The Covid-19 pandemic has changed everything we expected for 2020. Globetrender speaks with Jeremy Bowen, CEO of travel industry data and analytics company Cirium, about how it forecasts trends and accurately predicts the future of travel.
How reliable is using historical data to predict the future ?
People often ask me – can big data really predict the future? The truth is, there will always be an element of the unknown – as much as companies might wish we had a crystal ball to see how the travel industry will look in 2030 or 2050, forecasting is never an exact science.
But analysing patterns in historical data enables analysts to detect trends and forecast outcomes that could occur at specific times and within certain parameters.
The reliability of these future predictions depend on the quality of that data and the experts who analyse it. Great data scientists can test various hypotheses using techniques like statistical modelling and machine learning to generate possible future scenarios quite reliably.
Past data helps establish a baseline for potential outcomes in a given scenario and determine how likely each of those outcomes are to occur.
How can businesses make informed decisions about their operations in times of uncertainty?
Cirium supports airline, airports and other travel related companies to make these smart decisions. We are the go-to source for travel data, including airline schedules. We have the most comprehensive dataset of historical schedules in the world, that goes back 16 years, as well as 11 months into the future.
What are ‘predictive analytics’?
Predictive analytics are in high demand from data-driven companies, which take a proactive approach to future planning. Leveraging the power of machine learning, predictive analytics answer the question “What is likely to happen?” and enable businesses to anticipate future trends or changes likely to impact their bottom line.
But, just like the problems businesses are trying to solve, analytics have layers. There are actually four different types of data analytics; descriptive, diagnostic, predictive and prescriptive.
It’s only with a combination of these four layers, all working together, that you can really dig deeper into a company’s performance and better prepare for changes in the market.
The real game-changer for the industry is “prescriptive analytics”, which empowers businesses to not only prepare for the challenges ahead but anticipate problems before they arise. This means they can identify exactly what actions will help them maintain a competitive edge over their rivals.
How can someone ‘control the uncontrollable with data and analytics’, as it states on the Cirium website?
We are experiencing an extremely uncertain and uncontrollable time right now with the coronavirus pandemic. The travel and tourism sector has been heavily affected by government-enforced lockdowns around the world, which have dramatically restricted movement and reduced demand.
One way of managing this kind of disruption is by tapping into the value of big data and analytics – learning from real-time updates as well as from previous data patterns and modelling.
Of course, not everything is completely controllable. But we can gain more control by creating a research-driven response, with flexibility and adaptability built in.
Many companies are using data to deliver real-time alerts and provide flight disruption management, based on their traveller’s needs. Intelligent data capabilities enable them to enhance the traveller experience even when they can’t control all the aspects of travel.
The traveller is really in the driving seat. They expect personalization at every stage of a journey. This puts positive pressure on travel providers to better understand their needs and provide the right service, to the right customer at the right time. Using real-time and dynamic data sources means companies can monitor consumer trends and behaviours, as well as identify potential new markets and revenue streams.
While data is the most critical thing for travel service providers to plan their future operations in the short term, it’s clear that smart analytics will continue to establish itself as the backbone of the modern traveller journey.
What is unique about the coronavirus pandemic in terms of the data associated with it?
We have never seen such an unprecedented turn of events in the travel industry. The impact of the coronavirus pandemic has had a far greater effect than 9/11, the 2008 financial crash or any other seismic global event in recent history.
As the virus has grown exponentially and, in some cases, unpredictably around the world, the data associated with it has changed rapidly from day to day. We’ve had to adapt our operations to respond to the ever-changing landscape. For example, fusing multiple data sets, including flight status and schedules, to gain a complete picture of flight disruption, as airlines began to adjust their schedules on a minute-by-minute basis.
The key value in the data surrounding coronavirus is the insight it provides businesses to plan and prepare for recovery, such as aircraft rotation and how and when to bring aircraft back into service efficiently.
Everyone is currently looking to Asia-Pacific right now as we slowly begin to see “green shoots” of recovery, with a number of domestic flights returning to the skies. We are monitoring this data closely and using it to model the likely recovery path for other global regions.
However, the impact of regional lockdowns has already placed some airlines into crisis mode. Here, we can offer data and analysis around an airline’s operations, including the financial exposure of its fleet, route network and operating airport hubs. This prompts us to help answer questions around government support for our customers.
The second element of recovery that the industry needs to address is how to tackle the inevitable downturn in consumer confidence once lockdown restrictions are lifted globally. This means demand forecasting is likely to become the next challenge for data modelling.
What kind of data are you gathering and analysing in relation to the travel industry? From which sources?
Cirium has an unrivalled data platform – the Cirium Core – which houses data from more than 2,000 sources, from all corners of the aviation and air travel industry.
When air travel isn’t restricted, and is operating like it was earlier this year, we have the capabilities to track up to 120,000 flights in granular detail per day. We track global commercial flight activity in real-time by processing over 100 million flight updates in total per day and maintain an archive of flight activity worldwide.
Drawing from commercial flight schedules, real-time flight activity and the global aircraft fleet, most of our data comes from our trusted industry relationships including with airlines, airports, Global Distribution Service (GDS) providers and international government bodies.
We have direct data feeds of schedules from more than 900 global airlines and work with over 90 percent of the world’s top 50 carriers. Our data helps them to provide information directly to customers, enabling travellers to make more informed choices when on the move.
How do you turn ‘big data’ into ‘smart data’?
Sourcing data is just the beginning of the process. In order to ensure the accuracy, predictability, and fullness of our data, we invest heavily in data cleansing, verification, and scientific extrapolation processes. This is what transforms big data into smart data.
When new data enters our system, we compare overlapping sources to select the most accurate and reliable source. Then, we format data according to globally respected best practices and standards, using techniques such as machine-learning to generate the most complete picture.
If you combine this with the talent at Cirium, you create advanced views of data which become actionable insights for the user. It’s people who make big data truly smart and why we employ some of the best people from all backgrounds, from astronomers to tech giant alumni and airline network planners to aviation finance consultants.
For example, our consultancy team Ascend by Cirium, analyses industry trends and makes predictions for the travel and aviation sector. This helps customers navigate vast amounts of data and focus them on areas of strategic importance to their business.
How do you identify trends?
Sometimes it can be tough to tell the difference between hearsay and hard facts when we talk about travel trends. Staying ahead of the curve to prepare for shifts in the long-term outlook – well before they land – is the goal of predictive and prescriptive analytics.
Trends can also usually have far-reaching effects so it’s important to widen the scope in looking for trend development. For example, identifying and monitoring corporate travel trends could cause changes in the hospitality and MICE [meeting, incentives, conferences and events] sectors.
As trends tend to be multidimensional, a high degree of attention is required, and this needs an expert qualitative view that you can trust.
At Cirium, we put talented people with industry knowledge at the heart of our data and analytics solutions to make sure we capture the most relevant trends, apply the right analysis to project future scenarios and question what data means for the development of business strategies and decision-making.
Our team of more than 500 data scientists, technologists and market experts around the world, work closely to monitor patterns and anomalies in everything from flight schedules to aircraft fleet data – delivering timely insights with a personal touch.
What trends are you seeing for the future of aviation in the age of coronavirus?
The Cirium team of experts have been working with data surrounding the coronavirus pandemic and the industry since January this year, to help shed light on the impact of the travel disruption.
Based on IATA’s prediction of a 48 per cent drop in air traffic this year, we are projecting that we will initially see at least seven years of passenger traffic growth wiped out in 2020.
To return to the demand (traffic) base we enjoyed in 2019 would require 100 per cent growth over the 2020 demand base that the IATA scenario suggests. Achieving this growth could be impacted by many factors, for example travel restrictions outside of airlines’ control, as governments try to slow the pandemic.
It’s also likely we’ll see a drop in demand for leisure and corporate travel, as disposable income falls and former business travellers, or “road warriors”, learn to work from home.
We could start to see a very different airline sector, and by extension air travel market, post Covid-19. Our teams have been tracking the commercial passenger jets in service and in storage in recent months. Their analysis shows that of some 23,000 aircraft in service at the start of 2020, as many as a third could be immediately surplus to requirements based on IATA’s prediction.
Artificial intelligence (AI) will play a more pivotal role than ever, as airlines and the air travel market seek to embrace digital transformation to become more efficient.
We may also see changes in traveller behaviour as social distancing becomes the new normal for the industry. Greater hesitancy to engage with services requiring human contact could see a rise in self-serve tools, robotics and automated processes to avoid unnecessary contact when travelling.
To continue upholding the right targeted experiences, travel companies will be looking to smart analytics and AI. Data will need to flow between partners and traveller touchpoints to make that happen.
Of course, we are yet to start analysing the trends around consumer behavior and at this point it is uncertain how people will feel about travel once restrictions are lifted. What we can do now is work closely with the industry – across the board – to use data and technology to look at ways to manage the impact and develop strategies for recovery.
We are confident that the industry will recover as it has done so before. But some areas will return to health more quickly than others.
Are travel searches a good indicator that someone will actually book?
Under normal conditions, travel searches are a very useful indicator for consumer demand and future bookings. Airlines, hotels and destination marketers closely watch trends in online search to forecast demand and set pricing. Yet, current conditions are anything but normal.
According to Skyscanner, travel searches were up in April, week-over-week, by double digits in all major markets and mostly centered around national holidays.
Skyscanner survey data shows that travellers are increasingly optimistic about the safety of travelling six months from now. Consumers also more optimistic about domestic over international travel.
However, uncertainty about travel safety amid Covid-19 indicates that current searches may be simply aspirational. That is, they may or may not lead to a spike in bookings when lockdown measures are lifted.
There is still a great deal of risk that Covid-19 could rage again due to premature re-openings and lax behavior at travel destinations. A strong second wave of infections is the last thing the travel industry needs right now and could severely impact the long-term recovery of the sector.
Cirium sponsored Globetrender’s report on Travel in the Age of Covid-19, which is available to download HERE