From the very beginning, computers and their systems have required human input to function. The degree of human intervention required is constantly decreasing and changing, but in some form, it is always present – and RMS (Revenue Management System) is no exception.
RMS is tasked with analyzing huge amounts of data and making forecasts based on that data to help revenue managers in their work. Currently, even the most advanced artificial intelligence-based systems can’t think on their own – and that’s okay! – but they can be a huge help to professionals on a day-to-day basis. Let's see how.
What is an RMS?
An RMS (Revenue Management System) is a tool that helps revenue managers to work more efficiently. It has a number of useful features, such as 24-hour price monitoring, speeding up data collection, analyzing data in some cases, making recommendations, and even taking action.
A properly configured system can support the revenue manager’s work in many ways. Its main advantage is that it works 24 hours a day without getting distracted, which can be especially useful for hotel professionals who manage multiple properties. However, the system cannot function effectively without human intervention. Just as an artificial intelligence-based search engine is developed by software engineers, an RMS becomes functional and efficient by processing and understanding the data entered by a revenue manager.
Types of RMS
In general, there are two basic types of revenue management systems: rule-based systems and artificial intelligence-based systems. For the time being, the former is still more common, but the number of hotels using AI-controlled systems is constantly increasing.
- Rule-based systems
A rule-based system is significantly simpler than an AI-based one. Its operation is typically based on a basic algorithm using just a few variables. These variables may include occupancy, competitive prices, the level of market demand, or other basic factors which are usually quantitative in nature.
They usually work on an “if X happens then do Y” basis. For example, if occupancy increases by X percent, the price should increase by Y euros or Z percent. They can be used by almost anyone and require little or no experience in revenue management. Such systems are easy to use, but that simplicity also comes with limited results.
- AI-based RMSs
With a well-configured artificial intelligence-based RMS, revenue managers have a much more efficient tool to work with. This combined system takes into account all the variables that I mentioned in my article on dynamic pricing – and more. Artificial intelligence responds to change through machine learning. Of course, any machine-based system must be taught and constantly calibrated to assist with and partially take over the work of a well-trained revenue manager.
The need for human intervention
Once the revenue manager has set up an AI-based RMS, it can work quite effectively by following observed market trends and past patterns. Many may think that after a certain amount of time, human inputs become unnecessary. This isn’t true, however.
The dramatic events that have shaken the world and thus the hotel industry over the past few years, notably COVID-19 and then the war in Ukraine, took many unexpected turns and quickly revealed the limits of computer systems as well.
In the hotel where I worked, we had to completely shut down our artificial intelligence-based solution shortly after the COVID-19 outbreak, as we were receiving completely irrelevant recommendations from a system programmed based on past experience. It had no solution to these unexpected circumstances. Of course, not only our property but many other hotels that had used AI successfully in the past found themselves in this situation. Professionals everywhere had to intervene to find solutions to generate new demand, explore new markets, and develop new strategies.
Whether positive or negative, unpredictable events are an inescapable part of life. Because even the best RMS systems base their recommendations on past experience, any unexpected future event will confuse them, forcing revenue managers to shut down the system and take control. Software alone cannot provide the sophisticated predictions and creative ideas that a professional needs, at least for the foreseeable future.
How do RMSs handle subjective factors such as customer preferences and brand positioning in their decision-making process?
Revenue Management Systems (RMSs) primarily rely on quantitative data and algorithms to make decisions. However, they can incorporate subjective factors like customer preferences and brand positioning indirectly. For example, RMSs can analyze historical booking patterns and customer segmentation data to identify trends and preferences associated with different customer segments. Additionally, revenue managers can configure the RMS to prioritize certain customer segments or align pricing strategies with the brand's positioning. While RMSs may not directly interpret subjective factors, revenue managers play a crucial role in interpreting the system's recommendations in light of these subjective considerations.
Revenue managers can leverage RMSs in various ways to adapt quickly to unforeseen events beyond simply shutting down the system. Firstly, they can adjust pricing and inventory strategies dynamically in response to changing market conditions, using the RMS's forecasting capabilities to anticipate demand shifts. Additionally, RMSs can be used to analyze real-time data from alternative revenue streams or emerging markets, allowing revenue managers to diversify revenue sources and mitigate losses during crises. Furthermore, RMSs can facilitate scenario planning and sensitivity analysis, enabling revenue managers to simulate the impact of different crisis scenarios and develop contingency plans accordingly. Ultimately, RMSs serve as valuable decision support tools for revenue managers navigating unpredictable events.
Final words
There are those who say that RMS will completely take over hotel revenue management and there will no longer be any need for professionals. However, they often confuse revenue management with pricing, two concepts which are certainly not synonymous.
Although pricing can be fully automated in most cases, it is only one part of revenue management. Pricing is the end result of a complex decision-making process – and even if pricing is automated, there is always the possibility of another unexpected event that requires humans to take back control. Therefore, I am convinced that although RMS systems represent the future, they cannot function effectively without skilled revenue managers.