EyeforTravel San Francisco 2018

April 2018, San Francisco

Planet 'total offer': the coming revolution in airline merchandising

When three global technology trends combine with two industry-specific trends, can the planets align to transform the online relationship between airline and travellers? Tom Bacon investigates

For decades, airlines have worked to perfect network- and flight-based revenue management, forecasting demand by origin-destination market by fare level and employing sophisticated optimisation models to allocate scarce seat capacity across the network. Airlines may now seek instead to optimise individual customer interactions. 

The opportunity is revolutionary for airlines: designing and pricing the product around individual customers rather than the existing legacy methodology. The new process – sometimes termed ‘total offer optimisation’ - positions airlines more like real e-merchandisers. It means giving individual travellers the right product (not just the seat but the entire travel experience) at the right price (an attractive price for a bundle of services) at the right time. At Amadeus’ revenue optimisation conference in Monaco, ‘offer management’ was a key topic.

5 steps to totally optimising offers

’Total offer optimisation’ can’t realise its potential without five capabilities. Each of these has emerged over the past five to ten years bringing ‘total offer optimisation’ now much closer to reality. Two of them (ancillary revenue and NDC) are industry-specific breakthroughs. Three of them (big data, machine learning, and the cloud) are broad technology initiatives that complement the industry breakthroughs. But none of the above capabilities existed in the same way as ten years ago. The development and expansion of each of them now positions the industry for the most dramatic change since American Airlines introduced yield management in the early 1980’s.

These trends mark the beginning of a new era in the travel industry and, potentially, the beginning of a revolution in travel merchandising. 

Let’s first take a closer look at the three global technology trends.

  • Ancillary revenue streams 

In 2008, U.S. airlines introduced charges for checked bags; worlwide ancillary revenue has grown every year since. According to Ideaworks’ 2017 report, the top ten airlines in ancillary revenue generated $2.1 billion while in 2016, nine years later, the ancillary revenue reported by the top ten airlines had grown more than ten-fold – to more than $28 billion. Some of the leading airlines in ancillary revenue have almost as much revenue in ancillary options as in their base fares. 

  • Bigger data

Data has gotten bigger and bigger. The IDC Digital Universe estimated that data would grow by 50 times in the current decade, from 2011. For airlines, bigger data means:

  • More personalised data, including, potentially, social media information and travel preferences.
  • Combined search and booking data. Combining search data with booking data multiplies data available for analysis and also opens up new merchandising opportunities.
  • Combined base fare and ancillary purchase data. Initially, ancillary purchases were tracked separate from the flight booking. Airlines have worked to develop more integrated databases of customer purchase behavior.
  • Market data. Competitive offers, available on a real-time basis, form the set of choices against which an attractive personalised offer can be developed.
  • Machine learning

As data has grown, it has quickly outstripped the capacity for teams of analysts to manually analyse all the micro patterns. Machine learning is critical to exploiting the new data available in the airline industry; machine learning, not teams of analysts, is critical for developing more personalised approaches to product packaging and pricing.

IBM’s Watson is one example of the development of machine learning. In 2011, Watson beat a human expert at the Jeopardy! Game. It has since been deployed to medicine, weather forecasting, and a variety of commercial applications. Machine learning, as exemplified by Watson, is now ripe for use in airline e-merchandising.

New ancillary revenue streams, bigger and bigger data, and machine learning capabilities form the basis for new total offer optimisation. However, distribution of customised offers to customers across the globe in all channels is not possible based on decades-old distribution technologies. Two new distribution-related capabilities complete the emerging opportunity of total offer optimisation. These are:

  • Distribution – The cloud 

Responding to a customer’s search with a personalised offering would add unacceptable time and complexity to customer travel searches if airlines still relied on centralised data centres. Historically, airlines have relied heavily on caches to improve response time - but caching means the response isn’t real-time and can therefore become out-of-date or inaccurate. The cloud, on the other hand, can facilitate accurate real-time responses even in the face of dramatically increasing volumes. The cloud began in commercial form with amazon’s EC2 in 2006 but continues to grow in applications. Airlines and travel distributors use the cloud now to offer both accuracy and speed, and to synchronize search responses across channels/regions.

  • Distribution – New Distribution Capability  (NDC)

Five years ago IATA, an international airline organisation, developed a new XML-based protocol for distribution of airline content to third parties. Known as NDC, the move came in response to the need for the industry to offer broader personalisation and merchandising capabilities that would be fully available on airlines’ own websites but more limited on third-party sites. So far, it has been in test mode by a number of carriers, and today 41 airlines and 11 aggregators are certified at a Level 3 (‘offer management’).

Tom Bacon has been in the business 25 years, as an airline veteran and now industry consultant in revenue optimisation. He leads audit teams for airline commercial activities including revenue management, scheduling and fleet planning. Questions? Email Tom or visit his website

Related Reads

comments powered by Disqus