“I need to get more scientific about site selection.  What software should I buy?”

Given the massive closings of stores since 2008, many chain store executives are asking this question today in order to avoid a repeat performance as they start growing again.

The problem is, it’s the wrong question!

About 20 years ago it dawned on me that most real estate investment decisions were made using gut feel fueled by the desire to close deals, often in the face of contrary evidence.  For those old enough to remember, this was in the wake of the late 1980’s Savings and Loan meltdown, which seemed like a pretty serious crash until the Subprime Mortgage debacle of 2008 (estimated cost $7.7 trillion versus $1.5 trillion for the S&L Crisis).

I set out to create software that would help decision-makers gain fast access to market knowledge and insights that would increase the financial performance of their investments (this seemed easier than making software that would reduce greed on Wall Street or incompetence in government).

It took me another 10 years to realize that I had underestimated the importance of the decision-making process itself!  The best tool is of no value in the wrong hands.  The selection of research and predictive analytics tools is actually the LAST step in the quest for better real estate decisions, which should follow these steps:

1. Design the objectives and processes for decision-making
2. Find the right people for the team
3. Acquire the research and predictive analytics tools to support  the people and processes

Each of these steps is disruptive to an organization.  However, buying software is probably the least disruptive, so that’s where many companies choose to start! Gary Cokins, a veteran in the area of analytics and organizational change, says this:

“Organizations seem hesitant to adopt analytics. Is this due to evaluation paralysis or brain freeze? Most organizations make the mistake of believing that applying analytics is 90 percent math and 10 percent organizational change management with employee behavior alteration. In reality it is the other way around; it is more likely 5 percent math and 95 percent about people.” (source:  http://www.informs.org/ORMS-Today/Public-Articles/February-Volume-39-Number-1/Obstacle-course-for-analytics)

This explains why it is difficult to sell analytics, and it’s interesting to see how consultants and software vendors have adapted.  Here are some examples:

1. If a company is not willing to significantly change its process, they will not get much value from predictive analytics.  Therefore, they won’t want to pay much for it because the ROI will not be sufficient.Hence, some vendors optimize their product offering to be inexpensive, which usually means reducing the quality of the modeling services and the software deliverable that contains the model.Unfortunately, there is no such thing as a cheap model that is also reliable.
2. Some companies do not have the expertise in-house (at any level) to understand predictive analytics, which includes data quality, modeling methods, and the interpretation of results.In order to “meet the customers where they are,” some vendors create “black box” models with stunning graphics and user interfaces that insulate the customer from the messy, elusive realities of the complex retail marketplace.

   Although video games can be incredibly realistic, they are still not reality.

Everyone wants a quick, easy solution.  If you try to lose weight with diet pills, you’re not really dieting.  If you want to be more scientific about site selection, take the time to look at your decision-making processes, staffing, and finally, your predictive analytics tools.  Depending on the number of stores you are opening and the size of the investment, you might choose one of three approaches to making better real estate investment decisions:

1. Hire a consulting firm to build a model and provide an overall market strategy and an analysis of each site that you are seriously considering.  If your level of activity is not high, this might be more cost-effective than creating an in-house staff and equipping them with the tools.
2. Develop an in-house research team that can support the decision-making process, and provide them with models and tools built by outside firms who are experts in this highly specialized discipline.
3. Start with outsourcing and gradually build an in-house capability over time as the economies of scale justify the investment in the fixed overhead of people and tools.

Mapping and reporting tools are very good and affordable today, and every company should have the ability to analyze markets and sites with these visualization tools and generate descriptive reports.

However, predictive analytics require much more expertise to build and use.  If you aren’t ready to change the way you work, and invest in proven modeling methods, save your money for when the time is right!