Room Pricing: Using Relevant Data and Machine Learning for Precise Pricing
By Gino Engels Co-Founder, OTA Insight | November 04, 2018
Emerging technology is always a hot topic. It's gone from Big Data to data science, artificial intelligence, machine learning, predictive analytics, and most recently, blockchain. Keeping up with these technologies is never easy, especially for the hotelier focused on running a strong, profitable operation with a memorable guest experience.
As a third-party partner to the industry, we pride ourselves on being shepherds across this emerging technology landscape. There are new channels emerging regularly, such as alternative accommodations and niche booking engines; hoteliers are bombarded with new places to distribute their content.
And with each new channel comes an additional burden: ensuring parity and competitiveness of room pricing across channels.
For us at OTA Insight, we believe that Big Data is an overused-but-descriptive term that has helped us all understand the latent value within large datasets. The evolution of machine learning, which empowers decision-making to happen in real-time without further human input, now means that this value can be rapidly unlocked in real-time.
This evolution to real-time has created a world in which room rates are adjusted in seconds, rather than days. Dynamic pricing allows hotels to change prices rapidly in response to changing market demand.
While this maximizes profitability, it also dramatically increases complexity. It's not unheard of for there to be multiple price changes in a given hour. Consider the variables at play: seasonality, local events and conferences, room types, amenities, variations of packages (such as breakfast included)...the rate permutations across hotels in a given market can easily run into the millions.
So how do revenue managers adapt to this firehouse of data?
It's about having the most relevant dataset-you need to have a handle on how, when, and where competitors (and their algorithms) are changing pricing. And then deploy a combination of efficient machines and intelligent humans to derive precise pricing for a particular property. With that goal in mind, here are the good, better, and best tactics for leveraging data effectively.
Determine Your Realistic Compset
Segmentation is the most manual part of leveraging data science and machine learning. It requires a thorough understanding of a given market and a detailed understanding of your hotel's target demographics.
Compset is essential, as inputs dramatically impact automated technology. The data maxim truly applies: "trash in, trash out." So if you're monitoring the wrong set of pricing buckets, the pricing assumptions will also be wrong. This affects your forecasts, which will be improperly correlated to your property.
Another key point to crafting a valuable Compset: radius is not the only feature of a competitor. Don't hesitate to compare properties that appeal to similar demographics. Each property competes not just on location but on design, amenities, and overall aesthetic.
Good: A basic set of competitor hotels based on a radius around the property. These are the hotels whose rates matter most.
Better: A robust set of competitor hotels based not just on radius but also appeal. This includes hotels that might be on the other side of town but attract the same demographic. This requires thoughtful guest personas at the property level.
Best: All of the above, adjusted quarterly-and including periodic visits and stays at competitors to validate your instinct that properties resonate with similar target demographics.
Monitor and Verify Regularly
Today's technology is sophisticated enough to deliver real-time pricing recommendations. But this doesn't mean that all recommendations are correct for a given property's objectives across channel mix, ADR, occupancy, and other relevant metrics.
When it comes to using data to empower decision-making, particularly on yield and revenue management, it's important to always check the work. The computer doesn't have all the inputs-your gut still matters. For example, if booking windows are shrinking or expanding, you'll know best how to adjust overall pricing to maintain that optimal booking window.
Price parity is also a key part of revenue optimization. AI is the most efficient way to ensure that pricing is consistent across channels. Humans then review differences in price and determine a plan of action.
Good: A weekly cadence of analyzing your pricing relative to competitors. Adjust accordingly when things are out-of-line. Focus on outliers first.
Better: A daily review of pricing, coupled with a cohesive effort to make pricing a part of a team approach to operational excellence. This means tracking how things are going, and where the property is meeting (or falling short) of revenue goals.
Best: Optimize for both price and availability. Push inventory to the highest converting, lowest commission channels. Fix parity violations. Move from the outliers in, so that parity is consistently achieved across all channels.
Take Ownership of Data
Most hotels are already at a disadvantage on the data front, as OTA and metasearch platforms enjoy a massive volume of data to parse. Yet, given the pace of change today, whoever owns the data in this paradigm controls the narrative. While not always possible to own each piece of data that a guest's stay creates, the goal should always be to maintain as much ownership and control as possible.
"Data, I think, is one of the most powerful mechanisms for telling stories. I take a huge pile of data and I try to get it to tell stories." – Steven Levitt
Remember that those who own the data are able to make money beyond just the booking. Guest demographic data alone can fuel new products based around traveler insights. Each data point is a potential piece of a larger story, a story that explains why a consumer makes a decision-and how to get more positive business outcomes by encouraging more similar user behavior.
Data is a store of value, so if you're taken out of this equation, there's value lost. Fight for your data-and then clean it up so it's structured and ready to be leveraged in new ways. When you know your data, you are best poised for success.
Good: Be aware of any data sharing clauses in contracts. Manage expectations. Act on the "actionable" insights that existing data provides.
Better: Pull in data from other sources-such as marketing campaigns, on-property promotions, and email lists-and work to clean them up for future usage. Make your data structured and accessible for use across the operation.
Best: Capture as much data on-property as you can. Incentivize guests. Ensure that your PMS/CRS/CRM are as interoperable as possible. Train staff to keep records updated and detailed. Apply best practices learned from OTA partners as far as how they capture, store and use data.
The question stays the same-it's the answer that changes.
"How do I make the most profit out of each room?" This is the question that drives the hotel business. The answer to that question is different for each property. And it's an answer that evolves with shifting market dynamics, emerging technologies, guest expectations, and global demand trends.
The true skill for today's revenue manager is navigating this dynamic environment through both technology and humanity. The inputs are varied, the pace is relentless, and the competition intense. That's why it's more important than ever to rely on relevant data; those who have the cleanest, most accurate data for a particular query are always best positioned to make the best business decisions at the room level.
To be successful with data, seek out the partners you trust in terms of asking for advice, in terms of bringing the newest technology, and in terms of the data that's going to make your business profitable and ready for the next step in the digital revolution.
And never underestimate the power of your own gut instincts. As futurist Alvin Toffler reminds us, intuition matters more than ever: "You can use all the quantitative data you get, but you still have to distrust it and use your own intelligence and judgment."
HotelExecutive.com retains the copyright to the articles published in the Hotel Business Review. Articles cannot be republished without prior written consent by HotelExecutive.com.