← Back to the schedule

Three Advanced Analytics use cases in the travel industry

Calendar icon

Wednesday 14th

Time icon

17:50 | 18:30

Location icon

Theatre 18


Keywords defining the session:

- Advanced Analytics

- Use Case

- Travel

Takeaway points of the session:

- Advanced Analytics is one of the more promising source of value creation in the travel industry

- Artificial Intelligence has proven to be impactful in Customer Acquisition, Demand Optimization and Improving Customer Experience


Over the last years, the travel industry has been profoundly disrupted by digitalization. Digital technologies have changed the way in which people research content, compare options and book travel.
All these new digital technologies have produced an exponential growth in data generation, for instance a middle-sized hotel chain might monthly gather data from millions of visits to its own web page or an airline can gather up to hundreds of terabytes of data coming from the sensors of the airplane engines in just a single long-haul flight.
With the appearance of new capabilities such as Big Data Technologies and Artificial Intelligence, it is now possible to capture the full potential of this data. Machine Learning, Natural Language Processing, Image/Video Recognition or Parallel (Real time) Processing are reaching the sufficient maturity to be considered as one of the most promising sources of value creation in the travel sector.

Through the customer value chain, Advanced Analytics related use cases are starting to emerge:
– Customer Acquisition (advanced prospecting and smart re-marketing)
– Demand Optimization (Dynamic Pricing and Cancellations forecast)
– Customer Experience (Personalization and NPS understanding through NLP)

Customer Acquisition: impacting the Marketing ROI through micro audiences

With the appearance of Online Travel Agencies, the customer acquisition does not only consist in convincing the customer to travel with you, but also in convincing him to purchase the travel in your own website. In that context, optimizing the marketing budget to only invest in the most promising audiences is decisive.
Predictive models based on historical navigation patterns can be used to score every lead entering the travel company’s site and create qualified audiences.
Using Machine Learning to construct such advanced audiences and improve the re-marketing strategy has proven very effective, allowing an up to 50% reduction in marketing investments keeping stable the number of conversions.

Optimizing the Revenue: Dynamic Pricing

Being able to estimate the conversion probability of each customer (depending on several factors such as price) is the first step for a data driven pricing strategy. Machine Learning algorithms have proven to be very effective in such estimations and therefore in better understanding customer’s willingness to pay.
Moreover, more complex forecasting algorithms can increase price reactivity to global trends, limited offers or special events and techniques like self-learning algorithms (such as Deep Reinforcement Learning) have proven to be very effective in this field.
In such a competitive sector, revenue management has always been a strategic domain and dynamic pricing is seen as its next evolution. Applying such techniques has already proven to be able to boost revenue by several percent points compared to traditional revenue management systems.

Improving the NPS through Natural Language Processing

Customer surveys are a very common instrument in the travel industry to ask for customer’s feedback. Nevertheless, due to the unstructured nature of these forms, frequently only a part of this information can be translated into corrective actions impacting the Net Promoter Score.
Natural Language Processing techniques such as Sentiment Analysis or Topic Detection are being applied in the travel industry to dynamically understand which services are being well or bad perceived by the customer and automatically recommend corrective actions to boost the NPS.
Using such techniques can not only quickly produce very valuable and actionable insights, but also reduce the personal costs dedicated to such tasks.