Chris Hartmann Multimedia

Today's environment of phenomenal growth in both computing power and communications yields an amount of information that is staggering and often beyond human ability to deal with it. Leaving aside the Internet's infinite information space, even a typical hotel has far more internal data than it can effectively use. Hotel information shares the same four needs as all business data: collection/storage, retrieval/selection, analysis, and communication.

Collection/ Storage

Data comes in two broad types, that which you collect to run your hotel, mostly transactions, and that which you choose to collect to make the property run more smoothly and/or profitably. Collecting the required data is fairly straightforward, but there are several critical aspects:

First, the data must be as "clean" as possible. This includes making sure that when Jack Smith makes a reservation online using his credit card, on which his name appears as "John Smith" and returns a month later based on a phone reservation as "Jack Smith", there is a way to link that information. In addition, if the phone agent thinks Smith is spelled as Smyth, there should be a process to correct that before the information is locked in stone.

Second, the data must be stored in such a way that it's easy to access in the future. Of course the PMS or other system which uses the data as part of it's transaction processing will ensure that it can deal with the data appropriately. For other purposes, or for future use, it should be copied to another database and indexed in a way that makes it useful for it's non-transaction purposes. While this is especially true for transaction data, it may even be true for "voluntary" data, such as guest satisfaction, which may be needed one way (by guest) for follow-up and another way (by event type) for trend analysis. Retention period should be much longer for analysis purposes than may be necessary for transaction purposes.

Based on concerns from the old days when data storage was very expensive, there is often a fear of duplicating data. With vast amounts of inexpensive disk space now available, this is no longer a problem. What is a problem is copied data that is not an exact match. This can happen if data is cleaned or modified in one system and not in another. Ideally, data should leave the originating system completely cleaned and then be modified only by the addition of new fields or new access methods in systems further down the line.

Collecting data that is not needed to run the hotel but may be valuable for sales, marketing and operational analysis is both a challenge and an opportunity. Because its not defined by an existing system, you must identify it on your own, however you have the chance to collect data that is particularly valuable to your property and for your planned usage.

There are myriad options for storing data, however for any but the most basic needs, that storage should be in a database and some thought must be given to how the data will be used. Although there are exceptions, it is also best to store data at it's lowest level, hence not summarized. As with any computer system, data should be backed up as frequently as possible, with at least two generations stored off property.

Retrieval/Selection

Once data is in a database, tools are needed to access it in ways that are meaningful to the use of that information. Reservations may need to be quickly available by guest-identity, rate plan, day of week, specific date or group affiliation. Most important is to determine fields within data which are likely to be used as selection criteria and fields which will likely be sorted or subtotaled. The major difference is that selection fields may well be "keys", meaning one value can be immediately selected, such as guest name. Sort and total fields do not need to be keys since you are unlikely to say "let me see all reservations for Tuesdays", but may say "within this period, show me reservations sorted by day of week". In addition, "key" fields are needed to link two pieces of data together. For example, if you have a data record with each reservation and you want to tie that reservation data to the guest folio on checkout, there needs to be a "key" which ties that reservation to the folio. These are all decisions that will be made with the database developer or programmer but it is important to have a basic idea of how data will be retrieved and used when designing an information store. Another factor in data retrieval is whether to use standard codes and descriptions or free form text. Where possible, standard coding makes retrieval much simpler and is recommended. Such items as comments can be free form, however any free form data would be much more valuable with another field that indicates the nature (good/'bad, high/low, etc.) of the free form field.

Analysis

Analysis is where the fun starts. Once you have your data, some ways of retrieving it, aggregating it and reporting on it, you need to determine what to do with it. As you can imagine, looking at 5,000 reservation details on a 100 page report is not likely to provide much insight or actionable information. But how do you make sense of large amounts of data? There are several approaches.

The simplest approach is to use your knowledge of the data and the property to make some assumptions about how to best analyze it. If you know that filling the hotel on Sunday night is a problem, you may want to select only guests who stayed on a Sunday night and then see how they spread by arrival day of week. If you find that analysis provides little valuable insight, you may move to a seasonal breakdown, or one by market segment, etc. The key to this approach is having a flexible system for retrieving and reporting on data that allows multiple search criteria, sorting and totaling options.

At the opposite end of the spectrum are data cubes. Data cubes are sophisticated (read: more expensive and challenging to implement) tools that allow information to be analyzed along almost any line, summarized, exploded, and generally manipulated to the whim of each individual. Though there is a significant upfront investment of both time and money, data cubes provide the ultimate in data analysis flexibility.

Another powerful analytical capability is "drill down". Drill down (which makes sense only in the context of an online or screen analysis) allows you to select a single line of (summarized) data and see what makes up that line of data. A strong drill down capability allows multiple levels, where you can continuously drill deeper until you are at the transaction or lowest level data. Drill down is a standard capability of data cubes.

Automating the data analysis can sometimes help. In the simplest form, this means setting up rules, such as "When occupancy reaches 75% more than 60 days in the future, alert me". For those with a larger budget, there are tools from companies like Cognos, Business Objects and MicroStrategy that require less understanding of the data and provide even more automated analysis.

Communication

Finally, there is the need to get the data analysis, or less often, the data itself, to the people who can act upon it and translate problems and opportunities brought to light into improvement steps and processes. The older name for these capabilities is "EIS" for Executive Information Systems. These systems range from a simple set of automated reports that are generated and routed to the appropriate managers to a more sophisticated system where reports and again even data, are stored in locations and formats which can be accessed by appropriate managers as they see fit. More recently, the concept of a "digital dashboard" has been developed. Much like its namesake the auto dashboard, a digital dashboard provides simple data on key indicators which can be customized to individual need. While the DOS may want to see ADR and occupancy trends, the GM may care more about today's occupancy, guest complaints and daily profitability. Digital dashboards are meant to be simple, real-time and highly summarized.

Today, one of the most important business intelligence tools are called intranets. They are the modern synthesis of EIS and digital dashboards, and then some - actually, then a lot! An intranet is a browser-based internal web site which stores data, provides tools and archives reports, providing access to individual users based on their needs. A well designed intranet can provide critical information organization-wide instantly, incorporates both "push" (data sent automatically) and "pull" (data requested), with security built in. Since intranets are based on a standard Internet platform, with additional security, they can be made accessible from any Internet-connected location.

Pulling it all Together

Perhaps the best part of today's data storage is the low cost and great flexibility in "having it all". When I began my career, I was amazed that we had a 120Mb disk drive that was only the size of a small refrigerator. Now of course I have four times that storage or more in a piece of hardware that's the size of a stick of gum. Tempting as it is to fear "there's too much data", modern tools from data warehousing, to business intelligence to browser-based technologies provide almost unlimited flexibility to store and use data in meaningful ways that were unimaginable as recent as five years ago. The key remains in investing the upfront time to collect and organize it, and the vigilances to keep asking "What's important now for our organization to know?"

HVS Technology Strategies is a division of HVS International, the world's largest hospitality specific consulting firm. The division was formed in mid-2000, following two years of hospitality technology market research. Our findings revealed a growing demand for unbiased, technology-focused consulting throughout the hospitality industry. HVS Technology Strategies is comprised of consultants with hospitality operations experience. You will not have to spend valuable time educating our consultants on the principals of hospitality business practices. These individuals are constantly researching and keeping apprised of new technology products and services, as well as the practical applications of these products and services in the lodging environment. For more information visit www.hvsit.com.

HVS International is a hospitality services firm providing industry skill and knowledge worldwide. The organization and its specialists possess a wide range of expertise and offer market feasibility studies, valuations, strategic analyses, development planning, and litigation support. Additionally, HVS International supplies unique knowledge in the areas of executive search, investment banking, environmental sustainability, timeshare consulting, food and beverage operations, interior design, gaming, technology strategies, organizational assessments, operational management, strategy development, convention facilities consulting, marketing communications, property tax appeals and investment consulting. Since 1980, HVS International has provided hospitality services to more than 10,000 hotels throughout the world. Principals and associates of the firm have authored textbooks and thousands of articles regarding all aspects of the hospitality industry. Click here for more...

Chris Hartmann is Chief Technology Strategist for the HVS Technology Strategies division of HVS International. Chris has a computer science degree from Harvard as well as over 18 years experience as Chief Technology Officer at a large advertising agency in New York.

Stephen Rushmore
President
516-248-8828, ext. 278
HVS