Data is considered by many as the oil of the 21st century
Guest Post: Demystifying data in the travel sector
Daniel Cantorna, Director of Data Innovation, Collinson.
Data is considered by many as the oil of the 21st century. Its value cannot be understated as the insights derived can often make the difference between good and great, success and failure. However, with that said, many travel organisations are still not using data to their advantage. For example, a commissioned study conducted by Forrester Consulting on behalf of Collinson found that only half (50%) of travel organisations worldwide collect a wide range of customer data and augment it with third-party sources. Collinson consumer research also found that 61% of UK consumers say that they ignore the majority of communications from travel providers because they are neither relevant nor personalised.
So what is holding organisations back? In my experience, one of the biggest barriers to ‘data success’ is that both data collection and insights commentary is often overly complex – filled with jargon and/or buzzwords. As a result, it’s become a topic that many people shy away from. The first step is to ensure that employees understand the power of data and how they can make best use of it, by stripping it back to the basics.
Tackling buzzwords head on
As a starter for ten, below are five phrases I believe are often misunderstood or overcomplicated, along with my simplified view on what they really mean, and how and why they may be important to you and the industry moving forward.
First, second and third-party data
Let’s start with a relatively straight-forward one, focused on the origin of data. First-party is the customer data your organisation has collected itself – for example, through your website, sales channels or by asking customers directly. Second-party is someone else’s first-party data, usually acquired through a brand partnership, and third-party is generic data that is usually purchased off-the-shelf from another organisation.
When third-party data is applied with first or second-party data, it helps build a clearer picture of your customers and their preferences, enabling you to offer a more personalised experience throughout the customer journey. As companies seek to cut through the noise of all their competitors and deliver more personalised and relevant experiences, we expect to see more partnerships between travel organisations to not only boost sales, but to get more second-party data.
Extract, Transform and Load (ETL)
This is the process for sharing data between systems and companies. A classic and practical example of this is when a travel brand may look to market to its member base, as it may, for example, want to ETL from both an ecommerce system and an app through a CRM or ESP. It’s akin to copying text from one document to another and having the option to keep source formatting or comply to the formatting of the new document and from a brand perspective, it helps you to build a bigger picture of your customer and member base. In simple terms:
- Extraction is the process of picking the data points you want to incorporate and taking them from one source to the other
- Transforming is the process of re-organising or cleaning the extracted data so that it matches the same format your organisation uses
- Loading is the process of incorporating the data into your systems
Whilst ETL sounds somewhat industrial, it is a relatively simple concept once you unpick it. The challenge lies in the variety of processes and potential uses of that data by different brands. As data becomes more important to companies, data protocols will gradually become standardised in the travel sector to simplify the ETL process. Increased computer processing power will also help streamline the ETL process.
Shorthand for ‘origin and destination’, this data shows where a traveller has departed from and where they are travelling to. Sounds simple doesn’t it? That’s because it is.
Organisations in the travel sector and beyond want to access this data to build a greater picture of their customers. For example, almost every shop in an airport will ask for this information, not for tax-related discounts, but to gain OND data. We predict that more companies will partner with airlines to acquire OND data as a second-party data point as the desire to better understand consumer behaviours and deliver targeted propositions and dynamic content develops.
Machine learning is a way of using computers to analyse large amounts of data and reveal correlations and relationships in a very short space of time. For example, an airline may want to explore the relationships and trends between passengers that currently fly for both personal and business trips, with a view of better predicting and influencing more ‘bleisure’ travellers in the future. It involves a complex series of rules, statistical models and programmed commands to examine data and identify patterns – the system will then automatically perform new analysis based on what it finds.
What many people don’t know, however, is that machine learning is now becoming much more accessible to organisations of all sizes and budgets as more “off-the-shelf” solutions become available. Requiring minimal specialist knowledge, this is a great step forward in the democratisation of data analytics and will enable greater adoption of machine learning in travel brands and beyond.
Deep Learning or Neural Networks
Often referred to as a form of artificial intelligence, this is another computer-driven data analysis process that can allow travel brands to build better predictive models, however, the programmed rules are even more sophisticated and organised in a way that is similar to how humans process information. This means the computer is able to draw conclusions, adapt and make decisions on how best to analyse the dataset in a non-linear way, much like human intelligence.
A practical example and deployment in the travel space is airports, with the development of facial recognition scanners – a proactive move by the industry to make the airport experience quicker and thus, more enjoyable for consumers. Organisations may be tempted to leverage this form of data analysis, however it requires a significant investment and in-house technical expertise. Travel companies should consider exactly what they are trying to achieve with data before they proceed with deep learning or neural networks, as often machine learning is a much more viable option.
Conclusion – knowledge is power
By demystifying data and encouraging all employees to understand the basics of collection and the power of insights and analytics, companies can decentralise expertise and create an army of data advocates. By ensuring a top-down and company-wide commitment to data, travel organisations are only then in a better place to improve processes and utilise data effectively to build a robust view of customers and provide more highly personalised experiences – from the products or services suggested, to the offers and rewards given. In an era where experience is king, data is undoubtedly the smart route to the throne.