Holiday Extras is rethinking what it can build with Codex
With Codex, Holiday Extras is turning customer expertise into better apps, higher booking conversion and new possibilities for growth.
32.5%
increase in open-to-book conversion on the rebuilt Holiday Extras app.
10%
reduction in app error rate, from 19% to 9%.
6 months → 2 weeks
to achieve working app rebuild.
Holiday Extras(opens in a new window) helps customers search, book and manage trips. For CEO Matthew Pack, becoming an AI-based travel technology business means changing how the whole organisation works: giving the people who understand customers and operations the ability to improve them directly.
Company-wide ChatGPT Enterprise access established the foundation. Codex extended it into building software, from customer-facing apps to internal systems. Pack builds with Codex himself, using that experience to understand what colleagues need and raise his expectations of what the business can achieve.
“This isn’t about one or two people anymore. It’s about taking the entire team on the journey.”
Holiday Extras had spent years developing its mobile apps(opens in a new window). Major changes to its main app centred on an annual release. With Codex, it challenged itself to rethink the experience and how it was built.
Within two weeks, the team had a fully working rebuild, achieving development progress that previously would have taken over six months. Liam Thompson, VP of AI Innovation and formerly Head of Design, worked with engineers to translate his understanding of customers into software. Small, documented packages created reusable foundations, while engineers focused on architecture, APIs, security and development workflows.
The rebuilt app offered a more personal experience informed by customers’ bookings, trips and history. Codex enabled iOS and Android features to be developed together, bringing improvements to both platforms without Android following behind.
“It’s not a question of can I build it anymore. It’s how I would build it.”
Higher conversion: open-to-book conversion on the main app increased 32.5% following the rebuild.
Fewer errors: the main app’s error rate fell from 19% to 9%.
More frequent improvements: the main app moved from an annual major release to weekly releases in the first month after the rebuild, each containing multiple new features.
Faster iteration: Countdown moved from adding a feature roughly every six months to multiple additions a week.
Conversion and error-rate results compare the previous and rebuilt main app across different periods.
Faster development also gives customers more influence. When a Countdown customer complained about a missing list view, an agent prepared a fix. Liam Thompson approved it, the update was submitted and the customer changed their one-star review to five stars.
The same emphasis on acting on customer insight extends into operations. David Lee, Director of Innovation and Growth, describes an automated system that turns airport-transfer customer chats into actionable tickets, helping the team identify problems across its supplier network. Customer insight becomes work the team can act on.
“So, you can build things more quickly. You can test them more quickly, and you can learn from them because it can understand the data.”
Matthew Pack describes giving everyone permission to try AI, then actively encouraging them to use it. That permission comes with guardrails. Employees have access and encouragement to experiment; engineers build the packages, security controls and workflows that make that experimentation practical.
David Lee sees the effect in an experienced accountant who rebuilt finance systems around his understanding of the business and feedback from colleagues. The opportunity reaches beyond learning to code: people can apply years of operational knowledge directly to the tools they use. Leadership’s role is to make that possible, support experimentation and help successful ideas spread.
Holiday Extras values Codex’s ability to explore ideas, reason through complex problems, propose a plan and deliver independently. Liam Thompson describes being able to question an approach and receive suggestions throughout the work.
The wider partnership connects building with everyday work. ChatGPT Enterprise supports employees across the business, while OpenAI models power the app’s personalised experience. dots adds coordination and insight: Thompson recalls his dot proactively preparing a performance report after an app launch. David Lee values model quality and efficiency alongside the relationship behind the products.
“I love how OpenAI thinks differently. I love how you are open to feedback as well. We have a fantastic two-way relationship.”
Pack’s ambition is for more of Holiday Extras’ systems to be administered through AI, with employees directing the work and shaping the outcomes. The app rebuild demonstrates the potential: better booking performance, fewer errors and capacity to keep improving.
That gives Holiday Extras a new basis for deciding which opportunities to pursue across its products and markets. As more people can act on their expertise, Pack is asking the leadership team to reconsider the scale of its ambitions.
“Our dreams are not hard enough, clearly. We need to go back and rethink what we’re capable of doing.”
