Machine Learning and New Savings (In Business Travel)

Machine Learning to Realize New Savings in Business Travel
Travel News 18 May 2018

Research suggests that emerging technologies like AI and machine learning could contribute $15.7 trillion to the global economy by 2030 , of which 55% will be due to increased productivity.

AI is “a collective term for computer systems that can sense their environment, think… and take action in response to what they’re sensing and their objectives…As humans and machines collaborate more closely…the transformational possibilities are staggering.”  Machine learning is the term now more commonly applied to predictive analytics.


These days, data is worth a fistful of dollars – and then some. Most corporate travel professionals would agree that best-in-class travel programs are defines by exceptional traveler experiences. They’d also agree that data is the channel through which traveler feedback is gathered, and the analysis of that data delivers the insights that travel manager and TMC can put into action.


So, it follows that predictive analytics can provide the means to scale-up the volumes of data that can be mined and interpreted?  In theory, decision-making becomes better informed and the scope to influence traveler behaviors is better assessed. Oy maybe not. After all, lots of data doesn’t always guarantee better insights. After all, it’s not easy seeing the wood for the trees.  Then there’s the cost of buying-in the analytics and the brain-power to understand it all.


Additional data doesn’t always equate to faster or more insightful answers though. Nor does it necessarily warrant hiring an army of data scientists to deliver those answers. Two years ago, travel buyers were estimated to spend 40 hours per month reconciling expense and payment data .


Machine learning could provide the solution by drastically reducing the time and cost of data analytics, leaving travel managers free to focus on implementing the actions evidenced by travel and expense data.


TMCs investing in machine learning want to provide their clients with the facts to back decisions around travel spend and risk management. AI and machine learning enables travel buyers to understand how money is spent from region-to-region; to inform supplier negotiation and how to influence traveler behaviors.


Perversely, machine learning also enables corporate spend to be channeled back into travel rather than data analytics. Trend analysis and predictive decisions become quicker too, in turn delivering cost savings by that slickness of response.
The financial benefits to travel management of machine learning and AI are both tangible and hidden. For example, machine learning can mitigate traveler risk management by forecasting travel disruption and delays, enhancing the traveler experience through more timely updates.


Large hotel chains, such as IHG and Hyatt, have started leveraging their big data to better serve the customer and predict occupancy levels and room rates . One hotel, as a direct result of investing in predictive analytics increased revenues by 30%; average daily rate (ADR) by 10% and direct bookings by 109%. If it works for the supplier community, it stands to reason the buyer community can benefit too.


Machine learning can also help travel buyers to leverage the evolving nature of travel. AI can not only distinguish between business and leisure travel but can also help corporates to embrace the growing bleisure phenomenon by suggesting sightseeing or restaurant options to travelers or meeting delegates .


It can’t be that simple though, can it? Earlier this year Microsoft published its manifesto for corporate travel. At its core lies the view that inertia is undermining the fundamental objectives of traveler care, data security, employee productivity and savings.


Microsoft says that TMCs must apply machine learning and artificial intelligence techniques to create personas to tailor travel programs to the needs of travelers. By doing so, every business traveler can be translated into his or her collective data and then clustered together with other like-minded travelers.


Unfortunately, it’s not that easy. Corporate travel covers a wide range of personas, from meetings delegate to incentive travel, which all have different options, fare types and rules attached. As human beings, travelers’ motivations will vary. At a higher level, Microsoft’s generic approach flies in the face of the travel industry’s current fascination with personalization.


So where does corporate travel go from here. Here Microsoft’s manifesto has a valid point. There are so many stakeholders who must be convinced of the benefit of adopting new technology adoption for the sake of efficiency.


Ultimately, we believe that travel managers will invest in the technology to predict and resolve travel issues based on traveler behavior. The challenge, and the opportunity is to unlock the power of data and leverage this new currency. How long that takes is a whole different ballgame.

 

Sources:
https://www.pwc.com/gx/en/issues/data-and-analytics/publications/artific...
https://www.pwc.co.uk/economic-services/assets/macroeconomic-impact-of-a...
https://en.wikipedia.org/wiki/Machine_learning
https://www.sabre.com/insights/so-what-exactly-are-virtual-payments-and-...
https://www.eyefortravel.com/sites/default/files/1613_eft_predictive_ana...
https://www.computing.co.uk/ctg/opinion/3021863/how-machine-learning-boo...
http://www.travolution.com/articles/104109/trainline-using-machine-learn...