Retail Analytics

Recreating Success: How to Replicate your Business’s Best Locations

Jack Schroder


July 1, 2019

What is an area like?

People ask themselves this question every day. A visiting friend decides which neighborhood to stay in for the weekend. A tourist explores unfamiliar territory. A business expands into a new market.

For the visiting friend and the tourist, the stakes are low. A mistake costs some inconvenience and an Uber. Therefore, they can rely on simple guide books, online blogs, or even just looking at google maps.

For a business, the stakes are much higher. A bad location decision could lead to millions in missed opportunities or even closure. So, we turn to data. While businesses can’t be sure of the perfect location decision, we have data so that we can make an informed one. Traditionally, companies have used demographic data. Sometimes companies will also use psychographic data, but demographics and psychographics are really just two ways of slicing the same pie. More recently, movement data has come onto the scene and made a splash as well.

But, as any site selection expert could tell you, these data sources can’t tell the whole story. At Spatial, we use Geosocial data to add another perspective. This data is able to answer, “what an area is like?” better than any other dataset. For any business, especially those with a clear, distinctive brand, matching the feel of an area to their locations is essential.

Further reading: The Essential Guide to Geosocial Data.

Retailers and restaurants often know why their locations are successful and they can describe, in detail, what makes those areas unique. The problem is figuring out how to identify and replicate these strong neighborhoods in new markets.

A New Solution: Thumbprinting

In this blog post, we’re focusing on a technique we call Thumbprinting. This method of understanding neighborhoods is a novel technique for quantitatively understanding the feel, or the vibe, of an area. Here’s how it works:


That’s how thumbprinting works, but what is it good for?

Use Case: Retailers & Restaurants with fewer than 50 locations

There are a lot of potential use cases with this technique, but let’s focus on one in particular: High growth retailers or restaurants with fewer than 50 locations. These businesses are in an interesting spot. Their number of locations may make it difficult to find statistically significant relationships between location data and sales. Further, they may not have the capacity to investigate these relationships. These conditions make it difficult to quickly make intelligent location decisions, especially in new, unknown markets.

With thumbprinting, this business could replicate its ideal location(s) and identify the areas in a new market that best fit its brand or desired locale.

While we are focused on businesses with less than 50 locations in this post, it’s easy to see how other businesses could use this technique as well. Thumbprinting could apply in e-commerce (finding the best areas to market to) and more. The next section walks through a simple example of applying this technique.

Putting it together: Finding a new location using thumbprinting

In this example, a vegetarian restaurant chain has found that they perform best when they are in a culturally vibrant area. They identified an area in Williamsburg, Brooklyn as the prototypical example of the characteristics they link to success. This restaurant is growing like wildfire and wants to double its store count from 30 to 60 in the next 2 years. Using Spatial’s thumbprinting technique, it will be able to find new locations that not only have the right demographic makeup, but also the right community around it.

Step 1: Identify the social segments that define Williamsburg, NY

In this step, we process the social media generated within the targeted neighborhood and identify the social segments that stand out in the area. Unsurprisingly, the segments that stand out include Party Life, Trendy Eats, and LGBTQ Culture. All behaviors that Williamsburg is known for. In particular, the chain wishes to be near the LGBTQ Culture, which it sees as positive in 2 ways:

  1. It sees LGBTQ culture as a strong, positive indicator of vibrant & diverse culture.
  2. has identified that there is a nationally observed positive relationship between LGBTQ culture and vegetarian restaurants.

Any new locations for this vegetarian restaurant should score similarly to Williamsburg. In particular, they will likely score very highly for Party Life, Trendy Eats, and LGBTQ Culture (among others).

For a list of all our segments, explore the Taxonomy.

Step 2: Use the “thumbprint” to quantitatively see the similarity of other areas “feel” to that of Williamsburg, NY.

The restaurant chain, which has locations mostly on the East Coast, wants to expand in the West and chose Phoenix as the first market to enter. The map below shows similarity to the target neighborhood mapped across Phoenix, Arizona.

Heatmap showing the level of similarity to Williamsburg, NY

At first glance, it’s a bit overwhelming. But, this map has a lot of power. Each darkly colored area represents places that show a high degree of similarity to Williamsburg, NY. Let’s focus in on 2 of the darker areas (highlighted in orange for quick reference) and look what's there. The first area is in southeast Scottsdale.

Neighborhood in Southeast Scottsdale

It’s interesting to see a couple of landmarks that might be driving some of this similarity to Williamsburg. The Entertainment District, 5th Avenue Shops & Boutiques, and the Main Street Arts District all may have something to do with it.

The second area we’ll zoom into is in the north part of downtown Phoenix, near Broadway and Evans Churchill.

Neighborhood in North Downtown Phoenix

In the screenshot, you can see landmarks like the Phoenix Public Library and the Arizona School for the Arts. A quick cross-reference on google maps reveals an even more vibrant scene, including jazz clubs and the Phoenix Center for the Arts. Roosevelt Street brims with restaurants, boutiques, and culture.

This visualization helps give a better understanding of how it can be used to narrow down potential locations to just a few. The best part? These potential locations will feel more like Williamsburg than other locations in the market. Combining this analysis with demographic requirements and any other data necessary creates a powerful tool that will identify the ideal candidate locations for any business.

Step 3: Predict the best locations for a given decision using this data.

Using the data from step 2, we can now begin to make location decisions with an understanding of what an area is like. Depending on the type of decision being made, this step could look different. For our vegetarian restaurant, it knows that it needs to have a certain percentage of people aged 26-40 with household incomes greater than $70,000. Combining this requirement with the results of thumbprinting, it’s able to narrow the field down to 2-3 locations. And this boutique can repeat the process for each market it looks to enter.

Making location decisions is hard, and getting the right data to do so is not an easy task. Using Geosocial data to improve a decision can give retailers and restaurants an edge in finding the right locations for their business.

How PersonaLive Segmentation System Works

Personalive segmentation uses social media, mobile foot traffic, online activity, and individual-level demographics to organize every US household into one of 80 behavioral segments. These segments provide visibility to the online and offline preferences of the customers visiting any US property.

01 Append

Draw a polygon around a property to identify the behavioral segment of every visitor.

02 Analyze

Rank the top customer types visiting a location. Then match retailers based on online and offline activities of visitors.

03 Activate

Demonstrate visitor brand affinity to close deals. Activate marketing campaigns to drive target segments to your location.

Related posts

Retail Analytics

Ecletticos: Optimizing Facebook Campaigns With PersonaLive Segmentation

Lyden Foust


April 6, 2022
Retail Analytics

Startup Uses Geosocial Data to Analyze Consumer Behavior

Lyden Foust


July 2, 2020

Ready to get started with Geosocial data?