Case study

How we scaled easyJet holidays' intelligent pricing capacity by 40x

40x

capacity increase

a new

testing approach

18

months look-ahead

The upshot

Revenue management at scale with state-of-the-art testing technology

easyJet holidays needed to scale their pricing operations from hundreds of thousands to millions of daily pricing decisions—up to 40 times their existing capacity—whilst maintaining trader oversight and control. The challenge wasn't just about processing more pricing decisions, but making intelligent, nuanced changes to pricing that would be impossible to manage manually. Datasparq developed a comprehensive pricing model that generates strategic dynamic pricing recommendations rather than fixed prices, enabling traders to make informed decisions across millions of holiday combinations up to 18 months in advance.

One of the major innovations in the project came through implementing synthetic control methodology—a technique introduced to easyJet holidays for the first time. This technique solved the fundamental challenge of testing pricing models in competitive marketplaces. This innovation enabled easyJet holidays to accurately measure the impact of pricing changes with confidence that results were robust to outside interference, whether from competitor reactions, marketing activities, weather events or other external factors - something traditional A/B testing cannot achieve in marketplace environments.

‍Key benefits:

Expanded intelligent pricing capacity by 40 times, enabling sophisticated pricing changes previously impossible due to manual capacity constraints.
Developed innovative marketplace testing methodology using synthetic control techniques, creating a reusable methodology that easyJet holidays can apply beyond pricing—including testing website display changes, discount effectiveness, marketing campaign impact and customer responses to bundling options.
Achieved exceptional stakeholder buy-in, with trading teams integrating the solution into their workflow as the new standard approach to pricing decisions.

AI use case mapping for Leeds Bradford UK Airport

Our value and impact driven approach

AI strategy workshop output for UK regional airport

The opportunity

Scaling pricing operations in a fast-growing holiday business

easyJet holidays was experiencing rapid growth. This expansion created an urgent operational challenge: whilst the existing team of traders could make broad pricing changes across their inventory, they needed to reach millions of sophisticated daily pricing decisions to cover their expanding inventory effectively.

The traditional approach relied on human traders making pricing decisions for holiday packages—combinations of flights and hotels—with varying markup strategies. The team could apply broad changes, but the real challenge was making nuanced, data-driven pricing decisions that considered individual package performance, seasonal patterns, competitor positioning and demand signals whilst maintaining the oversight and control that traders required.

The complexity multiplied when considering booking windows extending 18 months into the future, creating millions of potential pricing scenarios that required sophisticated analysis rather than broad approaches.

The challenges we faced:

Scaling pricing operations up to 40x beyond current capacity whilst maintaining trader oversight and control.
Testing pricing strategies in competitive marketplace environments where traditional A/B testing fails.
Managing ambitious scope across all destinations and booking windows simultaneously.
Moving beyond broad pricing changes to sophisticated, data-driven recommendations that traders could trust and control.

The solution

Building intelligent pricing strategies with breakthrough testing methodology

Datasparq built a decision support system that blends human expertise with AI-driven insights to optimise pricing strategies. The model predicts conversion rates and generates contextual pricing recommendations—always within trading guardrails—while adapting to factors like booking performance, time to departure, historical patterns, and competitor activity, empowering traders to make intelligent, scalable pricing changes that would be impossible to manage manually.

01

Solving the marketplace testing problem

The project addressed a fundamental challenge in pricing model validation. Traditional A/B testing fails in marketplace environments because splitting all holidays in one category into train and test results in contamination effects. The size of the effect of any truly 'good' test prices from a model output may not be measured accurately as traders would quickly adjust their prices to follow suit. Likewise, the size of the effect of any truly 'bad' prices that happen to be lower may not be measured accurately, as we may just be self-cannibalising.

AI and data solutions development partner

Synthetic control for ab testing AI pricing

02

Implementing synthetic control methodology

Datasparq implemented synthetic control techniques to create valid testing frameworks.

Instead of comparing similar destinations directly, the methodology constructs synthetic control groups using weighted combinations of multiple destinations that historically behave like the test destination.

03

Validating model performance

The synthetic control approach was tested across 16 destinations, allowing measurement of model impact without marketplace interference.

The methodology proved both technically sound and powerful enough to detect meaningful performance differences, providing easyJet holidays with reliable testing capabilities.

data and ai solutions for airlines and travel

1. Proof of concept

First testing with a supermarket depot demonstrated capabilities beyond established industry tools, impressing transport leadership with performance that surpassed existing SaaS solutions.

2. First deployment

Applied to a major retailer's largest and most complex depot, exploring delivery types and time window flexing. This phase delivered substantial annual savings that continue to benefit operations today.

3. Extended capabilities

Extended to a more challenging network with multiple depots, handling unionised workforces, two-person deliveries and a wider variety of vehicle types. This phase demonstrated the optimiser's adaptability to different operational constraints.

4. Enterprise-ready

Standardised data formats and pipelines to create a system applicable across diverse customer networks, with further optimisation for a home improvement retailer adding wait time and compactification features that doubled performance.

Interactive example

Try it for yourself

Explore how the science works in this PlayML data science notebook

The impact

Breakthrough testing methodology with strong business adoption

The project delivered genuine transformation in easyJet holidays' pricing capabilities through the successful implementation of synthetic control methodology that solved their marketplace testing challenges. The testing approach proved both technically robust and comprehensible to business stakeholders, with easyJet holidays adopting the methodology for their internal testing dashboard. Initial testing produced encouraging results—and demonstrated that the testing methodology was sensitive enough to detect real performance differences.

The project achieved strong stakeholder engagement with trading managers and senior traders actively participating in technical discussions about model functionality, reasoning through why the system made specific recommendations and planning future improvements. The comprehensive handover programme included extensive documentation in addition to six detailed sessions covering model operation, technical implementation and improvement strategies, with easyJet holidays expressing strong satisfaction with the thoroughness of knowledge transfer.

Key impacts:

Developed a scalable pricing framework that expanded easyJet holidays’ capacity by 40×, enabling millions of intelligent daily decisions across 18-month booking windows.
Built price sensitivity and forecasting models to predict conversion rates, optimise strategies, and balance passenger growth with margin performance.
Introduced synthetic control testing to measure pricing impact reliably in competitive markets, solving the limitations of traditional A/B testing.
Integrated seamlessly into existing systems, automating pricing updates while preserving trader control with guardrails and operational levers.
Established a comprehensive decision support system that is now fully embedded in easyJet holidays’ revenue management infrastructure, with strong adoption by the trading team.

Ready to address AI opportunities that deliver real value? Our approach gets you from challenge to solution in weeks, not years.