Contact Us!

* indicates required

By submitting this form, you agree to receive promotional messages from PiinPoint about our products and services. You can unsubscribe at any time by clicking on the link at the bottom of our emails.

Contact Us!

By submitting this form, you agree to receive promotional messages from PiinPoint about our products and services. You can unsubscribe at any time by clicking on the link at the bottom of our emails.

Thank you

We will be in touch soon!
Oops! Something went wrong with our form.
Please send us an email at Info@PiinPoint.com and we will help you out.

"PiinPoint has become an integral part of my role as Retail Analyst at Cushman & Wakefield Waterloo Region. The platform allows me to put together professional looking reports and provide clients with the insights they need to make real estate decisions.

I honestly don’t know how I would do my job effectively without PiinPoint."

Jessica McCabe, M.Ed.
Retail Analyst

PiinPoint
PiinPoint Logo
Login

Predicting Success in New Markets

PiinPoint

-

January 19, 2022

Choosing the right market and the right real estate for Retail has certainly seen its fair share of change over time. From the early tactics of spotting steeples from a helicopter in the mid 20th century, to the advanced strategies employed by retailers today using GPS and AI technologies, there has been a constant pace of evolution. Now more than ever, site selection has to be driven by data.

For real estate and development executives at scaling retailers and restaurants, finding the “ideal” market fit for a concept, value proposition, and assortment/menu is an ongoing battle. Historically, supporting these decisions has been the domain of GIS Analytics teams armed with reams of internal and external data. Although GIS analytics has been around for more than 50 years - ESRI was founded in 1969 - the last 10 years of Big Data, ML and AI advancement have pushed big changes in the field of location intelligence.

Specialists in Location Analytics are working more holistically within their organizations. They’re taking their capability upstream to work cross-departmentally with store operations, store design, assortment design, marketing and even pricing!

The predictive models used to identify new “market white space” for growth are leveraging new machine learning tools to increase the probability that your next franchisee or store manager will be successful, thereby improving the average earnings from the network. Furthermore, these models are capable of informing decisions not only about what markets to expand to, but also simultaneously store size, format and internal design. 


Predicting White Space Market Performance

In any retail operation, there are store locations in the network that historically perform better than others. There are various reasons for that - site characteristics, management and operational efficiency, consumer demand, and competition/related real estate (market anchors). The first step is to ensure you are selecting the right “market.” The second step is to secure the right “site” in that market. 

The trick to finding the right white space market is to figure out which factors have the most influence on a market's potential value, independent of the site characteristics of a specific piece of dirt. If you can identify those factors, then you can score markets on the basis of the key factors that are associated with high performing markets in your existing network - that is, you want to find look-alike high-performing markets. Brokers will love you if you can narrow down their search and allow them to focus on what they do best: finding the right listings! 


The factors that drive success

In our experience, the factors that influence the value of a new market fall into one of four areas:

  1. Trade Area Profiles;
  2. Urban-Rural Classification of the Trade Area; 
  3. Consumer/Customer Preferences; and, 
  4. Competition and Other Retail Profile. 


Trade Area Profiles (demographics, socioeconomic)

The concept of a “retail trade area” has been around a long, long time. By broad definition, a retail trade area is the geographic area that a retail store draws customers from. 

For example, at a large home-decor department store outside of St. Catharine's, Ontario, most consumers drive a fair distance to shop there, likely up to an hour. A general industry standard for trade areas is the surrounding area of a store that represents 60% of customers. So for the St. Catharine's department store, that could be 25 minutes.

Not all trade areas are created equal

Traditionally, there are three methods you can use to define a trade area: rings, polygons, or drive/walk times. These methods are purely geographic-based distance measures. However, more advanced retailers are collecting postal codes at the cashier, while others have established loyalty or rewards programs that integrate with their POS system, giving them a wide range of insights about their customers' buying behaviour in addition to where they live. Having your customers’ postal codes at a minimum gives you the best accuracy about the size and dimension of your trade areas. If you have spend data as well, you have the gold standard!

Why does trade area definition matter?

Defining your trade area  provides you with a geographic zone within which you can profile the demographic and socioeconomic variables that define the variation and demand possibility in the area around any store. The assumption is that these are your potential customers and their profile variation will be a major component in driving market potential.


Urban/Rural Classification

When planning a market strategy, retailers and real estate professionals alike know that the way a store presents itself in a market can be vastly different depending on the characteristics of the market. One of the most obvious ways markets differ is through their urbanicity; where do they sit on a spectrum of rural vs. suburban vs. urban?  

During the COVID-19 pandemic, more and more people have been taking bets on life in smaller suburban or rural markets and as such, the urban rural continuum has begun to swing with more and more urbanites moving to more suburban and rural markets.  These choices create cultural, demographic, and economic diversity that could completely evolve the fabric of these smaller communities and their assumed customer profiles.

Most basic classification systems will just use the total population of the town or city a location falls in to determine the location class. The downside to this approach is that it does not account for other subtle differences that exist within any town, city or population centre. More sophisticated and more granular classification systems include a clustering approach on multiple attributes (say: total dwellings, daytime population, mobile devices, points of interest) across a broader territory in order to come up with an alternative classification across the urban vs. rural continuum.

Why does urbanicity matter?

Urbanicity classifications and digging into the micro-market profiles of a trade area can impact things like:

  • The size and format of the store
  • The type of customer that shops there
  • The products and services offered

The separation between urbanicity classes is a valuable measure in understanding the difference between markets, as it enables users to compare geographies using a consistent set of traits. 


Consumer/Customer Preferences

So far, we have been describing the area around the store based only on its location and the demographic profile of the population within the trade area and urbanicity classification. The next level of intelligence is using actual customer data or commercially available consumer research data to create “behavioural segment systems” attributed to the dissemination area or postal code level to provide another layer of demand variation over top of the demographic and socioeconomic profile.

Why do consumer preferences matter?

Consumer/customer segmentations based on preferences/behaviours provide real estate decision makers with important demand signals that can make or break the launch of a new location. It means very little if all you know is how much money a household makes and have no idea whether they prefer what you offer! As such, this kind of data contains rich variation in customer preferences, spending behaviour, and their potential loyalty to your brand that can increase the accuracy of market level predictions and improve your real estate choices.

The modelling process will identify which of the segments are aligned with higher performing stores and as such, can be critical in finding new white space markets where the same customer segments are present and plentiful.   

Competitive Landscape and Related/Anchor Real Estate 

The last dimension is one of the most critical: competition and related retail (market anchors or detractors). The previous variation buckets are more demand-related; now we need to discuss the supply side where constraints on performance are imposed by the co-location and density of competitors and related retail within the trade area. 

Why does competitive and related retail concentration matter?

Let's look at competitors first. This is likely obvious, but competitors congest the market unless you have a unique angle or competitive advantage. The truth is that the more competitors there are in a trade area or market, the more likely you are competing for the attention of your ideal customers which has a dampening effect on the potential for new location success. 

The other sub-dimension is related retail real estate. These can be considered “anchors” or “detractors” to your brand and its potential performance. Examples of retailers and restaurateurs that have the anchoring effect are Walmart or McDonald’s. Both these brands are typically destinations for shoppers in an area and being located near them, as long as they are not direct competitors, can have a positive halo effect on your performance. However, they can also be detractors to luxury or premium priced brands as their customers are not frequent shoppers of value brands and as such, they do not drive the “right” traffic to be a benefit. 

Modelling techniques can estimate the impact of the anchoring effect independently from demand side effects through the use of data sets that include the location of competitors and other points of interest. These data sets allow us to measure the distance effects from competitors and related retailers, but also measure the saturation of the market demand. 


Conclusion

Predicting potential white space performance using modern machine learning and statistical methodology depends on the availability of historical data at the dissemination area (Canada) or block group (US) that covers these four “buckets.” The data variables available within each bucket allow the analyst to correlate with the historical performance of the existing network to identify the factors that most influence high performing stores/markets. Each of these statistical impacts are then compiled into a prediction algorithm that can be applied to white space markets (trade areas) to provide a ranking and allow real estate professionals to prioritize markets for further investigation, and provide their brokers with guidance on finding the right “dirt."


Location analytics are not analytics you provide to retail. They are retail analytics with a location dimension. 


PiinPoint helps clients identify new trade areas that increase the probability of high performance and de-risk the location choice.

Check out PiinPoint and our Keto Cafe case study to find out how we have helped other retailers and restaurant brands improve their real estate planning, by helping them define their trade areas and produce a data-driven network roadmap.

Learn more about how PiinPoint can help your business evolve and thrive.