Building a Site Selection Process that Results in Successful Locations

Smart retailers know that site selection done right is a comprehensive process. We’ve written before on this blog about the importance of site selection that involves both site screening and site forecasting. This methodical approach helps reduce the risks of:

• Choosing poorly screened locations that are unlike stores/locations a retailer would normally consider
• Making site decisions based on inaccurate forecasts

The components of a site selection process

SiteSeer advises retailers and other multi-site chain businesses to embrace the two-step site screening and site forecasting approach, but what exactly should be involved in that process? Here are the essential components:

Data collection

Site selection begins with gathering quality, up-to-date data. That includes both internal data that a company collects about its existing locations and customers and third-party data that it purchases (such as demographic, psychographic, traffic count, business location, mobile location/foot traffic, or other data).

Location analysis

Companies must get an understanding of how their best locations differ from their worst locations. Those key performance indicators might include a minimum median household income, households with people of a certain demographic (i.e., children under the age of 18), specific proximity to competitors, or other factors. It’s critical to understand what traits make a chain’s locations "winners" in order to develop a method for screening sites that blends the art and the science of site selection.

Modeling

A predictive model helps estimate a site’s future sales performance and helps a retailer or chain rule out sites during the evaluation process and give them confidence in their location decisions. Forecasting models require following a strict and rigid set of rules and are statistically and mathematically sound.

Testing and learning

Building a site selection process isn't linear and shouldn't be a one-and-done endeavor. The best processes are rigorously tested by modelers and users/decision makers. A good site selection process combines data-driven tools with experience and judgment. That process should evolve and improve over time.

☑️Steps for effective site selection

Once a company builds its site selection process, key stakeholders involved in identifying prospective locations must be trained on how to put that process into practice. Here are SiteSeer’s recommended steps for good site selection (including the SiteSeer tools one can use):

☑️ Find sites.

New sites that are proposed by internal analysts or other staff, brokers or franchisees need to be screened at a high level. In SiteSeer, users can use Model Builder (and real-time Prospect Model option) to create Scorecards using a company’s key performance indicators/location success criteria. If a company doesn’t have very many locations, is just starting out, or is rebranding and wants to identify sites that fit the new business concept better, they can choose one of SiteSeer’s Location Profiles to give them a starting point. Next, they’ll apply a Scorecard’s rules to hundreds of thousands of locations (in whatever markets the user selects) and receive a list of locations with grades A-F, depending on how well each location fits the company’s success criteria. To visualize those locations on a map, the Hot Spot theme color-codes the map so that viable locations stand out. Other themes can also provide value when finding sites, such as a color-coded map of the best customer segments, important demographics, or competitors.

☑️ Screen sites.

A Hot Spot gives locations a score, but that’s not enough—companies need to screen them further. A site might score a B- on a Scorecard, but a deeper dive into the individual metrics might show that the site is risky. For example, maybe incomes in the trade area are much lower than the typical trade area income of the company’s other best-performing locations. During this step, retailers/chains need to address any concerning metrics. Additional research might be needed to understand whether a concept would be viable in an area.

☑️ Forecast sites.

Once a site is screened and any concerning metrics addressed, it’s time to move on to the forecasting stage. SiteSeer’s professional services team builds predictive models for retailers and chain businesses striving to make the best possible site and market decisions. We choose the right type of model for each business, whether that’s a machine learning model that “learns” why certain stores in a network perform well and others perform poorly and applies that knowledge to new site forecasts or an analog model that evaluates similarities between existing locations and the site being forecast.

☑️ Revisit forecasts.

At SiteSeer, we always recommend that companies revisit their forecasts after locations have opened and matured. Did reality match the forecast? If not, what can the company learn to make the model better? Successful companies understand that models are tools that need to be revisited and improved regularly.

Reduce the uncertainty of site selection

Like any forward-looking science, retail site selection has its challenges. Companies are betting millions of dollars on their next location, and no one can see the future. The good news is that implementing a systematic process for site selection can reduce the uncertainty and allow retail and other chain businesses to make more informed, profitable location decisions.  

Learn how SiteSeer can help your business implement a world-class site selection and forecasting process. Schedule a demo today!