just Book My Own Trip? Dual Identity, Dreadful Booking Tools and the Real Job of Corporate Travel.

Caitlin Gomez signs up for rideshare with her personal Gmail, like everyone else. Somewhere in a corporate travel program she is also a Head of Growth on the speaking circuit, and nothing in the supply chain reliably connects the two. That gap, between the person who books a holiday on their phone in thirty seconds and the business traveler who is told which tool to use and which channel to book through, is what this week's conversation is really about. I wanted it in front of those of you who read this because it explains why corporate booking tools have been so miserable for so long, and what suppliers should do about it.

She started on a trading desk, then moved into travel technology through Staples, then into partnerships at Lyft and CLEAR, then spent time at GBTA building out Innovation Row before joining a company that calls itself middleware on purpose. Apple Podcasts, also streaming on Spotify and YouTube.

Her answer to my opening question, why can't I just book my own trip, was blunt: you can, but do you want to be reimbursed? The program exists to keep people safe, control cost and get them home, and every corporate booking tool you have ever hated was built to enforce that. The mistake is treating travelers as if they forget who they are when they walk into the office. They use Uber and ChatGPT every day. The industry keeps asking them to become someone else the moment a trip is for work.

The gap between a demo and a departure manifest

That tension has a structural cause. A global corporate travel program might need to service 127 countries with different data reconciliation and standardization requirements, and no two TMCs or GDS setups handle it the same way, which is why Amgine builds itself to fit into existing infrastructure rather than asking enterprise clients to fit into theirs.

AI adds a new layer of pressure on top of that complexity rather than removing it. In most deployed setups, an AI agent is only as good as the antiquated GDS and API layer it has to plug into, and it works well for interpreting intent and layering policy preferences, but it still breaks the moment a real disruption hits and a human has to take over.

Caitlin Gomez, Head of Growth at Amgine, joins Alex to unpack the company's middleware positioning, its omnichannel booking flow across email, chat, Slack, and WhatsApp, the industry's blind spot on ground transportation, and what the recent BCD investment actually signals about the partnership.

Key Takeaways

• Program design beats traveler experience: Corporate travel optimizes for risk mitigation, cost control, and duty of care first, and a smooth consumer style booking flow is secondary to those constraints.
• Middleware over consumer brand: Amgine avoids becoming consumer facing and instead integrates directly into existing GDS and TMC infrastructure, including access through a company's own pseudo city.
• Attribution is the industry's open problem: Consumer platforms rarely have a reliable way to connect a traveler's personal persona to their corporate one, which weakens the case suppliers can make to a CFO.
• Ground transport remains a duty of care gap: Programs track travelers well in the air and at check in but lose visibility during transfers, which is where rideshare integration adds real value.
• AI needs domain fundamentals underneath it: Teams that lean on AI as a substitute for GDS and operational knowledge get stuck the moment a system breaks, since the technology still depends on a human who understands why.

🔧 Three to Build

Caitlin Gomez's argument boils down to one thing: corporate travel breaks where humans hand off to systems and systems go silent.

Wire your booking inbox directly into the GDS, no human relay.

Most TMCs still have an agent read an email, rekey the details, and push the reservation. That gap is where errors, delays, and liability accumulate. Robotic process automation can sit between the inbound request, whether it arrives by email, Slack, or WhatsApp, and the GDS, completing the reservation without replacing any existing infrastructure. The corporate client gets speed. The TMC keeps control.

Catch: the edge cases are not rare. Fare rules, unused ticket credits, and policy exceptions will each need a defined fallout path before you go live or the automation just moves the queue.

→ Automate corporate booking processing

Parse multi channel group air requests and route them to completion without a coordinator.

Group travel requests arrive fragmented: a forward from procurement, a spreadsheet attachment, a follow up on WhatsApp. Natural language processing can read across those channels, extract the structured data, and hand a clean request to the booking workflow. That removes the coordinator as a single point of failure and cuts the window between request and confirmed seats.

Catch: NLP accuracy is high enough in testing and brittle enough in production. Train on your actual channel mix, not benchmark data, or the model will fail on the requests that matter most.

→ Automate group travel inquiry processing

Surface a ground transport recommendation at the moment a flight lands, not when the traveller asks.

The stretch between the arrivals hall and the hotel is where corporate duty of care goes dark. The booking platform knows the flight. It often knows the hotel. It almost never closes the loop on what happens in between. By joining flight schedule data with travel patterns and a rideshare or transfer API, a product team can push a confirmed ground option before the traveller even clears customs.

Catch: the value is in the confirmation, not the recommendation. A suggestion the traveller still has to action themselves does not restore visibility. You need a completed booking, not a link.

→ Predict and recommend ground transport options

Alex Brooker, Host of the Travel Tech Podcast, Founder of Airside Labs