Software that performs sophisticated analysis of enterprise data to uncover patterns has enabled innovative organizations to more accurately predict their business outcomes, improve business performance and increase profitability. Companies wanting to further extend the knowledge they glean from complex predictive analytics algorithms are integrating location into their business intelligence (BI) solutions to provide business users throughout the organization with the ability to leverage the geographic component of their data to gain new insights into their customers, markets, and operations.
Location is valuable to an organization in multiple ways. When geographic visualization and data are combined with predictive analytics tools, companies become aware of the direct impact location relationships have on business operations' future business performance. From analyzing spatial dimensions of transaction data to demographic information, incorporating a location component into predictive analytics processes arms companies with more of the right information to respond to business needs quickly and appropriately.
Take for example, the retail industry. For this multi-billion dollar industry location is everything-from where to put a new store location to targeting key market clusters in marketing initiatives. From providing basic information and visualization capabilities to producing geographically tailored site reports and statistics, predictive analytics tools enhanced with "location intelligence" offer a far more powerful and accurate decision making tool. In the end, companies will improve their ability to determine optimal real estate strategies to maximize ROI, identify and prioritize markets for expansion, quantify and avoid cannibalization, and generate detailed site-specific sales forecasts for operations and strategic market planning.
Bringing BI into the predictive analytics picture
BI platforms are key to ensuring that the analytics are de-centralized and distributed to a large number of users for accurate decision making. By giving multiple users in the organization the ability to access sales and customer data, business intelligence technology shortens the distance between analysis and action.
In marketing for example, this means that regional, store and program managers can all access the data they need quickly-without having to know table structures, SQL queries, etc. This leads to the ability to segment markets and execute campaigns more effectively, resulting in increased ROI. Put simply, using business intelligence for marketing allows managers to fine-tune their message, fine-tune their medium, and fine-tune the target they are addressing. Rather than gut feel, marketers that leverage business intelligence use sound analytics to define a campaign.
To return to the retail example and to the retail real estate world in particular, the confluence of business intelligence software, predictive analytics and location intelligence is evolving the way businesses understand and perform. A company making strategic growth decisions such as selecting a new location for a store can incorporate predictive analytics approaches into daily operations and address critical business needs (i.e. identifying best markets for expansion, generating site-specific sales forecasts, etc.)--ultimately improving overall performance. Using predictive analytics for site selection needs is an excellent example of how companies can integrate data, software, modeling and consulting solutions to gain profound insight into customers and markets. Complex "built-in" predictive analytic algorithms can factor in the most statistically significant decision variables in an automated fashion to provide the user with a greatly narrowed field of choices and specific recommendations at the push of a button.
When BI tools are added to a company's predictive analytics process, it opens new doors of sharing, comparing and analyzing data across departments and the entire organization. For retail technology such as lease management software and site selection models, BI enabled predictive analytics means retail management professionals are no longer forced to erect silos of data, separating departments and decision making. Gone are the days of senior management needing hours to pour over reams of data from disparate departments in an attempt to determine which issues need immediate attention and which can wait. The ability for retailers to define and track interdependent strategies, through integration of BI systems, is resulting in more efficient use of capital and human resources which has a significant influence on bottom line performance. Senior retail managers are enabling departments to work together and evaluate data and information in a cooperative environment that produces decisions and direction that is in the global interest of the company and not just that of the departments that constitute the organization (e.g., operations, merchandising, marketing, real estate, IT, etc.). The most common integration of BI tools used by many companies today comes from firms such as MicroStrategy, Cognos, Business Objects, and Hyperion. Many of these organizations have penetrated major companies at the most senior levels, integrating information systems into daily operations. The use of informational executive dashboards allows senior management to constantly access company data transformed into information in the shape of graphs and charts and view it on a desktop. The information, and more importantly the visualization of critical performance metrics, was just a dream only a few years ago. The time and expense associated with information retrieval from an IT department used to bring executives to their knees, begging for the data keepers to get them information, and then when the requests were granted, the reams of paper that it was delivered in made it almost impossible to decipher.
Today, executives from the CEO to departmental heads can conduct predictive analytics and use location-enabled BI tools to turn data into information that can be used to track performance, shipping and receiving and any other business activity by market or region. These key metrics can be viewed in an instant, allowing the management team to identify potential problem areas, pull in the correct teams to address the issues and then quickly define a solution to address the problem areas. The same systems allow management to track the implementation of the agreed upon solution and see whether the results are positive. Given the pressures from the investment community for quarterly performance, C-level management in an organization must constantly monitor performance and identify potential problems quickly, and it is these BI tools that provide the catalyst for those quick evaluations and decisions.
Predictive analytics in action
To return to the world of retail to show predictive analytics in action, the advent of lease management and contract management software from companies such as Accruent, enable retail organizations to follow real estate transactions, site by site, from beginning to end. These systems can be integrated into the retailers' network environment allowing remote users to access the system for daily or weekly updates. The real estate vice president can evaluate the status of deals from lease negotiation through construction to ensure that the site plans are being submitted on time and that lease contracts are secure, as well as tracking whether there are delays in construction of projects due to architectural changes or delays in shipping of equipment.
The efficiencies introduced to companies that have to manage more than 50 store openings a year and roughly 15 percent of their real estate portfolio coming up for lease expiration in any given year is substantial. These efficiencies can reduce time to grand opening by weeks per store and allow the retailer to be more proactive rather than reactive to decisions wrapped around lease expirations.
In one such case, a prominent restaurant chain reduced its time from site identification to grand opening by 60 days per unit. Its annual growth plan was for 100 new units per year. When opening 100 units a year at a rate of 60 additional operating days per unit due to a quicker to market delivery, the chain was able to add 6,000 operating days per year to the chains revenue stream. These types of results are bound to put a huge smile on the face of any CEO and shareholder and, clearly, more money in their pockets.
The realization of retail leaders that sound real estate strategies are paramount to the long term health of a chain has led to a significant increase in the use of site selection models to assist in defining when and where to open stores. Site selection! applies predictive analytics approaches to uncover the most profitable and sound business decisions.
The critical use of point of sale data or customer transaction information leads to the development of customer profiling that arms the retailer with information to understand and predict their expected contribution from consumers in any market from all demographic and lifestyle segments. Knowing the characteristics of high and low value customers gives the retailer the ammunition by which to evaluate any site in a market and develop a sales forecast that will yield a high degree of confidence. Modeling techniques today run the gambit from regression based, gravity based, neural net based and analog based. The retailer needs to be sure that the source of the modeling technology is a company that is familiar with current modeling technologies and can advise them of the strengths and weaknesses of each methodology and its effectiveness for their business.
The use of site models by companies such as The Home Depot, Brinker International, Kohl's, Burger King, Talbot's, Target, and others has been a catalyst in the development of strategic real estate plans. Every retailer is pressured by investors to ensure that they minimize brick and mortar mistakes that can have a devastating affect on profits.
The development of sophisticated modeling techniques to define market strategies and forecast individual site sales is the best insurance policy against catastrophic development mistakes because decisions are based more on impartial fact (real customer, business and market data) rather than intuition. The lack of quality real estate, the competition for land and the pressure from Wall Street has pushed senior management to find ways to enhance decision making to a level that significantly affects profits.
The technologies cited above afford management the opportunity to improve strategic real estate decision making, increase the pace of store deployment, increase revenue, and create early warning programs that allow them to quickly identify and act to correct negative trends in the business performance.
Although this article has focused on real estate and retail businesses, other industries such as communications, financial, healthcare, insurance, mobile, supermarket, government, hotel and media also benefit from the use of predictive analytics and expand on the capabilities of these processes by adding in location components and BI platforms. While some companies have extensive data from POS and loyalty programs in a central data warehouse, others possess only small amounts of in-house data. With the recent explosion of business data and accelerated market dynamics, more decisions must be made in compressed time frames. More and more businesses are turning to predictive analytics as the means to meet this demand for real-time decision making and to improve corporate performance.
biography
Mark Zygmontowicz is the Managing Director of analytical services segment of MapInfo, a location intelligence company.
