If you are a hotelier, you should worry about Big Data about as much as you worry about missile defense. That’s because the chance of you working on either is slim to none. Sure, you should be informed about the subject, and be able to speak intelligently about how it will change the traveler experience, but always be conscious of the fact that property level hotel data is not Big Data. Understanding what Big Data is and what it is not will prevent you from making the mistake of being mesmerized by the vendors who are just now gearing their sales pitch to convince you that you need to invest in yet another “solution” to take advantage of Big Data opportunities. Remember these three words when they come knocking at your door – source, size, and systems.
Source
Over the next decade most Big Data projects will involve searching weblogs for patterns in user behavior. Every website has its own weblog where it stores detailed information about each user session. Each click, each data entry, each unique user ID is tracked. Google Analytics, a free solution, is the overwhelming leader in summarizing this data into charts and graphs – so they have already taken care of that problem. The only part of the user experience that Google Analytics does not track is the user activity on the third party booking engines that most independents use (i.e. Travelclick, Synxis). That data is kept on the weblog of the booking engine, where Google Analytics cannot search and therefore you do not have access to it. Sure, these booking engine vendors provide some analytics, but not the kind that is needed by RM and Marketing to be able to make strategic or tactical decisions.
For now and the foreseeable future, your data sources for analysis are the PMS, POS, labor systems, Inventory tracking systems, and any other service or transaction system at your property. At the property level, none of the databases in any of these systems are big enough to be labeled as Big Data. Why?
Size
Look up any legitimate source of information on the subject and you will learn that Big Data is now counted in petabytes. This is 1 million gigabytes of data. In terms of rows of information, according to Google, Big Data starts at about 500 million rows of data with multiple columns of variables. To see how property data measures up, let’s do the math on reservation information for a 400 room property.
400 rooms x 365 nights x 5 years = 730,000 rows of data.
Therefore, even if you were to create a data file to analyze 5 years worth of reservations for this property you could easily fit it in an Excel spreadsheet. Now, if we tracked 100 variables for each reservation(most hotels look at less than a dozen), and each variable was 50 characters long(most names are less than 20 characters), we would have a data file that is 1.5 megabytes. Definitely not Big Data potential. Even if you were to take every line of every receipt from every outlet for a few years at one hotel property, the file size would not even come remotely close to a petabyte of information.
Systems
With such “Small” data, a spreadsheet or even a free statistical package like R, is sufficient for any property to run most summary analysis. Throw in some SQL Grouping into your queries and you could easily handle analysis for multiple properties on one Excel sheet. Granted, you may have to upgrade the processing power of your desktop to cut down some of the calculation time. You may even want to invest in a small data warehouse, but none of these investments will break the bank. Those people working on Big data problems, mostly academics for now, do need massive parallel processing hardware and software, but again, property data will never be that “Big”.
Finding insights into guest behavior across many properties, social networks and weblogs is a legitimate Big Data challenge for very large hotel chains. This is the exception, however, and I can’t think of any other analytics project that would involve Big Data in the hotel industry. Even in this case, large hotel chains will inevitably find that looking at massively commingled property information is less valuable than analyzing single property data sets. Therefore, to all hoteliers I say, use your unique data and analytics in order to deliver memorable guest experiences, but don’t worry about Big Data.
Follow @owlanalytics Join the Revenue Manager Development LinkedIn group
[jetpack_subscription_form]
Robert Hernandez, Hotel Profit Engineer
Author Info: Robert Hernandez is an expert in the field of mathematical Hotel Optimization and Analytics. He has spent the last 17 years building data-driven forecasting and optimization models for companies in over 20 different industries, from tech to tourism. Robert possesses a very unique skill set including cross-disciplinary experience, advanced mathematical and analytics skills, data transformation, industry-specific knowledge and business-process improvement expertise. Robert began his career at the Walt Disney Company in Revenue Planning. Read More+
