In recent years, the tourism industry has been disrupted by a strong push for innovation, thanks to the introduction of new technologies and greater managerial awareness of their necessities. As is often the case, new technology evolves very quickly in times of lack to solve/address the issues that the industry is facing at the time, so it’s important for us to recap some of the most important developments, to better understand the parts of technology that are necessary for hoteliers and which ones can be put aside for another day.
A Brief History
Big Data is a relatively new concept in the tourism industry and, as such, there is a still a great deal of confusion as to what Big Data is and how it helps hoteliers get more bookings and revenue; to make it easy, I’ll give you a simpler definition and explain why it’s key for the future success of revenue management.
Big Data is, quite simply, a huge amount of data – including email data, data from images, text, search data, etc. – that can be either structured (i.e., presented using tools like Excel) or unstructured.
The vast use of cloud services combined with the volume of data has resulted in the term Big Data – simply because the scope and complexity of the data available are beyond the human mind to compute in a time-effective manner. The interconnection of systems, combined with the infinite capacity to store data at low prices, has opened the door to the democratization of Big Data, which was previously the exclusive domain of only a few.
Until recently, Big Data was the key element missing, but highly necessary, to complete and perfect the modern revenue management processes, revenue management systems and strategies – but thankfully, today, we have the technology to support revenue management completely.
What Exactly is Big Data in Revenue Management?
The application of Big Data in the travel industry can be very extensive and varied, so let’s focus on its impact on the field of revenue management, where it is ideal for the analysis of demand, as a basis for pricing decisions.
The classic elements and key actions involved with effective revenue management are:
● Selling your room at the right price…
● Selling your room at the right time…
● Selling your room to the right customer.
The evolution of revenue management analysis techniques has always revolved around these concepts.
We often hear about the importance of elasticity of demand, as a way to understand how much a customer is willing to pay for a hotel room, but this data has always been difficult to obtain. The simplest attempt has been to analyse the hotel's history (via PMS data) to study the purchasing patterns of customers in the past, and assuming they will remain consistent, using them to make future decisions. In the past, that might have provided successful at times, but there are many case studies based on this model of analysis, which have glaring weaknesses that we can explore simply by re-evaluating the strategy, based on our earlier definition of revenue management:
1. Selling at the right time
Q: When is the right time to raise or lower a room rate?
A: Without Big Data, the answer is that we don't know for sure.
According to the traditional approach, we analyze the history, define a starting rate and based on the past year, we adjust prices by taking into account the evolution of occupancy, along with other more or less important variables (which also change according to how the revenue manager applies his/her decisions). This method of analysis is correct but not very precise.
Without real-time demand data, we don't know any precise data about potential guests’ interest in traveling to your country, destination and/or property – or how many people will be coming to the destination in the coming days/weeks/months/year – because we only have a projection from the past.
We don't know when to raise the price, or whether to wait, because we don’t have any idea of what the future will bring, in terms of demand.
As such, many revenue managers would choose to raise their room rates at the wrong time or choose to not raise them when there are customers who are willing to buy at higher prices. Alternatively, other revenue managers blindly try to match or beat competitors' prices, a practice that often leads to a downward spiral, depressing the entire market.
Q: What is the right price for guests to want to book a room at your property, instead of the competition?
A: Without demand data, it is very difficult – if not, impossible - to answer that question.
One method is to, again, analyze price patterns over history but here’s an important flaw in that strategy: can we ever be completely sure that the correct price was applied in the past?
Unfortunately, we can’t. Like in the previous example, this method makes it very likely that you will inherit some errors in the present, due to past pricing errors.
This strategy isn’t just a problem for revenue managers; it also provides highly inefficient when using technology that uses this type of algorithm to calculate room rates.
Let’s look at an example: in machine learning, models of classification, such as automatic image recognition, the algorithm is 'trained' to recognise an image because it is given the correct image beforehand; as such, it can identify an apple, as an apple, when it sees one later, because it was initially shown an optimal image of an apple (as a basis for comparison).
In the case of hotels, revenue management algorithms that base their pricing on this strategy, will encounter the same problem; by using potentially flawed (or less than optimal) historical data to predict the future, you might get suboptimal pricing suggestions, which don’t factor in current market conditions.
There is no dichotomy between Small Data (PMS data) and Big Data (demand data) because, the two combined, provide the complete picture that is needed to develop data-based revenue management strategies and establish accurate rates, in real-time:
● In revenue management, Big Data offers the opportunity to analyze, in real-time, how many customers are searching for your destination, from which country of origin and with what spending power. This is crucial information to know in order to set the right price, to attract the right customer; however, without an RMS with large infrastructure and advanced analysis models to accurately process the data, it’s impossible to capture and analyze this data manually.
● Small Data makes it possible to observe, in detail, the main data elements of the hotel, including the evolution of key metrics, which are essential for defining sales strategies – of course, that is assuming that the data that is being analyzed was entered correctly into the PMS.