Is Modern Data Warehouse/Data Lake with Machine Learning the way to go?


Hotel technology has been around for decades, specifically the PMS.  I was fortunate to work for one of the pioneers of this technology in a now defunct company, but through the years I always struggled to understand why the technologies became so segmented. At one time a PMS was your folio management, guest profile management, back office, housekeeping and inventory system.  With the advent of the “above property” systems technologies became more segmented, fragmented and ultimately disjointed.  Add on to that with the specialization of data Channel Management systems, revenue management systems, ORM, Guest Messaging, etc. I could go on.

So, we went from a centralized system, though limited, to an amalgamation of information, data sets and chaotic reports.  How exactly is anyone supposed to keep track of all these systems, reports and yes, monthly bills?  At what point does it become too much information to be actionable by the on-site staff?  We love to think that it is all under control but with all these specialized systems what important data sets fall through the cracks.  Even the most experienced hotelier has this challenge with the new information being created by ingenious programmers dreaming up the next big idea.

Why not Modern Data Warehouse/Data Lake with AI/ML?  Think of it along the lines of the old school IBM Mainframe where all of the information is kept in a centralized location and can be accessed at any time.   Proper interfacing to the essential software’s has always been important but why not interface different system to pull both structured & unstructured data, and draw all the information into one point that organizes the data specifically to the property’s needs? It may seem like a daunting task, but properly organized actionable data could be at the fingertips of on-property operators to make informed decisions based on historical data leading to, at worst a mitigation of liability and at best an increase in revenue. The key is to leverage some of the in-memory technologies along with Machine Learning to consolidate the data after cleansing and running on cloud or hybrid mode.

Now let’s look at how this information can be used on the guest services side.  Imagine a guest walking into the property and you already know what their preferences are based on past purchases. Being able to marry PoS data from the outlets, their past survey responses, social data and any other relevant data from past stays would be helpful in servicing these guests, not to mention what the marketing department could do with this data to entice them for a return visit.  While geo-fencing for larger properties can utilize guest flow on the operations side and proximity marketing on the marketing side all being tracked into a unified guest profile to give you a complete picture of the guest journey.

We could take this a step further and merge the data from the revenue management system, all of the comp set data, online reputation management creating a single picture without entering into three different systems.  One report that is fed by three systems with one click.  The increased value of this is these reports can be set by property based on the needs and value placed on different data sets.

Information flow is vital to any organization but in the unique world of the hotel environment it could be a game changer in the delivery of guest service and operational efficiency.  Different systems seamlessly passing information in real-time is an achievable goal.  VenueLytics is a world-class AI/Data Analytics platform that will transform the way you view hotel data.

If you would like to learn more about VenueLytics and our customer experience platform, powered by AI and analytics, please visit

Gino Ferrario

Senior Director of Sales